Akoglaniz para Immortal

First published in 2015 by Gloucester Publishers Limited, Northburgh House,

10 Northburgh Street, London EC1V 0AT

Copyright © 2015 Cyrus Lakdawala

The right of Cyrus Lakdawala

to be identified as the author of this work has been
asserted in accordance with the Copyrights, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a

retrieval system or transmitted in any form or by any means, electronic,
electrostatic, magnetic tape, photocopying, recording or otherwise,
without prior permission of the publisher.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

ISBN Kindle: 978-1-78194-258-1

ISBN epub: 978-1-78194-259-8

Distributed in North America by National Book Network,

15200 NBN Way, Blue Ridge Summit, PA 17214. Ph: 717.794.3800.

Distributed in Europe by Central Books Ltd.,

99 Wallis Road, London E9 5LN. Ph 44(0)845 458 9911.

All other sales enquiries should be directed to Everyman Chess,

Northburgh House, 10 Northburgh Street, London EC1V 0AT
email: [email protected]; website: www.everymanchess.com

Everyman is the registered trade mark of Random House Inc. and is used in this work under licence
from Random House Inc.

Everyman Chess Series

Chief advisor: Byron Jacobs
Commissioning editor: John Emms
Assistant editor: Richard Palliser

Typeset and edited by First Rank Publishing, Brighton.

Cover design by Horatio Monteverde.
Printed by TJ International Limited, Padstow, Cornwall.
About the Author
Cyrus Lakdawala is an International Master, a former National Open and
American Open Champion, and a six-time State Champion. He has been
teaching chess for over 30 years, and coaches some of the top junior players
in the U.S.

Also by the Author:

Play the London System
A Ferocious Opening Repertoire
The Slav: Move by Move
1 ... d6: Move by Move
The Caro-Kann: Move by Move
The Four Knights: Move by Move
Capablanca: Move by Move
The Modern Defence: Move by Move
Kramnik: Move by Move
The Colle: Move by Move
The Scandinavian: Move by Move
Botvinnik: Move by Move
The Nimzo-Larsen Attack: Move by Move
Korchnoi: Move by Move
The Alekhine Defence: Move by Move
The Trompowsky Attack: Move by Move
Carlsen: Move by Move
The Classical French: Move by Move
Larsen: Move by Move
1 ... b6: Move by Move
Bird’s Opening: Move by Move
Contents
About the Author
Series Foreword
Bibliography
Introduction

1 The Cochrane Gambit

2 The Scotch Petroff
3 The Old Main Line
4 Fifth Move Alternatives
5 The New Main Line: Shirov’s Variation
6 The Three Knights Petroff
7 Odds and Ends

Index of Complete Games

Series Foreword
Move by Move is a series of opening books which uses a question-and-answer
format. One of our main aims of the series is to replicate – as much as
possible – lessons between chess teachers and students.
All the way through, readers will be challenged to answer searching
questions and to complete exercises, to test their skills in chess openings and
indeed in other key aspects of the game. It’s our firm belief that practising
your skills like this is an excellent way to study chess openings, and to study
chess in general.
Many thanks go to all those who have been kind enough to offer inspiration,
advice and assistance in the creation of Move by Move. We’re really excited
by this series and hope that readers will share our enthusiasm.

John Emms
Everyman Chess
Bibliography
Books
A Vigorous Chess Opening Repertoire for Black, Or Cohen, (New In Chess 2013)
Kasparov vs Karpov: 1986-1987, Garry Kasparov, (Everyman Chess 2009)
Petroff Defence, Alexander Raetsky and Maxim Chetverik, (Everyman Chess 2005)
Starting Out: Open Games, Glenn Flear, (Everyman Chess 2010)
Starting Out: 1 e4, Neil McDonald, (Everyman Chess 2006)
The Four Knights: Move by Move, Cyrus Lakdawala, (Everyman Chess 2012)
The Petroff: an Expert Repertoire for Black, Konstantin Sakaev, (Chess Stars 2011)
World Chess Championship: Kramnik vs Leko, Ray Keene, (Hardinge Simpole Publishing 2004)

Electronic/Online
ChessBase 10
Chesslive database
ChessPro (Annotations by Konstantin Landa)
ChessPublishing.com (Annotations by Paul Motwani, Victor Mikhalevski, Richard Palliser, David
Vigorito and Olivier Renet)
Chess Today (Annotations by Alex Baburin, Mikhail Golubev and Maxim Notkin)
Introduction
“Out of this nettle, danger, we pluck this flower, safety.” – William
Shakespeare

W
Soldiers of ancient Rome, when on the defensive against the Barbarian
hoard, huddled into a tortoise-like defensive formation with shields thrust
outward to protect them from the incoming arrows and spears. They called it
the ‘shield wall’. Petroff’s Defence borrows this strategy. “I hate the Petroff.
It seems to me that Black only wants a draw, and he usually has memorized
long variations that bring him close to his goal. As (GM) Gufeld would say,
‘This is not chess!’” wrote IM Jack Peters to me in an email. The name of our
opening, to some, signifies stupefying boredom, but only to those who don’t
relate to it. I remember throwing an anti-Petroff’s/Berlin Lopez tirade to
some friends at the chess club about 15 years ago (your moody writer throws
many such rants on various subjects quite often, to all who are willing to
listen) about how “the Petroff’s and Berlin Lopez are conspiring to ruin 1
e4?!, until the move is worked out to a forced draw”. I raved on, adding that
Black’s overly respectful interpretations were akin to playing a piano
concerto with just one finger, on a single key. Our opening is one of cold-
hearted – almost reptilian – pragmatism, where we begin the game with deep
deference to White’s first move, with the humble objective to merely not
lose, rather than win. You may ask: why play in such fashion?
A Petroff-ignorant member of the San Diego Chess Club proclaimed the
opening boring and lacking in imagination. I innocently queried him, “You
mean like a person who names his dog, “Dog?” He proceeded to claim that
defence of the many ‘+=’ positions can be as joyless as attending a funeral to
which I countered with “What if you attend the funeral of an enemy?” But
this point of view lacks perspective. In the Petroff’s, we use our imagination,
not so much to wrest the initiative, but more to frustrate and neutralize
White’s first move advantage and create an unbreachable defensive firewall.
The overall goal is not to energize our own positions, but instead, to suppress
and deflate the opponent’s vitality – and most of all, his or her hopes of a full
point. If every opening has its own distinctive personality, then the Petroff’s
is one of introverted humility. A chess opening is defined, not by its thrill-
level, but what we want from it. Activity? Solidity? Attacking or
counterattacking potential? In the case of the Petroff’s, what we want most of
all is to avoid loss. We are willing to suffer 25 or so moves of an opponent’s
mild ‘+=’ to achieve it. We seek to create a self-sufficient barrier to White’s
ambitions. Also, don’t be fooled: Our drably inconspicuous opening can –
and often does – quickly erupt into chaos at a moment’s notice. We may have
the preference for a quiet draw, but may well receive adventure, whether we
like it or not.

When life deals us an unsavoury hand (i.e. the black pieces!) we have two
choices:
1. Rebel against reality by playing an ultra-sharp line.
2. Accept the reality and dig in solidly to attain equality in time.
Petroff’s Defence – like the Berlin Ruy Lopez (which appears to be a bit of an infringement of the
Petroff-copyright, as the most difficult anti-e4 opening for White to crack), is an example of the latter
choice, while openings like Sicilian Najdorf and Dragon, ignore White’s first move advantage and go
for the full point with increased chances of a loss. Those who live in the hostile environments, like the
Sicilian Dragon or Najdorf, the King’s Indian and King’s Gambit et al., are inured to violence and
expect it in virtually every game they play. Petroff’s is a different place, where our opponent chases
while we dance, just out of reach. When we win, it is almost always because our opponent overpress.

Petroff’s: The Theoretical Nightmare

Some openings, like the London System and Colle (considered by many as the least adversarial
openings in chessdom) are built for utility and thus relatively easy to learn, the way a TV dinner is
designed for the divorced man who still hasn’t learned to cook. The theory-intensive Petroff’s Defence
is the absolute opposite. The opening is not conducive to those who study by osmosis and blindly
absorb theory by going over a large number of database games. So unforgiving is Black’s side that even
a seemingly innocently subtle variance in the position can bring immediate pain. Our lines are
incredibly involved, one looking similar to the other – yet requiring a different plan. The knowledge
and ideas from a chess book do not magically jump off the page and into our minds permanently. We
must go through the lines over and over again for this to occur. Through the magic of repetition, our
brains begin to construct highly specialized neural pathways, dedicated to remembering even the most
specific patterns. It is one thing to memorize our lines, and quite another to distil the knowledge into a
coherent middle game plan. However, there is good news as well: one advantage of playing 2 ... Nf6
over the more traditional 2 ... Nc6 is that apart from skipping the heavily analysed Ruy Lopez, we don’t
need to learn any of White’s tricky gambits and specialty sidelines, such as the Goring Gambit, the
Danish Gambit, the Scotch Game, the Max Lange Attack, et al.

Our Pedigree
Any opening endorsed by a World Champion tends to elevate its brand. In the case of Petroff’s, we
have towering advocates like Karpov, Kramnik and Anand, as well as numerous super GMs. So I can
wholeheartedly assure you that our opening is sound!
We must acquaint ourselves with the flora and fauna of each opening we study, so here are the
critical positions we meet in Petroff’s Defence:

Cochrane Gambit

“Truth is beauty,” wrote John Keats. If we extend this to chess, then all semi-sound sacrifices and
attacks are ugly. I don’t know if Keats took time out from his busy poetry schedule to play chess, but if
he did, he would most certainly have frowned upon the presumptuousness of the Cochrane Gambit
(Keats wrote an unfinished fairy tale- the Cap and Bells, mentioning a Princess who “…castled her
King with such a vixen look”). White suffers from a selective interpretation, a preconceived belief that
he or she has a strong attack, ignoring the possibility that the gambit is borderline unsound if Black
knows the theory. White’s sacrifice at first glance feels so incomprehensible, that we suspect our
opponent of conduct of one playing some game other than chess, where material doesn’t matter. White
refuses to pay even marginal homage to rationality, while unabashedly embracing chaos. In today’s
comp-driven chess world, the sacrifice is deemed semi-sound at best.

The Scotch Petroff’s

In the Scotch Petroff’s, White wants an open game. He or she can have it. But an advantage?
Maybe not.

The Old Main Line

 Here we arrive at the Old Main Line, perhaps the most theoretically-dense portion of the book,
where our lines are indexed, referenced, cross-referenced and cross-cross referenced, until our heads
(and our opponents’ as well!) spin with data overload. One variation looks much like another, as they
blend into an unclassifiable miscellaneity of theoretical confusion.

The Shirov Attack

In 2002, GM Alexei Shirov electrified the chess world by defeating GMs

Morozevich and Motylev in the Russian versus the World rapid match. He
singlehandedly turned a dusty, seemingly harmless line into what is now
considered White’s main line against Petroff’s Defence. The divide comes
here. White will either post his or her bishop on the e3- or f4-square. We
need to understand the differences and know how to react properly to either
plan.

“I Offer a Draw”
One warning to the reader: Petroff’s isn’t your best choice when you as Black
play for the win. If a water colour artist were to paint the mood of the next
two diagrams, she would do so with a limited palette of mainly grey tones.
The following two diagrams are examples of two very drawish lines White
has at his or her disposal:

In this annoying version, we are forced to respond with ... Qe7, after which
White forces queens off the board. However, our position isn’t quite an exact
replica of White’s since we are minus a move in the symmetry and must play
accurately to equalize.
 Sometimes counterfeiters try their luck by playing Petroff’s from White’s side, except a move up. In
this position White will play d3-d4 next, insisting upon an Exchange French type of position when we
respond with ... d6-d5, in the hopes of reaching an intolerably dull version. If we are happy with a
draw, then we have no worries. If we have hopes for a win, the ritual of seemingly eternal sameness can
finally end if we play ... c7-c5 at some point and take on an isolani.

Just to show the reader that the Petroff’s isn’t as wimpy as its reputation suggests, watch how Pillsbury
dismantles the great Emanuel Lasker in the following game:

E.Lasker-H.N.Pillsbury
St. Petersburg 1895

1 e4 e5 2 Nf3 Nf6
Petroff’s Defence is an ideal weapon against an ambitious opponent.
3 Nxe5 d6
One of the earliest examples of the Petroff’s didn’t end well for our side,
with the incompetent NN playing the Black pieces against the best player of
the era, Gioachino Greco: 3 ... Nxe4? 4 Qe2 Qe7 5 Qxe4 d6 6 d4 f6? (NN
should just forget about the pawn, and try to develop with 6 ... dxe5 7 dxe5
Nc6) 7 f4 (Morphy would have played the superior 7 Nc3!) 7 ... Nd7 8 Nc3
dxe5 9 Nd5 Qd6? (NN adopts a defiant stance, when he should run, squealing
like a frightened mouse, so 9 ... Qd8 was better) 10 dxe5 fxe5 11 fxe5 Qc6
(11 ... Nxe5 loses instantly to 12 Bf4) 12 Bb5! Qc5 13 Be3 (the bishop likes
to decorate the execution chamber with bright, cheery colours – in absolute
antithesis of the unfortunate trapped black queen’s mood) 13 ... Qxb5 14
Nxc7+ Kd8 15 Nxb5 1-0, G.Greco-NN, Europe 1620. In this book, I hope to
improve upon NN’s handling of the Petroff’s!
4 Nf3 Nxe4 5 d4 d5 6 Bd3 Be7 7 0-0 Nc6 8 Re1 Bg4! 9 c3 f5
When we don’t fake it and actually play in our style, we reveal our true
nature to our opponent. Pillsbury, a natural attacker and the father of the
Stonewall Dutch formations from White’s side, would have felt very
comfortable with Black’s Stonewall-ish d5- and f5-pawn structure.
10 Qb3 0-0!
Amazingly, this position is still a main line today, found in Chapter Three.
11 Bf4!?

Pillsbury’s new-fangled opening confuses Lasker who plays an inferior

move. This is the first deviation from current theory. Today, 11 Nbd2 is
normally played.

Question: Can White grab the b7-pawn and get away with it?

Answer: No. Please fast forward to Chapter Three to see why not!
11 ... Bxf3!
Black inflicts serious damage to White’s kingside structure. White’s light-
squared bishop doesn’t provide enough compensation.
12 gxf3 Ng5 13 Kg2
 White’s kingside is getting draughty, and the paucity of defenders around
his abode does not bode well.
13 ... Qd7?!
This natural move may not be best since the b7-pawn is now really under
attack. Stronger is 13 ... Na5! 14 Qa4 c6:
a) In one ear the voice of reason and conscience whispers, and in the other
ear, we hear the voice of greed and irrationality. Then why is it we are so
tempted to heed the latter?
15 Rxe7? Qxe7 16 Qxa5 Ne6 17 Bd2 Qh4 (White is grossly
underdeveloped with his queen taking vacation on the a5-square while Black
threatens to activate his rook into the attack with ... Rf6)18 Na3 Rf6 19 Kh1
f4 (White sets up the ... Rh6 threat though even more crushing is 19 ... Qxf2!)
20 Rg1 Qxf2 21 Rg2 Qxf3 22 Be2 Qe4 0-1, Richter-Methner, Breslau 1889.
b) 15 Nd2 b5 16 Qc2 Bd6 17 Bg3! Qd7 18 h4 Ne6 is messy and even according to Houdini.
14 Qc2?
In this case, the adverse side effects of the cure may surpass the difficulties
of the disease. Lasker, one of the all-time great pawn grabbers,
uncharacteristically declines the b7-pawn when he should have taken it.
After 14 Qxb7! Ne6 15 Rxe6 Qxe6 16 Bb5 Na5 17 Qxc7 Nc4 18 Bxc4
dxc4 19 Nd2 Rac8 20 Qxa7 Bd6 21 Bxd6 Qxd6 22 Re1 Qg6+ 23 Kf1 Qh6 24
Re7! Rce8 25 Ke2, Black’s threats, like grains of sand in our shoe, are
annoying but unable to inflict real damage. I actually prefer White’s position,
since his king looks safe enough and he has an awful lot of pawns for the
exchange. If the c4-pawn falls, then White will have four scary queenside
passers.
14 ... Ne6 15 Bc1
The bishop awakens to a sense of obligation and duty to his king, and then
realizes: “The hell with it! It’s every man for himself.”

Question: Why did Lasker undevelop like this?

Answer: Clearly the bishop refuses to chip in and do his part for the war effort, but I don’t see
reasonable options as the bishop does not have a better retreat square. For example:
a) 15 Bd2 clogs his queenside development.
b) 15 Be3? just loses time to 15 ... f4! and if White gets greedy with 16
Bxh7+??, 16… Kh8 17 Bc1 Ng5! is curtains, since the h7-bishop hangs and
Black also threatens mate, beginning with ... Qh3+.
15 ... Bd6
Seizing control over the f4-square. It’s safe to declare Lasker’s opening a
spectacular disaster.
16 Nd2
Not 16 Bxf5?? which fails to 16 ... Rxf5 when White is unable to
recapture, due to the ... Nf4+ discovery.
16 ... Rae8
Pillsbury calmly brings in another attacker.
17 Nf1?
Also, 17 Rd1 Nf4+ 18 Kh1 Rf6 preparing a hit on the h-file, gives Black a
winning attack, but this line still puts up greater resistance than Lasker’s
choice.
 Exercise (combination alert): In miserable positions, we become
a person afflicted with cataracts, who can only make out vague
forms devoid of all detail or colour. A dejected Lasker’s last move was
an error in an already wretched position. What did White miss?

Answer: Discovered attack/double attack. Black wins a key pawn with this tactic.
17 ... Nexd4! 18 Qd1
The queen nervously attempts to change the subject. Not much choice in
the matter since taking the knight drops the e1-rook.
18 ... Rxe1 19 Qxe1
 Exercise (combination alert): Find the way to sustain Black’s attack.

Answer: Annihilation of defensive barrier/piece sacrifice.

19 ... Nxf3!
White’s king won’t last long in the forthcoming onslaught.
20 Kxf3
In such positions, we often wish we could retract our last few key
decisions. But now there is no reason to flee, when there is no home to return
to.
20 ... f4!
A previously underground organization is exposed to the light. Now
Black’s queen is granted entry into the attack via the chronically-weakened
kingside light squares. This ensures White’s endless cycle of suffering.
21 Qd1
Also futile is 21 Ke2 Ne5 22 Bc2 Qg4+ (irresistible commands leap from
the g4-witch’s mind, directly into the ensorcelled white king’s thoughts,
compelling him to leap to her bidding against his will) 23 Kd2 (even though
the king still has a few weary remaining defenders around him, he realizes he
is alone) and 23 ... Nf3+ wins the queen (while for the purists among us, 23 ...
Nc4+! forces mate in eight).
 21 ... Ne5+ 22 Ke2 Qg4+
White’s king awakens from a terrible nightmare where his sister attacks
him with a knife. But unlike a dream, he feels the sting of very real pain from
a new scar on his cheek.
23 Kd2

Exercise (combination alert): This one is easy. Black to play and win a piece.

Answer: Deflection/overloaded defender.

23 ... Qxd1+
“I always knew that love would be your undoing,” the queen tells White’s
heartbroken king.
24 Kxd1
The king’s right foot trips over his left.
24 ... Nxd3
When you lose something of value and then later regain it, its acquisition
is so much sweeter the second time around. White is down two pawns in the
ending and remains with a rotten position. Of course Lasker could have
resigned here. I’m pretty sure he just sat there, reflecting woefully upon his
previous folly, the way a small child just learned that a hot stove burns.
25 Ke2 Ne5
 Pillsbury refuses to exchange off White’s lousy remaining bishop.
26 f3 Re8 27 b3 Ng4+ 28 Kd2 Ne3 29 Bb2 Ng2 30 h3 Bc5 31 Nh2 Bf2
32 c4 dxc4 33 bxc4 h5 0-1
Summary: Welcome to one of the most solid defences in all of chess:
Petroff’s Defence.

Acknowledgements
Many thanks to my editor GM John Emms, and to CM Junior Tay for
the final edit. And thanks also to my wise and all-powerful Khaleesi,
Nancy, for her proofreading and her constant supply of soothing, piping
hot tea.

May our Petroff shield-wall hold eternally against the barbarian

invaders who seek to knock it down and enter our realm!
Chapter One
The Cochrane Gambit

This is the starting position of the rather impulsive Cochrane Gambit after
White has ditched his knight for the f7-pawn. This is a position loaded with
intriguing potential for both attacker and defender. At first glance, it feels as
if White’s sacrifice fails to meet even rudimentary standards of soundness,
and that this is true generosity in a gravely hypertrophied form.
We look at three lines from this position: 5 d4, 5 Bc4+?! and 5 Nc3. When I comped the lines, they
always worked out in Black’s favour. However, as we all know, it’s one thing to wrest the advantage
theoretically, with a soothing cup of tea and a 3200- rated chess computer at our side. Over the board,
we face considerable practical difficulties, especially if White plays the sacrifice in a rapid game.
According to my database, White scores respectably – just over 53%, so let’s give the Cochrane some
respect (even though in our hearts we may still disrespect it!) and study it well.

Game 1
N.Short-A.Shirov
FIDE Grand Prix, Dubai 2002

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nxf7

 The opening salvo of the mean-spirited Cochrane Gambit. I must take my
hat off to both Short and Topalov in this chapter, since they had the colossal
nerve to play the gambit on world class players.

Question: Surely it’s unsound to hand over a full piece for merely two pawns
on the fourth move, against one of the most solid openings in chess, isn’t
it?

Answer: Will Black’s king ford the river, or drown? At this point it’s hard to say. I agree that the
sacrifice is semi-sound at best, but from a practical standpoint, it’s eminently dangerous, and the line
deserves its own chapter. It scores a respectable 53.1%, so clearly we must factor in White’s practical
chances – especially if the sacrifice is played on you in a rapid game, as in this one. There is no way to
work out (at least with our limited human brains) every permutation to such a speculative sacrifice, but
with the assistance of those wonderful/terrible comps, we can come close. One can’t shake the feeling
that White unwisely declared war, when lacking in manpower and armaments.

Question: Let’s check if you have been reading the introduction

notes to the chapter. What does White get for the piece?

Answer: For the piece, White gets the following:

1. Two pawns, since his kamikaze knight martyrs itself for the f7-pawn
and he also gets to hang on to the e4-one.
2. Black’s castling has been disrupted and our king remains in the centre
for some time to come.
3. The loss of our f-pawn means that White may eventually play for d2-d4,
f2-f4 and e4-e5, sending our f6-knight packing.
4. Bc4+ is in the air. We sometimes respond with ... Be6, which means
that after White swaps, our nervous king suddenly finds himself on e6, dead
centre of the board with the vast majority of white attackers looking in his
direction.
4 ... Kxf7 5 d4
In Game 2, we will look at 5 Bc4+?!.
5 ... c5!
Black undermines White’s sole compensatory asset: The big pawn centre.

Question: Why can’t Black take the e4-pawn?

Answer: That would be a big mistake which returns the piece after 5 ... Nxe4? 6 Qh5+ Ke7:
a) 6 ... Kf6 7 Qh4+ g5 8 Qxe4 Qe7 9 Nc3 leaves Black down a pawn in an
awful position.
b) 6 ... Kg8?? is a helpmate after 7 Qd5+ Be6 8 Qxe6 mate!) 7 Qe2 and
White regains the piece with Black’s king still bouncing around in the centre.
6 dxc5
a) 6 Bc4+ d5! (Cochrane Gambit principle: Always meet any Bc4+ with ...
d5!, if in doing so, White is forced to take d5 with a pawn) 7 exd5 b5! (White
reels from this energetic counter)
 8 Be2 (after 8 Bxb5? Qxd5 9 0-0 Bb7 10 f3 Qxd4+, queens come off the
board and White has only two pawns for the piece) 8 ... cxd4 9 0-0 Bc5 10
Bg5 Nbd7 11 c3 Qb6 and White lacks compensation, since Black’s forces are
exceedingly active and his king remains safe, R.Pegg-J.Helbich,
correspondence 2008.
b) 6 Nc3 Nc6 7 Bc4+ d5! (this move sometimes even works when White is
able to capture the d5-pawn with a piece) 8 Nxd5, H.Blauhut-N.Selyanchin,
correspondence 2008. Once again, White lacks full compensation for the
piece after 8 ... Be6 9 0-0 cxd4 10 Bg5 Be7 11 Bb3 Re8 12 Qf3 Kg8 when
Black successfully completes his development and escorts his king to relative
safety.
6 ... Nc6!
 Stressing rapid development over material issues.

Question: White now has three pawns for the piece. Does he have compensation?

Answer: GM Paul Motwani answers your question: “Black’s lead in development and his extra
piece easily outweigh the three surplus pawns that White has as (inadequate) compensation”.
Other options:
a) Black’s last move is much stronger than the greedy 6 ... Qa5+?! 7 Nc3
Qxc5 (7 ... Nxe4?? is met with the double attack trick 8 Qd5+) 8 Be3 when
White enjoys a considerable development lead and may well have full
compensation for the piece.
b) Worth considering is the central counter 6…d5!? 7 e5 Ng4 8 Bc4 Bc5 9
Qxd5+ Qxd5 10 Bxd5 and here, Black has the ambitious 10…Be6! (Cohen
analysed 10…Kf8, leading to a slight advantage assessment) 11 Bxb7 Nd7 12
Bxa8 Bc4+ 13 Kd1 Rxa8 gives Black a powerful attack, thanks to his flurry
of hyperactive minor pieces and the precarious position of the white king.
7 Bc4+
This lures Black’s king out ever further, yet he has fully adequate
resources, since it is Black, not White who leads in development.
7 Nc3, “jkidjr22”-C.Lakdawala, Internet (blitz) 2014. Now Black should
refrain from recapturing the c5-pawn, which gives White some play for the
piece, and instead play 7 ... Be6! 8 cxd6 Bxd6 9 Bd3 Re8 10 0-0 Kg8 when
Black’s king is safe and White’s pawns are not enough for the piece.
7 ... Be6 8 Bxe6+ Kxe6 9 0-0
GM Konstantin Sakaev gives 9 Nc3 Kf7! 10 Qe2 Qd7 11 Be3 dxc5 12 f4
Nd4! (Sakaev’s novelty) 13 Bxd4 cxd4 14 0-0-0 Rc8 15 e5 Rxc3 16 bxc3
Ba3+ 17 Kb1 Nd5 18 Ka1 Qc6 and “Black has an ‘overwhelming advantage”
(Sakaev).
9…Kf7!

Black’s king, ungraspable as a shadow, proves to be an elusive target for

White’s still underdeveloped forces. Black’s last move was a powerful
theoretical novelty at the time, which I believe buries White’s 5 d4 sub-
variation of the gambit – at least theoretically, if not on a practical level.
Black’s king remains safe and untroubled – for now! 9 ... d5 isn’t quite as
accurate. It conforms to the tenets of principle, yet remains ineffective. 10 e5!
Ne4 11 Qg4+ Kf7 12 Qf5+ Ke8 (12 ... Kg8?? 13 Qe6 mate would be a
wonderful gift for White) 13 Nc3 Nd4, A.Vitolinsh-R.Dautov, Minsk 1988.
Black should still be better here but must play very accurately. Following 14
Qh5+ g6 15 Qg4 Nxc3 16 bxc3 Nf5 17 Rb1 Qc8 18 Qf3 Qc6 19 e6, (White
threatens g2-g4! with Qf7+ in the air) 19 ... h5!, I don’t believe in White’s
full compensation for the piece.
10 Qe2
 a) 10 Qd3 Qd7 11 Rd1 Qe6 12 cxd6 Nb4 13 Qb5 Bxd6 14 Nc3 Rhd8 15
Bg5 a6! 16 Qxb7+ Kg8 (Black’s king is safe, while White’s queen is in grave
danger prompting the next move) 17 Bf4 Bxf4 18 Qxb4 Rdb8 19 Qa3 Be5 20
Rab1 Qc4 and White’s four pawns for the piece are of little use here. Black
applies tremendous pressure to the queenside and White looks busted,
J.Bentancurt-H.Blauhut, correspondence 2004.
b) 10 Nc3 dxc5 11 Qe2 Qe8 (covering against Qc4+) 12 Re1 Nd4 13 Qd3,
H.Starke-D.Hudak, correspondence 2009. Black looks better after 13 ... Kg8
14 Bf4 Qf7 15 Rad1 Re8 as I don’t see full compensation for White, although
Black must still deal with the irksome issue of how to activate the sleeping
h8-rook.
10 ... Qe8 11 Re1
11 Nc3 dxc5 is similar to the Starke-Hudak note from above.
11 ... d5!
Principle: Meet an attack with a central counter. Alternatively:
a) Black wisely steers clear of 11 ... dxc5? 12 e5 Rd8 13 Nc3 and Black
should return the extra piece, since 13 ... Nd5?? is met with 14 Nxd5 Rxd5 15
Qh5+! (creating a weakness on g6) 15 ... g6 16 Qf3+ Ke6 17 Qf6+ Kd7 18
e6+ Kc7 19 Qxh8 and Black can resign.
b) Sakaev recommends 11…Qe6! which is strong as well.

However, my take is the resulting positions are harder to defend than the
11…d5 lines, especially for club-level players. After 12 Nc3 Rd8! 13 Bg5
dxc5 and here White can try 14 f4 (Sakaev analysed 14 e5 h6 15 Bxf6 gxf6
16 f4 f5! and he stated that “Black’s blockade is secure and the extra knight
will tell”) 14…Be7 (14…Nd4! might be even better and after 15 Qf2 h6 16
f5 Qc6 17 Bxf6 gxf6 18 Nd5 Bd6, Black has better chances) 15 e5 Nd4 and
here, White has a choice of two queen moves:
b1) 16 Qf1 Ng4 17 Bxe7 Kxe7 18 h3 Nh6 (Black is also slightly better
after 18 ... Nxc2 19 hxg4 Qxg4 20 Qc4 Nxe1 21 Qxc5+ Ke8 22 Rxe1) 19 Qf2
Qf5 20 Ne4 g5! 21 Nxg5 Qxc2 22 Rac1 Qxf2+ 23 Kxf2 Rc8 24 g4 and White
does have some play for the material deficit but the ball is in Black's court.
b2) 16 Qd1 Nd5 17 Ne4 Rhf8 18 Nd6+ Kg8 19 Bxe7 Nxe7 20 Nxb7 Rb8
21 Nxc5 Qd5 22 Nb3 Rxf4 23 c3 Nxb3 24 Qxd5+ Nxd5 25 axb3 Rxb3 26
Rad1 Ne7 27 Rd7 Kf7 28 e6+ Kf6 29 g3 Ra4 30 Rc7 Rxb2 (30 ... Rb5! is
stronger, keeping the extra knight while preparing ... Ra2 and start the a-
pawn running and after 31 Re2 Ra2 32 Rf2+ (to drag the b-rook away, thus
protecting the b2-pawn) 32 ... Rf5 33 Re2 h5 34 Kg2 a5, Black has winning
chances) 31 Rf1+ Nf5 32 Rf7+ Kxe6 33 R7xf5 Raa2 34 Rh5 and White
managed to survive this ending to nick a draw, T.Schwetlick-J.Lehtosaari,
correspondence 2005.
12 e5 Ne4 13 Nc3!

Short is willing to pay a premium to eliminate Black’s powerful e4-knight.

In doing so, he damages his structure beyond repair and reduces his attacking
force further. Even still, it’s his best move since by opening the b-file, he
activates his a1-rook and applies pressure to the b7-pawn. 13 Bf4? makes no
sense at all, since White simply loses a tempo after 13 ... Bxc5 14 Be3,
B.Koen-Z.Sanner, correspondence 2005. Black is winning after 14 ... Qe7 15
e6+ Kg8 16 Bxc5 Qxc5 17 Nc3 Nxc3 18 bxc3 Ne7 19 Rad1 h5! (Black plans
to activate his h8-rook via the h6-square) 20 c4 dxc4 21 Rd7 Rf8 22 Rxb7
Rh6 and ... Rhf6 is coming and Black consolidates.
13 ... Nxc3 14 bxc3
White’s structure begins to clot with long-term weakness. White has a
couple of zwischenzugs to consider:
a) Perhaps White should toss in 14 e6+ which prevents Black from a ...
Qe6 light-square blockade, as in the game’s continuation. After 14 ... Kg8 15
bxc3 Bxc5 16 Rb1 b6 17 c4 dxc4 18 Qxc4 Ne7 19 Be3 Bxe3 20 Rxe3 h5!
(remember his manoeuvre, which enables us to activate our dormant h8-rook)
21 Rd1 Rc8 22 Qb3 Qg6 23 Rd7 h4! 24 Qb4 (24 Rxe7?? is met with the
sneaky back rank cheapo 24 ... Qf6! 25 Qb4 a5! when White’s queen is
overloaded and he loses a rook) 24 ... Nf5 25 Re2 h3 26 g3 Kh7, Black is
well on his way to consolidation.
b) Sakaev considers that White has sufficient compensation after 14 Qf3+
Kg8 15 Qxc3 15 ... b6 16 Be3. However, Black can play 15…Qe6 instead
and following 16 b4 h6 17 Bd2 g5 18 a3 Bg7 19 Rad1 Qf5 20 Qb3 Kh7, he
can complete development with good prospects.
14 ... Qe6!
Two windows simultaneously open on Black’s mental computer:
1. Black sets up a light-squared barrier. If he prevents the f2-f4-f5 pawn
thrust, then he renders White’s 4:2 kingside pawn majority meaningless.
2. Black buys time for ... Be7, ... Rhf8 and ... Kg8, castling manually.
If he manages both, he emerges with a winning position. Shirov’s move is
more accurate than 14 ... Bxc5?! 15 Qf3+ Ke6 16 Qg4+ Kf7 17 Qf3+ with
perpetual check.
15 Rb1 Rb8
The b7-issue is easily dealt with, but it does effectively tie down Black’s
rook.
16 Be3 Be7 17 f4 Rhf8
Black is ready to castle manually with ... Kg8 next.
 18 Bd4
On 18 f5?, Black sidesteps the cheapo and naturally takes the e-pawn,
rather than the f-pawn. After 18 ... Qxe5 19 Qd2 Kg8 20 Bf4 Qxf5 21 Bxb8
Bxc5+ 22 Kh1 Rxb8, Black has bishop and knight for only a rook, with a
winning position.
18 ... Kg8
We note a waning quality to the energy level of White’s attack. Black has
obviously consolidated. It appears as if he castled all along.
19 Rf1
Short still harbours dreams of activating his majority with g4 and f5.
19 ... g5!

Undermining White’s imposing centre.

Question: Isn’t Black overshooting his mark, by volunteering to expose his own king?

Answer: In such unforgiving positions, the tingle of imminence courses through our minds,
reminding us that one mistaken idea may be our last. I like this excitable if unforced addition. Shirov is
willing to expose his own king as in doing so, he breaks up White’s central pawn monopoly. However,
he gravely increases the complexity level of the game, which can only benefit White, who remains
objectively in deep trouble. Safer but less promising is 19 ... Rf7 20 g4 g6 21 Qf3 Rbf8 22 Qg3 (22
Rxb7?? is met with 22 ... Rxf4 23 Qxf4 Rxf4 24 Rxf4 Bg5 and when the g4-pawn falls, it’s curtains for
White) 22 ... Kh8 (admittedly this is a strange spot to put one’s king if the opponent has a bishop
staring at it from d4 but e5-e6+ isn’t available, and Black can eliminate the d4-bishop any time he
pleases with ... Nxd4) 23 h3 Na5 24 Rf3 and now White will just sit with rooks doubled on the f-file,
hoping to enforce f4-f5 at some point. It doesn’t look all that easy for Black to make progress.
20 fxg5
After 20 g3 Rf5 21 Kh1, Black has 21…Bf8! intending ... Bh6, which adds
further pressure to bear on White’s now-wobbly centre.
20 ... Bxg5
“A good servant must anticipate the master’s needs,” recites the g5-butler,
who has his work cut out for him protecting his king. Houdini prefers the
counter-intuitive 20 ... Rxf1+! 21 Rxf1 Bxg5 perhaps because Black’s rook is
now free to roam away from its b7-burden. Also, Black’s king isn’t displaced
in this variation.
21 Rxf8+ Kxf8?!
Up until now, Shirov has handled Black’s side in model fashion. The b7-
pawn isn’t important and Black should let it go with 21 ... Rxf8! 22 Rxb7 Qf5
23 Rb1 Nxd4! 24 cxd4 Qf4 25 c3 Qe3+ 26 Qxe3 Bxe3+ 27 Kh1 Bd2 28 Rd1
Bxc3 29 h3 Rf2 30 Rd3 Rc2! 31 Kh2 Kf7 and White is busted in the ending,
since his pawns will soon fall.
22 Qh5 Qg6?!
Now clocks get low, and mistakes and inaccuracies bloom from both sides.
More accurate may be 22 ... h6! 23 h4 Be7 24 Rf1+ Kg8 25 Rf4 Rc8 and
White has no way to continue the attack, since 26 Rg4+? Kh7 doesn’t bother
Black at all.
23 Qf3+! Kg7
The only way to play for a win as 23 ... Qf7 24 Qg4 Qg6 25 Qf3+ merely
repeats the position.
24 Qxd5 Qxc2 25 e6+
The long suffering bishop is released from his penance.
 25 ... Kg6?!
The king attempts to muster aplomb for the grim task of survival which
lies ahead of him, but picks the wrong square. Black’s earlier 19 ... g5! thrust,
which once appeared so ingenious, now displays its darker side, showing
cracks of uncertainty. When you deliberately expose your king, the following
moves must be played at a ruthlessly accurate level.
Instead, 25 ... Kh6! was correct. Now if 26 Re1 Nxd4 27 cxd4?? (27 Qxd4
Rd8 28 Qe5 Rf8! 29 Qd4 Qd2 30 Qxd2 Bxd2 31 e7 Re8 32 Re6+ Kg5 and
White can resign) 27 ... Rf8 (there is no defence to the ... Qf2+ threat) 28 Kh1
Qf2 29 Rg1, White gets mated after 29…Qxg1+! 30 Kxg1 Be3+ 31 Kh1 Rf1.
26 Rf1?!
White had to try 26 Re1 Qf5 27 Qb3 Rf8 28 Qd1 h5 29 Qe2 with some
chances to survive.
26 ... Qd3?
Ah yes, the joys and sorrows of rapid chess, where GMs and patzers alike
make fools of themselves when clocks run low. The evaluation swings by a
whopping nine points after Shirov’s last move.
Black missed 26 ... Nxd4! 27 Qxd4 (27 cxd4? is met with the back rank
trick 27 ... Be3+ 28 Kh1 Rf8! when White’s rook is out of safe squares and
after 29 Ra1, 29…Bc1! forces mate) 27 ... Qe2 28 Qd1 Be3+ 29 Kh1 Qxd1
30 Rxd1 Bxc5 and Black consolidates his extra piece.
 Exercise (critical decision): An attack, once commenced, is a self-sustaining
entity which feeds upon its own success. In one move, the evaluation
shifted
from losing to winning for White. White’s choices: a) 27 Qd7, intending
Qf7 mate! b) 27 Rf3, lifting another attacker into the mix. c) 27 h4, giving the
g5-defender a bump. One line loses, one line draws and another line wins.
We must be like
Goldilocks and pick the one which is ‘just right’. Which one would you
play?

27 Qd7?
Confusion to the nature of the position’s truth seems to be both players’
lots, as clocks run low and tensions run high. Now Black is better again.
Answer: White missed 27 h4! which disrupts Black’s defensive flow and continues to coerce and
generally inconvenience Black’s king. After 27 ... Bxh4, 28 e7! leaves Black’s king fatally exposed
after 28 ... Nxd4 29 Qf7+ Kh6 30 Qf8+! Kg6 31 e8Q+ Rxe8 32 Qxe8+ Kh6 33 Qf8+ Kg6 34 Qd6+
Kh5 35 g4+! Kxg4 36 Qf4+ Kh5 37 cxd4 and White wins. If he had played 27 Rf3 instead, he will only
get a draw after 27 ... Qd1+ 28 Kf2 (28 Rf1?? walks into the dirty trick 28 ... Be3+! winning on the
spot) 28 ... Qd2+ 29 Kf1 (29 Kg3?? Qe1+ 30 Bf2 Qe5+ 31 Qxe5 Nxe5 and White’s unfortunate rook
has no place to go) with perpetual check.
27 ... Be3+?
Black defenders drift aimlessly as snowflakes. A misguided winning
attempt. Black should have settled for 27 ... Nxd4 28 Qf7+ Kh6 29 cxd4 Qg6
(or 29 ... Qxd4+ 30 Kh1 Qe3 31 Rf6+ Bxf6 32 Qxf6+ Kh5 33 Qf5+ Qg5 34
Qxh7+ Kg4 35 Qe4+ Qf4 36 Qg6+ Qg5 and the game ends in perpetual
check) 30 d5 Be3+ 31 Kh1 Qxf7 32 Rxf7 Bxc5 33 h4 Kg6 34 g4 Rd8 35 Kg2
Rxd5 36 h5+ Kg5 37 e7 Rd2+ 38 Kf1 Bxe7 39 Rxe7 Rb2 40 Rxh7 Kxg4 41
a4 with a likely draw.
28 Bxe3 Qxe3+ 29 Kh1 Rf8?

Exercise (combination alert): Shirov’s last move hopes to capitalize on

White’s weak back rank, which however, isn’t as weak as it appears.
White
has a shocking way of short circuiting Black’s plan. White to play and
win:

30 Rg1?
Short must have been very low on the clock.
Answer: Queen sacrifice/pawn promotion.
Both sides missed 30 Qf7+! (recent events add misery to the black king’s
already cheerless disposition) 30 ... Rxf7 31 exf7 Qxc5 32 f8Q Qxf8 33 Rxf8
when Black can resign.
30 ... Ne5
 Black’s most accurate consolidation method is 30 ... Rf6! 31 Qxb7 Qxe6.
31 Qd6
After 31 Qd4, Black has 31 ... Qe2 32 h3 Nd3 33 Qd6 Rf6 34 Qg3+ Kh5
35 Qc7 Qe4! (guarding both the b7- and h7-pawns, as well as taking aim at
the white kingside) 36 Kh2 Nf4 37 Qg7 Rxe6 and Black should consolidate.
31 ... Rf6?
Correct was 31 ... Re8!, keeping the e-pawn honest.

Exercise (combination alert): Black’s last move was inaccurate.

White to play and force a draw with a multi-themed forcing line:

31 h3?
Back to losing! If two world-class GMs can blunder so liberally in a
Cochrane Gambit game, then for us, the ordinary, we had better prepare for a
blunderfest of epic proportions.
Answer: Queen sacrifice/pawn promotion/smothered mate/pin/perpetual check.
White missed 32 e7! (the queen cannot be captured due to the promotion
zwischenzug check) 32…Ng4! (threatening a smothered mate on the move)
33 Qg3 (pinning Black’s knight, whose mouth droops in disappointment from
having been robbed of his f2-mating target) 33 ... Qxg3! (Black’s queen
flaunts insolent defiance, allowing White to promote to a new queen with
check) 34 e8Q+ (Hi! I’m back!) 34 ... Kg5 35 Qg8+ Rg6 36 Qxg6+ (instead,
36 Qd8+ Nf6! and White is unable to capture on g3, due to 37 hxg3?? Rh6
mate!) 36 ... hxg6 37 hxg3 Nf2+ 38 Kh2 Ng4+ 39 Kh3 Nf2+ ending in an
odd perpetual check. What an amazing variation!
32 ... Nd3!
The mating threat on f2 consolidates Black’s position.
33 Qd5

Exercise (combination alert): 33 ... Nxc5 should consolidate and win.

But Black has a much more deadly method to end resistance. How?

33 ... Nxc5
The prosaic continuation.
Answer: Mating threat/removal of the guard.
More effective is 33 ... Qg3! (threat: ... Nf2 mate!) 34 Rb1 (now Black has
a free hand to attack) 34 ... Nf2+ 35 Kg1 Nxh3+ 36 Kh1 Nf2+ 37 Kg1 Ng4
38 Qe4+ Kg5 39 Qd5+ Kh4 40 Qd6 Qf2+ 41 Kh1 Qf1+ 42 Rxf1 Rxf1 mate!
34 e7 Qxe7 35 Qg8+ Kh6 36 Rd1 Ne4 0-1
Finally, Black consolidated his material gains and White has to lay down
his arms.
Summary: Some writers claimed that the Cochrane gambit is unsound – which may well be so on
the omniscient comp level. For us easily confused humans, consolidation of our extra piece won’t be so
easy. So let’s respect the Gambit and study it well.

Game 2
D.Reinderman-M.Bosboom
Groningen 1995

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nxf7

White continues with the Cochrane Gambit philosophy: When persistent,
reasoned argument fails, then what other choice do we have but the use of
direct force?
4 ... Kxf7 5 Bc4+?!

This path is unlikely to conform to White’s attacking desires. The move

alters the sacrifice’s original intent into an unintelligible facsimile. Sakaev
calls this move “stupid”, leading “to a position in which White is almost
immediately lost!” and so something is definitely wrong here.

Question: Why is the move so bad? Doesn’t it bring out a piece with check?

Answer: It also allows Black a strong central counter with our next move, which effectively clogs
White’s attacking chances, while allowing Black freedom and easy development.
5 ... d5!
In the martial arts, one is taught: To shift a powerful oncoming force, we
only require a mild counterforce to send it spinning off track. This is pretty
much a refutation of White’s last move, which clears the d6-square for our
bishop and clogs the a2-g8 diagonal.
6 exd5
White has also tried the bishop retreat 6 Bb3 Bg4 (6 ... c5! is an unplayed
idea which looks rather harsh for White as Black threatens ... c5-c4) 7 exd5
Qe8+! 8 Kf1 (8 Qe2 Bd6 9 d3 Qxe2+ 10 Kxe2 Re8+ looks like a lost ending
for White who has two unimpressive pawns for the piece) 8 ... Bd6 9 d4 Na6
10 Na3 cxd4 11 Qxd4 Rf8 12 Nc4 Bc5 (White clearly lacks compensation for
the piece, since Black is about to consolidate with ... Kg8 and then turn his
attention to White’s unfortunate king with ideas like ... Ng4, homing in on the
f2-pawn) 7 f3 Be6 8 e5 (8 d4 c5! also looks highly favourable for Black, who
isn’t even behind in development) 8 ... Nh5! (seizing control over the f4-
square and threatening an annoying check on h4) 9 0-0 g6! (preparing the
overprotection of the f5-square) 10 d4, S.Reinhardt-S.Eichner, Goerlitz 1999.
White lacks compensation for the piece after 10 ... c5 11 f4 Ng7 12 c3 Nc6.
6 ... Bd6 7 0-0 Rf8 8 d4 Kg8

Question: I don’t see even an iota of compensation for White’s piece shortage. Though he has three
pawns for it, they look ineffective. Is he busted?

Answer: The position looks like something arising from a Ruy Lopez Marshall Gambit, where
Black often gets compensation for a missing pawn in the form of a developmental lead and open lines
to White’s king. Here, however, Black is the one up a piece for three pawns, while he still enjoys all the
Marshall Gambit-like benefits. Conclusion: White’s opening has been an absolute disaster.
9 h3
After 9 Nc3 a6 10 Bg5 Bf5 11 Re1 Nbd7 12 Bb3, M.Ganzer-K.Holschuh,
correspondence 2003, White clearly lacks full compensation for the piece,
following 12 ... h6 13 Bh4 b5.
9 ... Bf5 10 Bb3

Question: Why did White retreat the bishop, rather than continue developing?

Answer: White is desperate to activate his central majority with c2-c4-c5, so he clears the way for
his plan.
10 ... b5!?
Attempting to mess with White’s intent by fighting for control of the c4-
square. More promising is to simply allow White to execute his plan and
instead go for a direct assault on White’s king with 10 ... Qe8! and if 11 c4?,
Black has 11…Qg6 with the threat of ... Bxb1 as well as ... Bxh3.
11 a4 bxa4
Black can also play to slow White’s majority-activation plan down after 11
... b4 12 c4 bxc3 13 bxc3 Nbd7 14 c4 Bb4! 15 Bd2 Rb8 when I fail to see
White’s compensation.
12 Rxa4 Qe8!
 The queen remains outwardly respectful of White’s king, yet mocking and
plotting against him in the isolated privacy of her mind. Bosboom identified
the correct plan, which is to simply build up a kingside assault, to which
White has no reasonable defence. What’s ironic is that White can’t even try
to mobilize his central pawn mass with 13 c4?? as the Nb1 would hang.
13 Nd2 Qg6
Threat: ... Bxh3.
14 Kh1

Exercise (planning): It grows apparent that White is unable to

effectively govern the territory adjacent to his king. Come up with a
clear plan for Black to stage an assault on White’s lightly defended king:

Answer:
Step 1: Seize control over the g4-square by offering a piece, which either
weakens White’s king further, or forces him into a pin.
14 ... Bg4!
The presumptuous bishop’s audacity makes White’s queen uncomfortable.
Well, the bishop isn’t a likely candidate for a commendation for bravery in
battle, mainly since his offer of his life cannot be accepted by White, without
suffering painful consequences.
 15 Nf3
Probably the best of White’s miserable choices, regretfully played with the
philosophy: When we are unable to fix a defect in our position, the next best
thing is to try and conceal it. White must have agreed to walk into this pin
with a heavy heart, since the alternatives lose:
a) 15 hxg4?? Qh6+ 16 Kg1 Qh2 mate!
b) 15 f3 Bd7 16 Ra1 Qg3 17 f4 Bxf4 18 Nf3 Bd6 (even stronger than 18 ...
Bxh3 which also wins) 19 c4 (or 19 Ne5 Bxh3 20 gxh3 Qxh3+ 21 Kg1 Qg3+
22 Kh1 Nbd7 23 Qf3 Ne4! and Black wins) 19 ... Ng4! and if 20 hxg4, 20…
Rxf3 forces mate.
c) 15 Qe1?? Re8 costs White heavy material.
15 ... Qh5?!
Black stumbles, missing:
Step 2: 15 ... Bxf3! (lure White’s queen to the f3-square, after which Black
gains numerous attacking tempi) 16 Qxf3 Ne4 17 Qe2 (or 17 Qe3 Nd7 18
Kg1 Rae8 19 Qd3 Rf5 and if 20 c4?, 20…Nxf2! forces mate after 21 Rxf2
Re1+) 17 ... Qg3!! (this is the kind of sweet shot we all pray for in our
games) 18 f4 (also, 18 fxg3 Nxg3+ 19 Kh2 Nxe2+ is game over for White
top) 18 ... Qxb3! and Black wins by clearing the g3-square for a knight fork.
16 Kg1! Nbd7!
Black correctly rejects the ending line 16 ... Bxf3?! 17 Qxf3 Qxf3 18 gxf3
Nbd7 19 c4 when White’s central pawns are activated and he may even
generate play on Black’s isolated a-pawn.
17 Re1
After the howler 17 hxg4??, Black wins with Nxg4 18 Re1 Bh2+ 19 Kf1
Bg3! 20 d6+ Kh8 since 21 fxg3 is met with 21 ... Qh1+ 22 Ke2 Qxg2+ 23
Kd3 Nf2+ with a royal fork.
17 ... Kh8
Now would be a good time to cash out with 17 ... Bxf3! 18 Qxf3 Qh4 19
Be3 Nb6! 20 Raa1 Kh8 and there is no defence to the coming ... Nfxd5,
picking up the d5-pawn.
18 hxg4!?
Reinderman succumbs to temptation. Declining the proffered bishop was
no picnic either, as after 18 Re3 Bxf3 19 Rxf3 g5! 20 Qd3 g4 21 Rf5 Qe8 22
Kf1 Qg6 23 hxg4 Nb6 24 Ra1 Nxg4 25 Rxf8+ Rxf8 26 Qxg6 Rxf2+ 27 Ke1
hxg6 28 Rxa7 Rxg2, White can’t save himself.
 18 ... Nxg4
White is now up two pawns. However, Black’s attack is worth more than
those bits.
19 Re4
Shielding the e-file with 19 Be3 doesn’t change much as after 19…Rae8,
Black still has a winning attack.

Exercise (combination alert): Find Black’s most promising attacking continuation.

19 ... Bh2+
A hissing projectile grazes the king’s face, leaving a deep scar as a
permanent reminder. This is not bad, but Black can do better.
Answer: Much stronger was the 19 ... Nh2! pin when after 20 Re3 Nxf3+ 21 Rxf3 (21 gxf3 Bf4! is
decisive as well) 21 ... Rae8 (threatening mate in two moves, starting with a queen check on h2) 22 Be3
Rxe3! (removal of a key defender) 23 fxe3 (or 23 Rxf8+ Nxf8 24 Qxh5 Re1 mates) 23 ... Qh2+ 24 Kf2
Bg3+ 25 Kf1 Qh1+, Black wins.
20 Kf1 Ndf6 21 Re6
 Exercise (combination alert): Find Black’s next shot which will generate a
winning attack (Hint: This theme has already been displayed in the notes
above):

Answer: Clearance/double attack.

21 ... Bg3!
The bishop continues to dance elusively out of reach of those who wish
him harm. White is unable to meet Black’s dual threats on f2 and h1.
22 Be3
Defence of the f2-pawn represents the greater urgency. White loses
quickly after22 fxg3?? 22 ... Qh1+ 23 Ng1 Ne4+.
22 ... Nxe3+ 23 fxe3 Ng4!
The text move is even stronger than 23 ... Qh1+ 24 Ke2 Qxg2+ 25 Kd3
Ng4 26 Qe2 Nf2+ 27 Kc4 Rxf3 28 Rxa7 Raf8 which should also win.
24 Ke2
 Exercise (combination alert): Black’s bombing campaign isn’t carried out in
precise strikes, but rather in megatonnage, dropped randomly over White’s
territory, without regard to targeting or civilian casualties. Continue
Black’s attack:

Answer: Removal of the guard/clearance/attraction.

24 ... Rxf3! 25 Rxa7
A good try, playing on Black’s weak back rank. The alternatives are
equally bleak anyway:
a) 25 gxf3 Qh2+ 26 Kd3 Nf2+ forks king and queen.
b) 25 Kxf3 Rf8+ 26 Ke2 Nxe3+ wins White’s queen and forces mate.
25 ... Rff8
The human move. The showoffy comp found 25 ... Rxe3+!! 26 Rxe3 (26
Kd2 Rxa7! 27 Rxe3 Bf4 pins the rook) 26 ... Nxe3+ 27 Kxe3 Re8+ 28 Kd2
Bf4+ 29 Kc3 Qxd1 and Black wins.
26 Qa1 Rxa7 27 Qxa7 Qh1!
The queen exudes an aura of expertise in the art of getting what she wants
out of life.
28 Kd3
 “Conniving hag! You will not get your way this time!” declares the
slippery king, whose prediction turns out to be inaccurate.

Exercise (combination alert): This one doesn’t require the analytical

abilities of Sherlock Holmes or Monsieur Poirot. Even Dr. John Watson
(I mean Sherlock Holmes’ sidekick, not the notable chessologist, IM John
Watson who lives three blocks from me, and who could solve the exercise
as well) can solve it. White to play and force mate in three moves:

Answer: 28 ... Qf1+!

“Ah, me, what a day! A woman’s work is never done,” exclaims the
exhausted queen, who is finally close to completing her execution quota for
the day.
29 Kc3
It becomes quite clear that White’s king never really had a knack for the
ladies:
a) 29 Ke4 Nf2 mate! The f2-executioner looks upon White’s king with the
impersonal appraisal of a bored clerk who is just doing his job.
b) 29 Kd2 Be1+ 30 Kd1 Bc3 mate!
29 ... Be1 0-1
Summary: 5 Bc4? (Yes, I downgraded it to a full question mark) is the worst of White’s fifth move
options in the Cochrane Gambit. Black does indeed come close to reaching a winning position after the
powerful central counter 5 ... d5!.

Game 3
V.Topalov-V.Kramnik
Linares 1999

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nxf7

I just can’t wrap my head around why so many people are willing to enter
this semi-disreputable sacrifice. Why risk all for a mere passing thrill? I
suppose the answer is: When in pursuit of our dreams, do we not all at times
grow impervious to reason? Topalov thinks to himself: “Let’s play a semi-
shady gambit on the world’s leading authority on Petroff’s Defence” – not to
mention a guy who was to become world champion the following year.
4 ... Kxf7
The knight’s death is merely a tiny fragment of Topalov’s larger design,
which is to take out a world class opponent with a notorious opening line,
which in turn would yield rich psychological benefit for future encounters –
but only if he succeeds, which he partially did by drawing the game.
5 Nc3

Question: What is White’s idea behind White’s move?

Answer: It’s played with the idea that Black is unable to undermine a centre which doesn’t yet
exist. By holding back on d2-d4, White keeps us guessing as to the final formation of his centre. White
can play for Bc4+, followed by d2-d4, or even d2-d3 and f2-f4.
5 ... c5
Given an exclamation mark by Sakaev. It’s still in our best interest to
challenge the d4-square, whether or not White occupies the square with a
pawn or not. However, Black has reasonable alternatives in:
a) 5 ... Be6 (Black cuts out Bc4+ once and for all) 6 d4 g6 7 Be2 Bg7 8 f4,
J.Tait-A.Richards, correspondence 1999. Oh, the shame of it! The man
playing White is one of my editors, Jonathan Tait, apparently caught red-
handed playing the shady Cochrane Gambit! However, I must add that he
went on to win, so we are once again reminded of just how dangerous this
gambit really is over the board, despite what the comps say. I still prefer
Black after 8 ... Re8 9 0-0 Kg8 as White’s one pawn just doesn’t feel like it’s
enough for a full piece, no matter how impressive his central domination
seems.
b) 5 ... c6 (in order to meet Bc4+ with ... d5. 6 d4 Be7 7 f4 Re8 8 Bd3)
“donalddraper (GM)”-C.Lakdawala, Internet (blitz) 2014. Black should
continue 8 ... Bg4 9 Qd2 Kg8 10 0-0 Na6 11 h3 Be6 12 Ne2 when White will
achieve some kingside play with f4-f5 and maybe even g2-g4.
c) 5 ... Be7 6 d4 Re8 7 Bc4+ Be6 8 Bxe6+ Kxe6 9 Qf3 Kf7 10 Be3 Nc6 11
g4 (threat: g4-g5) 11 ... Kg8, M.Santo Roman-A.Hauchard, Toulouse 1995.
White certainly gets reasonable practical chances after 12 g5 Nd7 13 Rg1
with a long term kingside initiative for the piece, although objectively, Black
must still stand better.
6 Bc4+
The only move. Instead, 6 d4?! Nc6 7 Bc4+ (7 dxc5?! Be6 8 cxd6 Bxd6
looks terrible for White, who is down material and behind in development) 7
... d5! 8 Nxd5 Be6 transposes to a note we covered in Short-Shirov (Game 1)
where Black holds an undisputed advantage.
6 ... Be6

Question: Does our standard ... d5 counter work for us here?

Answer: The ... d5 counter doesn’t work as well here.

6 ... d5?! has never been tried here, and for good reason. After 7 Nxd5 Be6
8 0-0! (8 Ne3 looks fine for White as well) 8 ... Bxd5 9 exd5 Bd6 10 b3 Nbd7
11 f4 Re8 12 Bb2 a6 13 Qf3 b5 14 Bd3 c4 15 Bf5 (15 bxc4?? bxc4 – Black
threatens ... Qb6+, as well as ... cxd3 – 16 Bxf6 Qb6+ 17 Kh1 cxd3 18 Bc3
dxc2 looks bad for White) 15 ... Kg8 16 Rae1, White has three pawns for the
piece and approximately even chances.
7 Bxe6+ Kxe6
Black’s king reminds us of the excited game show contestant who is told
by the emcee to “Come on down!” to the stage.
8 d4
White’s main move.
Instead, 8 g4!? is an interesting novelty with the idea of displacing Black’s
f6-knight, which then enables Qg4+.

Question: Is g4-g5 really a threat?

Answer: It may not be. Perhaps Black can just keep developing with:
a) 8 ... Nc6!? 9 g5. Now the counterintuitive 9 ... Ng8!? is also possible. The idea is to be able to
challenge White’s control over the d5-square with …Nge7. After 10 d3 Kd7 11 Qg4+ Kc7 12 Be3 Rc8
13 0-0-0 Kb8 14 f4 Nge7 15 h4 g6, White has almost, but not quite enough compensation for the piece,
since Black’s king is now perfectly safe.
b) Black can also nip the g-pawn advance with 8…h6 9 d3 Nc6 10 f4 Kd7!
(the queenside clearly feels like safer territory for Black’s king than the
hostile kingside) 11 Be3, M.Molner-B.Finegold, Saint Louis 2012. Black has
unravelled and is on his way to consolidation after 11 ... Kc7 12 g5 hxg5 13
fxg5 Ne8 14 Nd5+ Kb8 15 Qe2 Nc7.
8 ... Kf7!
Kramnik is understandably anxious to escort his king from the open
centre. His move is more accurate than 8 ... cxd4 9 Qxd4 Kf7 10 Qc4+ d5 11
Nxd5 Nxd5 12 0-0 Na6 13 Rd1, V.Destruels Moreno-B.Lindstrom,
correspondence 2007. The position feels dynamically balanced after 13 ...
Rc8 14 Qe2 Qe8 15 Rxd5 Kg8 16 Bf4 h6 17 Rad1 Kh7.
9 dxc5 Nc6 10 Qe2 Qd7!
Black’s key to survival of the Cochrane Gambit is to fight for control over
the light squares and this gives the queen access to the important e6-square.
11 Be3!

Topalov’s move, providing the option of castling queenside, is superior to

11 0-0 Re8 12 cxd6 Bxd6 13 f4 Bb4 14 e5, C.Koch-F.Hoegerl,
correspondence 2006. Now Black should probably play 14 ... Qe6 15 Qf3
Bxc3 16 bxc3 Nd5 17 Ba3 g6. White has full compensation for the piece,
with three pawns and a slightly insecure black king.
11 ... dxc5
I prefer Black’s position. Kramnik isn’t going to allow White to castle
long so he picks up another pawn at the cost of time. Also, a coming ... Nd4
may irritate White.
12 f4
 A looming pestilence threatens to envelop the countryside around Black’s
king. Topalov activates his 4:2 central/kingside pawn majority. Also, 12 Rd1
Qe6 isn’t a problem for Black.
12 ... Re8
Kramnik seeks to induce e4-e5, which although disruptive for Black, also
weakens White’s light squares.
13 e5 Ng4 14 Rd1 Qf5!

Question: Why not post the queen to the e6-square, where it covers against Qc4+?

Answer: It’s a trap! When our position slowly overcomes a long, painful strategic illness, the thing
we dread most is a relapse:
The problem with 14 ... Qe6? is that it allows White to open lines with 15
f5! Qxe5 (15 ... Qxf5?? hangs the queen to 16 Rf1) 16 Qc4+! Ke7 17 Qxg4!
and if 17 ... Qxe3+, Black is busted after 18 Kf1 Ne5 when:
a) 18 ... Kf7?? 19 Qc4+ wins heavy material, due to a coming d5-knight
fork.
b) 18 ... Qh6 19 Qe4+ Kf7 20 Qd5+ Kf6 21 Ne4+ Rxe4 22 Qxe4 Nd4 23
c3 Nxf5 24 Re1 (Black’s king finds himself in a desperate situation as after
24 ... Ng3+ 25 hxg3 Qxh1+ 26 Kf2, 26…Qxe1+ is forced and Black is
unlikely to erect a saving fortress after 27 Kxe1 Bd6 28 g4 Rb8 29 Qxh7) 19
Nd5+ Kf7 20 Qh5+ g6 21 fxg6+ hxg6 22 Qxh8 Qg5 23 Qh3 and Black lacks
compensation for his material deficit.
15 0-0
 Exercise (planning): There are long, dusty roads yet to be
travelled before Black’s hopes of coordination achieve full
implementation. How did Kramnik make use of the fragmented data
and cobble together an effective unravelling defensive plan for Black:

15 ... h5!
Answer: Kramnik decides to exercise his authority by jealously hoarding the central and kingside
light squares, the underlying theme which saturates his coming moves. Kramnik inoculates himself
from future h2-h3, g2-g4 pawn storm ideas. The move does weaken the g5-square, yet the move’s
virtues overrule this minor concession. Kramnik’s move has a dual purpose:
1. As mentioned, it seizes further control over the kingside light squares.
2. Black prepares to develop his h8-rook, via the h6-square.
We sense a quality of stark simplicity in Kramnik’s defensive idea, which
implies efficiency, despite its outer appearance of limited means.
16 Bc1?!
The freeloading bishop hasn’t earned an honest living since his mother
paid him a quarter to take his cod liver oil when he was seven years old.
White loses the initiative with this meek retreat. He should try 16 h3 Nxe3 17
Qxe3 Rh6 18 Ne4 Kg8 19 c3 when he remains only slightly worse.
16 ... Nd4 17 Qc4+ Kg6!?
Kramnik’s dexterous king hopes to weave his way around the dangers. My
inclination would be to head for the g8-square. Black is on his way to
consolidation after 17 ... Qe6 18 Qd3 Kg8.
18 h3 Nh6 19 Nb5!?
Topalov realizes that the d4-irritant must be removed if he is to generate
counterplay. Moreover, after 19 Be3 Nxc2 20 Bf2 Nd4 21 Qd5 b6 22 Bxd4
cxd4 23 Ne4 Qe6 24 Qxd4 Be7 25 Qd3 Qf5 26 Qf3 h4, I still don’t sense full
compensation for the piece.
19 ... a6!
Up until now, the queenside has been as devoid of activity as a bank on
Christmas day. Black is not adverse to a knight trade as the resulting
exchanges will allow him to hit the white queenside. Also strong is 19 ...
Nxc2! 20 Nxa7 Be7 21 Nb5 Rd8.
20 Nxd4 cxd4 21 Qxd4

One glance and we clearly see that White’s kingside inquiries lead to a
dead end, and it is Black who rules this sector of the board.
21 ... Rc8?!
Black sets up ... Bc5 and ... Rxc2 threats. However, I don’t think it’s
necessary to give White a third pawn for the piece. Black looks clearly better
after the more direct 21 ... Qxc2! 22 Qb6+ Qc6. (the point: Black hangs on to
his b-pawn in this line) 23 Qxc6+ bxc6 24 Be3 Nf5 and White’s boneless and
filleted position sags further. He will be hard-pressed to hold this ending.
 22 Qb6+ Kh7 23 Qxb7 Rxc2 24 Be3
Now White looks only slightly worse.
24 ... Qg6
Covering the a6-pawn, while eyeing the g2-soft spot.
25 Rc1!
Eliminating Black’s most active piece. White’s position, now dangerously
near the ‘busted’ mark, somehow manages to coordinate, despite every
indication of imminent collapse.
25 ... Rxc1 26 Rxc1 Nf5
I slightly prefer Black after 26 ... Qd3 27 Bf2 Nf5 28 Rc3 Qd1+ 29 Kh2
Qe2 30 Qd7 Qe4 31 Qd2 Be7.
27 Bf2 h4 28 Rc7
After 28 Rc6 Qh5 29 Qxa6 Be7 30 Qd3 Rd8 31 Qe4 Rd2 32 b4 Rxa2, the
game is probably still slightly in Black’s favour.

Exercise (planning): With his last move, Topalov disallows

Black’s kingside to develop. How did Kramnik force a draw?

Answer: Take advantage of White’s back rank. Black threatens a nasty queen check on b1.
28 ... Ng3!
Menacing a dastardly mate from the h1-square.
 29 Kh2!
Most certainly not 29 Bxg3?? hxg3 (threatening a big check on b1) 30 Rc1
Bc5+! 31 Kh1 Be3 32 Rd1 Bxf4 and Black is winning.
29 ... Nf1+ 30 Kg1 Qb1
Threatening mate, with ... Ng3+ next. Black’s queen feels it would be a
better world with fewer white kings living in it.
31 Bxh4 Bc5+ ½-½
This move sets into motion a self-perpetuating cycle. There will be no
negotiated terms of surrender for either party in this game. After 31 ... Bc5+
32 Rxc5, 32…Nd2+ forces 33 Kh2 (33 Kf2?? Ne4+ 34 Ke3 Qg1+! 35 Kf3
Qf1+ 36 Ke3 Nxc5 and Black wins) 33 ... Nf1+ 34 Kh1 (the king describes
his recent conduct as a “strategic retreat,” while everyone else calls it
“running for his life”) 34 ... Nd2+ with perpetual.
Summary: 5 Nc3 may be White’s least unsound line in the Cochrane Gambit
– if we can call it that – but I feel it still falls just short of equality for White
if Black plays the opening perfectly, as Kramnik did in this game.
Chapter Two
The Scotch Petroff

We arrive at our first real test of the book, which is commonly known as
the Steinitz variation. However, I prefer to call it the Scotch Petroff since
White declines to play into the even more heavily analysed 3 Nxe5 lines, and
instead opens up the position with 3 d4 in Scotch Opening fashion. White
plays this way, mainly hoping to detach us from the familiar contours of the 3
Nxe5 line. We can reach the following two diagrams:
 In this version, White creates an immediate imbalance of kingside versus
queenside pawn majorities. The structure resembles an Open Ruy Lopez, and
we get fully equal chances in the sharp middle game to come.

This is the main line of the chapter. A conspicuous anomaly juts out: the position is symmetrical,
except for the fact that White has the extra move Bd3. According to current theory, this factor alone
isn’t enough to provide White a stable edge.
Game 4
D.Jakovenko-Wang Yue
Pearl Spring Chess Tournament, Nanjing 2009

1 e4 e5 2 Nf3 Nf6 3 d4
We enter a major branch, the ‘Scotch’ Petroff. White seeks a livelier open
game and plays an early d2-d4 in Scotch Opening fashion.

3 ... Nxe4
Black’s main move. Due to space concerns, we won’t be covering either:
a) 3 ... exd4 4 e5 Ne4 5 Qxd4 d5 6 exd6 Nxd6 7 Nc3 Nc6 8 Qf4,
G.Kasparov-A.Karpov, World Championship (Game 10), New York 1990.
This is the kind of open position with slight development lead that White
generally wants when entering the Scotch Petroff. Black can probably
equalize but must be exceedingly accurate to achieve that.
b) 3 ... d6 is an odd Philidor’s/Petroff hybrid line . I think 4 dxe5! (4 Nc3
Nbd7 transposes to the Hanham Philidor’s) puts some degree of pressure
upon Black. Following 4 ... Nxe4, 5 Qd5! transposes to the Philidor line more
commonly reached after 1 e4 e5 2 Nf3 d6 3 d4 Nf6 4 dxe5 Nxe4 5 Qd5. After
5 ... Nc5 6 Bg5 Qd7 (6 ... Be7 7 exd6 Qxd6 8 Nc3 leaves White with a
nagging developmental lead) 7 exd6 Bxd6 8 Nc3 0-0 9 0-0-0, White has a
meaningful development lead in this open position, R.Byrne-T.Petrosian,
Biel Interzonal 1976.
 4 dxe5!?

This move, which used to be just a surprise weapon, is now growing in

popularity.

Question: What is White’s intent?

Answer: White’s reasoning for the move:

1. White creates a slight imbalance in the pawn structure, hoping the e5-
point may lead to a future kingside attack.
2. White doesn’t want to lose time capturing on e5 with his knight, only to
get it booted off by d7-d6. We look at the White’s main line 4 Bd3 for the
bulk of the chapter.
4 ... d5
Black’s soundest response. Alternatives:
a) 4 ... d6 5 Qd5! transposes to that rather unpleasant Scotch/Philidor
hybrid mentioned in the above notes.
b) 4 ... Nc6 doesn’t make sense to me as it avoids our freeing ... d5-break.
After 5 Bc4 Nc5 6 Qd5 Ne6 7 0-0 Be7 8 Rd1, Black’s development remains
bottled up and ... d7-d6 will be dangerous with White leading in
development, A.Strikovic-G.Pestic, Internet (blitz) 2003.
c) In Game 8, we will look at the ultra-sharp 4 ... Bc5!?.
 5 Nbd2
Now we reach opposite-wing pawn majorities, and the position begins to
resemble an Open Ruy Lopez. In the next game, we will look at 5 Bd3.
5 ... Nc5
First played by GM Andor Lilienthal in 1938. Let’s look at other tries:
a) I have my doubts about Black’s alleged equality after 5 ... Be7 6 Nxe4
dxe4 7 Qxd8+ Bxd8 8 Nd4 c6 9 Bf4 Ba5+ 10 c3 Bc7 11 Bc4 0-0 12 Bg3,
S.Karjakin-Wang Yue, Beijing 2011. Black failed to equalize. Not only are
e5-e6 tricks in the air, White continues to lead in development.
b) 5 ... Nxd2.

Question: Doesn’t this help White develop?

Answer: If Black refuses to move his knight, then he must deal with Nxe4 consequences. So Black
simply swaps on d2, reasoning that every swap decreases the intensity of White’s future kingside
attacking plans.
b1) 6 Qxd2! (the idea is to transfer the queen to f4, which may get Black’s
king nervous later on if he chooses to castle kingside) 6 ... Be7 7 Bb5+ (7 Qf4
is also logical) 7 ... c6 8 Be2 0-0 9 0-0 f6 10 c4 fxe5 11 cxd5 cxd5 12 Nxe5
Nc6 13 Nxc6 bxc6 14 b3 c5 15 Ba3 Bb7 16 Rac1 Rc8 17 Bg4 Rc7 and Black
achieved dynamic equality, P.Svidler-V.Kramnik, Dos Hermanas 1999.
b2) 6 Bxd2 Nd7!? (a rare move, hoping to throw White off his intended theoretical track) 7 Qe2!?
Nc5 8 0-0-0 Be7 9 Nd4 0-0 (the game sharpens considerably) 10 g4!? (to enable a future Nf5) 10 ...
Ne4, V.Gashimov-B.Gelfand, Odessa 2009. White looks a shade better after 11 Be1! with the idea of
f2-f3 next.
6 Nb3
White unravels, preparing to challenge Black’s c5-outpost next with Be3.
6 ... Ne6

“The play of both sides looks very unusual. Instead of developing the other
pieces, both sides play with just one knight,” writes GM Viktor Mikhalevski.

Question: Isn’t Black’s e6-knight blocking development?

Answer: Black can always fianchetto on the queenside. He prepares ... c7-c5 and ... Nc6, annexing a
good chunk of the centre. Other alternatives:
a) Also possible is the simplifying 6 ... Nxb3 which I don’t like, since it
improves White’s structure by opening the a-file. After 7 axb3 Be7 8 Bd3
Nd7 (intending to chase White’s bishop with ... Nc5) 9 c3 Nc5 10 Bc2 Bg4
11 b4 Ne6 12 Qd3 (White breaks the pin, while menacing Qb5+ and Qxb7)
12 ... Qd7 13 0-0 g6 (blunting White’s battery, while seeking exchanges via
... Bf5) 14 Bh6 Bf5 15 Qe2 Bxc2 16 Qxc2 Bf8, V.Ivanchuk-
R.Kasimdzhanov, Elista 1998 White continues to apply pressure to Black’s
game with 17 Be3.
b) The pinning 6 ... Bg4 can be met with7 h3 Bh5 8 Be2 Be7 9 0-0 0-0 10
Nfd4 Bg6, R.Zelcic-T.Martinus, Croatia 2014. White looks a shade more
comfortable after 11 f4 Be4 12 Be3.
7 Nbd4!
White moves his knight before Black restricts access to the d4-square with
a coming ... c7-c5. Other options are not as threatening:
a) The milquetoast 7 c3 c5 allows Black equality, A.Goloshchapov-
G.Fuchs, Austria 2001.
b) Also, Black has no issues after 7 Be3 c5 8 c4!? d4 9 Bd2, I.Kurnosov-
D.Andreikin, Moscow 2006. As suggested by Mikhalevski, Black emerges
well from the opening after 9 ... Nc6 (instead of Andreikin’s 9 ... f5) 10 Bd3
g6!.
7 ... Nxd4 8 Nxd4 c5!?

A novelty at the time and a hoped-for improvement.

Previously played was 8 ... Be7 9 Bd3 c5 10 Nf5 and the knight, a superb
actor, puts on his most falsely submissive face for the black king, all the
while plotting his death:
a) Here, Black erred with 10 ... 0-0?. One can sometimes tamper with an
interpretation, but never with the facts themselves. In this position there was
but a single solution to Black’s problems – and therefore zero leeway for
creative interpretation. There was no time for a dress rehearsal as Black had
to get the performance right on his first try.
 Exercise (combination alert): Gelfand misses his opponent’s
intent. White to play and launch a powerful offensive:

Answer: Annihilation of defensive barrier.

Gashimov decides upon a sharp reprimand for Black’s king with 11
Nxg7!! Kxg7? (this definitely wasn’t Gelfand’s day as 11 ... c4 is forced
when 12 Nf5 Kh8 13 Nxe7 Qxe7 14 Be2 Qxe5 is better for White, but vastly
preferable for Gelfand than the game continuation) 12 Qh5 Rh8 (12 ... f5? 13
exf6+ is suicidal for Black, since it uncovers an attack on the h7-pawn) 13
Bh6+ Kg8 14 e6! (freeing up the e5-square for his queen) 14 ... fxe6 (14 ...
Bxe6?? 15 Qe5 Bf6 16 Qg3+ and White’s queen closes her eyes with the
blissful expression of a connoisseur sniffing the bouquet of an old, rare and
expensive wine) 15 Qg4+ Kf7 16 Qg7+ and White’s earlier attacking idea
comes to fruition in glorious technicolour, V.Gashimov-B.Gelfand, Lugo
2009. Yes. Even a leading expert on the Petroff can have a horrible day and
get crushed in what is considered one of the safest openings in chess!
b) Instead, Black should settle for 10 ... Bxf5 which is necessary and after
11 Bxf5 Nc6 12 e6! g6! 13 Bh3! Qd6 14 exf7+ Kxf7 15 0-0 Rhe8, White’s
control over the central light squares is mostly offset by Black’s superior
development.
 9 Bb5+!
This disruptive move retains the edge.
9 ... Nd7!
The careless 9 ... Bd7? walks into 10 e6! Bxb5 11 exf7+ Kxf7 12 Nxb5
Qe7+ (12 ... Qe8+? is met with 13 Be3! when Black is unable to play 13 ...
Qxb5?? (13 ... d4?? 14 Qf3+ Kg8 15 Nc7 wins for White) 14 Qxd5+ Ke8 15
0-0-0 Qc6 16 Qf5 and Rhe1 is coming, with a winning attack for White) 13
Be3 d4 14 0-0! with a strong White initiative.
10 e6!
Principle: Create confrontation when leading in development.
10 ... cxd4 11 exd7+
Conversely, 11 exf7+? works out badly for White after 11 ... Kxf7 12
Qh5+ g6 13 Qxd5+ Kg7 14 Qxd4+ Qf6 when White lacks full compensation
for the material investment.
11 ... Bxd7 12 Bxd7+ Qxd7 13 Qxd4

Exercise (planning): At first glance it appears as if Black is in deep

trouble. After all, he can’t castle queenside and also any bishop move
allows Qxg7. How did Wang Yue deal with his developmental issues?

Answer: Pawn sacrifice. Black gets full compensation by offering the g7-pawn for a development
lead and open lines.
13 ... Be7!
Desperation is the last and highest court of appeal on the chess board. If
we lose the case, there are no more stays of execution. One may question the
proportionality of Black’s response, but in this case the comps say it is fully
justified. Moreover, 13 ... Rc8?! 14 0-0! f6 15 Re1+ Kf7 16 Qd3 is
unpleasant for Black.
14 Qxg7!?
White indulges in a risky pawn grab. It’s a difficult balancing act to avoid
erring on either the side of meekness, or its opposite, overconfidence. After
the safer 14 0-0 0-0 15 Rd1 Rfd8, Black is fine despite the isolani, since his
pieces all have freedom of movement.
14 ... Qe6+

Exercise (critical decision): White is at a crossroads: . He can try 15 Be3, offering

a piece sacrifice after 15 ... Bf6 16 Qg3 d4. Or he can take a calmer route
with
15 Kd1. One line is clearly superior to the other. Which one would you
play?

Answer: In this instance, the calmer route is also the better one.
 15 Kd1!
“I never guess. It is a shocking habit—destructive to the logical faculty”,
declared Sherlock Homes in The Sign of the Four. If we eliminate all
superfluous factors, then the remainder line must be the correct one. The king
eases his way to the door, desperate not to call attention to himself as he
clears the e1-square for the rook. White correctly avoids 15 Be3? Bf6 16 Qg3
d4 17 0-0-0 dxe3 18 Rhe1 e2! 19 Rd6 Qf5 and if 20 Rxe2+ Kf8 with
insufficient compensation for the piece.
15 ... Bf6 16 Qg3
After the robbery, the thief melts away into the back alleys of the city.
16 ... 0-0-0
Black’s idea produces the following corollaries:
1. White’s king is insecure in the centre.
2. This fact means it will be hard to connect rooks and complete
development.
3. Black enjoys numerous open lines for his rooks.
4. Black leads in development.
17 Re1 Qb6!
Keeping the tension by taking aim on the b2- and f2-pawns. Houdini prefers 17 ... Qc6 after which
White should keep the edge with non-stereotypical development via 18 a4! Rhg8 19 Qf3 Kb8 20 Ra3
Rg6 21 Rd3.
18 Qe3
Instead, 18 Qb3 Qxf2 is met with 19 Qh3+! Rd7 20 Be3 Qh4 (the only
retreat) 21 Qxh4 Bxh4 22 Bd4 Bxe1 23 Bxh8 and White has all the chances
in the ending, since he has two pawn islands to Black’s four. Still, White is a
million miles away from winning after 23 ... Re7 24 Bd4 Ba5 25 c3 Bc7 26
h3 Re6 27 Kd2 f5 and I think Black holds the ending without too much
trouble.
18 ... Bd4!
 Cashing in on his initial investment.
19 Qb3 Bxf2
I would have been sorely tempted to keep queens on the board with 19 ...
Qf6!? 20 Qh3+ Rd7 21 Re2 Rhd8 (21 ... Bxb2? walks into the trick 22 Rb1
Bxc1 23 Rxb7!) 22 Rb1 Bxf2 23 Qh6 when the game remains tense and
approximately even.
20 Qxb6
Whew! White’s king breathes a sigh of relief once queens are removed
from the board.
20 ... Bxb6 21 Bf4
 Mikhalevski claims an edge for White due to his superior structure. I’m
not so sure about this, since Black still leads in development and I think he
can maintain the balance without too much difficulty.
21 ... Rhe8!
Intending ... Re4.
22 Rxe8
Also, 22 Kd2 Re4 23 g3 f5 24 Rf1 Rde8 25 Rae1 Kd7 is just fine for
Black.
22 ... Rxe8 23 Kd2 Kd7
Black immediately centralizes his king, planning ... Re4 and ... Ke6.
24 Rf1 Re4! 25 Rf3 Ke6 26 Be3! Bd4!
Safer than 26 ... Bxe3+ 27 Rxe3 Kf5 when White can play on in the rook
ending after 28 Ra3.
27 Rh3
 Exercise (planning): When confronted with a critical problem, a player must avoid:
1) An overexcited mind, which runs one extraneous idea after another. 2)
A dull, unfocused lazy mind – the one which brings forth the dangerous
thought: “No need to analyse. Everything draws.” Black’s numerous pawn
islands are a cause for concern. Come up with an airtight drawing plan for
Black:

Answer: Play ... h7-h5-h4!, which enables ... Rg4, after which White is too tied down to play for the
win.
27 ... h5!
White’s chances of winning fall to zero if Black’s h-pawn reaches h4.
28 Bxd4 Rxd4+ 29 Kc3 Rc4+ 30 Kd3 h4
Intending ... Rg4. Black stands no worse.
31 Re3+ Kd6 32 Rf3 Ke6 33 Re3+
Hey, will one of you please offer the draw?
33 ... Kd6 34 Rf3 Ke6 35 Re3+ ½-½
Summary: After 5 Nbd2 we reach a dynamic position with opposite wing pawn majorities, which
has characteristics of an Open Ruy Lopez.

Game 5
D.Navara-V.Kramnik
CEZ Trophy match, Prague (rapid) 2008
1 e4 e5 2 Nf3 Nf6 3 d4 Nxe4 4 Bd3

Question: I thought we were covering 4 dxe5 lines?

Answer: Be prepared to reach our position via dual move orders, the one played in Kramnik’s
game, and also 4 ... d5. In the next game, we will examine 5 Bd3 where we intersect on move five in
both versions.
4 ... d5 5 dxe5

There we go. We transpose to our line, as if White played dxe5 on the

fourth move. In this version White skips an early Nbd2 and plays his bishop
to d3, challenging our e4-knight and enabling early castling.
5 ... Nc5
This move makes the most sense to me, when White may either lose the
bishop pair, or lose time moving his d3-bishop:
a) 5 ... Nc6 6 0-0 Bg4 (the pinning move is sharper than 6 ... Be7) 7 Nc3!
(playing to trade off Black’s best piece) 7 ... Nxc3 8 bxc3 Be7 9 Bf4 Qd7 10
Rb1, P.Svidler-A.Karpov, San Sebastian 2009. Black looks okay if he
continues 10 ... Rb8 11 h3 Bh5 12 c4 Nd4 13 cxd5 Bxf3 14 gxf3 Qxd5 15
Be4 Qc4.
b) 5 ... Be7 6 0-0 Nc5 7 Be2 0-0 8 Be3 c6 9 c4 dxc4 10 Bxc4 Qxd1 (10 ...
Nbd7 is riskier, since it is in White’s best interest to retain queens) 11 Rxd1
Nbd7 12 Rc1 Na4! 13 b3 Nab6 14 e6 fxe6 15 Bxe6+ Kh8 16 Nbd2 Ne5 17
Bxc8 Nxf3+ 18 Nxf3 Rfxc8 and Black more than equalized, since his
queenside pawn majority is a slight asset in the ending, V.Gashimov-
B.Gelfand, Bursa 2010.
6 Nc3
White’s most principled move. Other options are:
a) In the next game we look at 6 Be2 which preserves the bishop at the
cost of lost time.
b) 6 0-0 Be7 7 Be2 with a transposition to the Gashimov-Gelfand note just
above.
c) 6 Bg5 Be7 7 Bxe7 Qxe7 8 0-0 Bg4 9 Be2 c6 (9 ... 0-0 is a possible
improvement, since 10 Qxd5 Nc6 11 Nc3 Rad8 12 Qc4 Bxf3 13 Bxf3 Nxe5
is just fine for Black) 10 Re1 and White should probably play his knight to
d4, with an equal position, M.Dojcinovic-Z.Spicka, correspondence 2003. At
this point, Black can play 10 ... Bxf3! 11 Bxf3 Nbd7. I prefer Black’s
queenside majority over White’s, and also his knight pair over White’s knight
and clunky f3-bishop.
6 ... c6
Kramnik stabilizes the d5-pawn. There is no need to chop on d3 just yet,
since White gets nothing if he preserves the bishop by backing it to e2. The
careless 6 ... Be7? walks into a little 7 Nxd5! cheapo, winning an important
central pawn.
7 Nd4
 Clearing the way for a future f2-f4 push, while avoiding Black’s potential
... Bg4 pin.
7 ... Be7

Question: I understand that capture on d3

loses time, but isn’t the bishop pair worth it?

Answer: Principle: In open positions, a development lead tends to trump the advantage of the
bishop pair, so I would avoid it. Let’s look: 7 ... Nxd3+ 8 Qxd3 Bc5 9 Be3 Na6 10 0-0-0 (now we see
an added benefit for White – he can castle long and prepare a pawn avalanche against Black’s king) 10
... 0-0 11 Rhe1 Nb4 12 Qd2 Be7 13 a3 Na6 14 Nb3 (seizing control over the c5-square and at the same
time, suppressing ... c6-c5) 14 ... Be6 15 h3 Qc7 16 Bg5 Nc5 17 Nd4 a5 18 Bxe7 Qxe7, E.Sutovsky-
L.Rettore, Moscow 2008. After 19 f4, White clearly has a more potent pawn majority.
8 0-0
White has been forced to castle kingside, and now ... Nxd3 is on the table
for us.
8 ... 0-0 9 f4 Nxd3

Question: White’s g2-f4 push looks risky with his

king still on g1. Can Black exploit it with 9 ... Qb6?

Answer: Your interesting and risky suggestion hasn’t been tried, yet looks fully playable. Houdini
now suggests 10 Bf5 (also possible is the slightly safer line 10 Kh1 Nxd3 11 Qxd3 Re8 12 Be3!? Qxb2
13 a3 (threatening to snare the queen on the next move) 13 ... Qb6 14 e6 Qd8 15 exf7+ Kxf7 16 Qxh7
Rh8 17 Qd3 Kg8 18 f5 and Houdini rates this unbalanced position at dead even) 10 ... Ne6 11 Be3
(White offers the b2-pawn, egging Black to “go ahead, make my day”) 11 ... Qxb2 12 Qd3 Nc5 13 Qd2
Nbd7 14 Rab1 Qa3 15 Rf3 g6 16 Bf2! Qa5 (16 ... gxf5?? is suicide for Black after 17 Nxd5) 17 Bh3
and now f4-f5 is in the air with White having full compensation for the pawn in the form of a
dangerous build up around Black’s king.
10 Qxd3 f5!

A non-conformist’s mind is generally open to the unusual. This looks like

good strategic judgment.

Question: Why? Doesn’t it violate principles by handing White a connected,

passed e-pawn and reducing the scope of his light-squared bishop?

Answer: Exceptions don’t disprove the general rule, but they certainly are annoying! What you
claim is true, but what you left out is that Black also gets something more important in return: the ... f5-
push/blockade removes the energy of White’s kingside pawn majority, and that in turn makes the
Black’s king safer.
Black can also try 10 ... f6 11 Bd2 Na6 12 a3 (covering against ... Nb4 and
... c6-c5 tricks) 12 ... Nc5 13 Qg3 Qb6 14 Kh1 a5 (14 ... Qxb2?? 15 Rfb1 Ne4
16 Qe1 will cost Black his queen), G.Berlinger-A.Dikmen, correspondence
2009. The position remains sharp after 15 Rad1 Kh8 (15 ... Qxb2!? 16 exf6
Rxf6 17 Rb1 Qxa3 18 Ra1 Ne4 19 Nxe4 Qxg3 20 Nxf6+ Bxf6 21 hxg3 Bxd4
22 Rxa5 Rxa5 23 Bxa5 and Black looks okay with his bishop pair plus one
pawn for the exchange in the ending) 16 b3 fxe5 17 fxe5 Bd7 18 Be3 Qc7
and I slightly prefer Black’s position. This is due to the bishop pair and
having the more solid pawn structure.
11 Nb3 Na6
Black intends to transfer the knight to e6, where it effectively nullifies
White’s kingside majority with a light-squared blockade. After 11 ... b6 12
Rd1 Na6 13 Be3 Nc5 14 Nxc5 bxc5 15 Na4 Qa5 16 Qa3 d4 17 Bf2 Ba6 18
b3 (threat: Bxd4! followed by Qxe7) 18 ... Rae8 19 Qc1 Kh8 20 c4!, I
slightly prefer White’s position since her knight will transfer to the d3-
square, with an ideal blockade. If queens come off the board, then the
weakness of the c5-pawn may become a cause for concern for Black,
Y.Dembo-T.Mamedjarova, Chalkida 2009.
12 Be3 Nc7 13 Ne2
White anticipates the coming ... b6 and ... Ba6 and transfers his knight to a
more useful post.
13 ... b6
Kramnik begins the process of slowly activating his queenside majority.
14 Nbd4 Ba6 15 Qd2
Now Nxd6 is a threat. Instead, 15 Nxc6?? Bxd3 16 Nxd8 Bxe2 willl cost
White material.
15 ... Qe8 16 c3
In order to contain Black’s strategic threat of ... c6-c5, followed by ... d5-
d4.
16 ... c5
Black’s majority rolls and I already slightly prefer his position.
17 Nf3 Rd8 18 Rfd1
The rook escapes the a6-bishop’s pin.
18 ... Ne6
 Notice how White’s kingside majority is rendered useless. “White does not
have any active plan of his own, so all he can do is prevent the development
of Black’s initiative,” writes Sakaev.
19 a4
In order to keep Black’s majority in check and also to keep him guessing
about a4-a5.
19 ... h6
Keeping White guessing about a potential ... g7-g5.
20 a5?!
I would play 20 b3 and be satisfied with a slightly inferior game for White.
 Exercise (combination alert): White’s tempting and seemingly active
last move weakened the already tender light squares. This allowed
Kramnik a mini-combination which won a pawn. Do you see it?

Answer: Double attack. The e2-knight and a5-pawn hang simultaneously.

20 ... Qb5! 21 Ng3
21 axb6? fails to provide White compensation for the piece after 21 ...
Qxe2 22 bxa7 Qxd2 23 Bxd2 Be2 24 Rdb1 Bd3 25 Rd1 Ra8. Black should be
able to convert the full point.
21 ... bxa5
Not only did Black win a pawn, but he also opened the b-file for future
pressure to the b2-pawn. However, even stronger was 21 ... d4! 22 cxd4 cxd4
23 Bf2 Bb7 (Black’s grip on the light squares increases) 24 axb6 axb6 25
Ra7 Bxf3 26 gxf3 Bc5 with strategic domination.
22 Qc2
Attacking the f5-pawn.
22 ... g6 23 Qa4!?
Navara adapts to the unfamiliar rhythm of the altered situation as best he
can, hoping to regain material by picking on Black’s weak a-pawn. 23 h4 is
simply met with 23 ... h5.
 23 ... Qxa4 24 Rxa4 d4!
Principle: Passed pawns should be pushed. Black gets away with it due to
tactics along the g1-a7 diagonal and also the d-file pin.
25 cxd4 cxd4 26 Nxd4
If White decides not to take chances and avoid the d-file pin with 26 Bd2,
Black can continue 26…Bb5 27 Rxa5 a6 and his bishop pair rules, along with
his passed d-pawn.

Exercise (critical decision): Black has two tempting candidate moves:

a) 26 ... Bb4, cutting off the a4-rook’s coverage of the d4-knight and
consolidating his a5-pawn. b) 26 ... Bc5, adding pressure to the pinned
d4-knight. One line is clearly stronger than the other. Which would you
choose?

26 ... Bc5?!
This is like buying a $25,000 engagement diamond for your fiancée, and
then insisting that the jeweller place it in a $25 brass ring setting. Black wins
an exchange in this line but gives away too many pawns to do so. In this case,
initiative is more important than material.
Answer:
The correct move is a) 26 ... Bb4! as the quest for initiative rises above
contending motivations, like material. This move places White’s position
under tremendous strain after 27 Kf2 Nxd4 28 Rxd4 Rxd4 29 Bxd4 Rc8!
(threatening infiltration to the c2-square, thus forcing White’s next move) 30
Bc3 Bb5 31 Ra1 Rc4 32 Ne2 Bc5+ 33 Kg3 Re4 34 Ng1 Be3 and White has a
decisive advantage. If 35 Nh3 g5 36 fxg5 Bc6! (threatening mate on g4) 37
Nf2 f4+ 38 Kg4 Bxf2, White is down a piece and busted.
27 Rxa5! Nxd4 28 Rxa6

28…Nf3+!
Kramnik’s point. This discovery wins an exchange but fortunately for
White, it comes at a high cost (pawn-wise) to Black. White should still hold
the game.
29 Kf2 Bxe3+ 30 Kxf3 Rxd1 31 Kxe3
White has one pawn for the exchange and is about to win a second pawn.
31 ... Rfd8?!
Kramnik continues to spend at a luxury level which led to Babylon’s fall.
This allows a grave weakening of his kingside pawns. Black should try 31 ...
Kg7 32 Rxa7+ Rf7:
a) 33 Rxf7+!? when White’s best drawing chances is to play the exchange
down four versus three pawns ending after 33 ... Kxf7 34 Ne2 Rb1 35 h3 h5
36 g3 Rxb2 37 Ng1 Rb3+ 38 Kf2 Ke6 39 Nf3 Rb2+ 40 Ke3 Rg2 41 Ng5+
Ke7 42 Kf3 Ra2 and this won’t be so easy for Black to win.
 b) 33 Ra8 Rb7 34 Ra3 Rb1 35 Ne2 R1xb2 36 h4 R7b3+ 37 Rxb3 Rxb3+
leads to pretty much the same ending.
32 Rxg6+ Kh7 33 Ra6 R8d3+ 34 Kf2 R1d2+ 35 Ne2 Rd7 36 e6!
Active defence. White risks his e-pawn to put Black’s rooks in a more
defensive position. Following 36 b4 Rb2 37 Ke3 Rb3+ 38 Kf2 Rxb4 39 Kf3
Rb2 40 Ke3, I think White should hold the ending.
36 ... Rb7 37 Ke3 Rdxb2 38 Nd4 R2b6
Also, 38 ... Kg6 39 e7+ Kf7 40 Nxf5 R7b3+ 41 Ke4 Rb4+ 42 Kf3 R4b3+
43 Ke4 Rb4+ is perpetual check.
39 Rxb6!?
I would not have swapped away the remaining rook, since it makes it
tough for White to deal with Black’s passed a-pawn. I think White is at least
equal after 39 Ra5 Kg6 40 Re5 (threat: e6-e7) 40 ... Re7 41 h4 h5 as the
position remains in dynamic equilibrium.
39 ... Rxb6

Exercise (combination alert): White to play and pick off a pawn:

Answer: Deflection. The promotion threat diverts the rook and costs Black the f5-pawn.
40 e7!
This trick picks off Black’s f5-pawn.
 40 ... Rb8 41 Nxf5 a5

Exercise (planning/critical decision): Ah, yes. That small matter of the outside passed a-pawn.
White should still hold the game with accurate play. White can play: a) 42 Nd4, intending Nc6. b) 42
Kd4, intending to transfer his knight to c2 and his king to e6. c) 42 Nd6, intending to promote the e-
pawn. One of the lines draws, one loses, while the other puts White in jeopardy. Which one should he
play?

42 Nd4?
The clunky knight is also culpable in White’s steady decline. However,
this is the wrong plan.
Answer:
White draws with b) 42 Kd4! a4 43 Kd5 a3 44 Nd4 a2 45 Nc2 Kg6 46 Ke6
Rb6+ 47 Kd7 Rb7+ 48 Ke6.
He should most definitely not play the grossly overoptimistic 42 Nd6??
which loses to 42 ... a4 and the intended 43 e8Q? (43 Nc4 Re8 44 Kd3 Rxe7
is also an easy win for Black) 43 ... Rxe8+ 44 Nxe8 a3 allows Black to
promote first.
42 ... a4 43 Nc6
Threat: Nxb8, as well as Nd8.
43 ... Re8?!
This inaccuracy endangers the win. Black gains a future tempo with 43 ...
Rg8! 44 Kd4 Kg7 45 Kc4 Kf7 46 Kb4 (46 g3?? walks into a deadly pin with
46 ... Rc8! 47 Kb5 Rxc6! 48 Kxc6 Kxe7! and the a-pawn promotes) 46 ...
Rxg2 47 Kxa4 Rxh2 48 Kb5 h5 and the h-pawn wins.
44 Kd4
Also insufficient is 44 Kd3 Kg7 45 Kc3 Kf7 46 Kb4 Ra8 47 Ka3 Ra6 48
Nb4 Re6 49 Kxa4 Rxe7 50 Nd3 Re2 51 g4 Re3 52 Ne5+ Ke6 53 Kb5 Re4 54
Nd3 Rd4 55 Nc5+ Kd6 56 Nb7+ Ke7 57 g5 h5 and Black wins, since White’s
kingside pawns fall.

44 ... Kg7?!
A trifle too cautious. Black does better to distract the White king with 44 ... a3! as 45 Kc3 Ra8 46
Nb4 Kg7 47 g4 Kf7 48 Kb3 Kxe7 49 Nd5+ Kf7 50 Nc3 Rc8! 51 h3 Rc6 52 Na2 Re6 53 Kxa3 Re3+ 54
Kb4 Rxh3 55 Kc5 Rg3 wins for Black.
45 Kc4 Kf7 46 Kb4 Ra8 47 Ka3 Ra6 48 Nb4 Rd6
The more accurate 48 ... Re6! allows entry to the third rank.
49 Kxa4
Also insufficient is 49 g4 Kxe7 50 h4 Rd1 51 h5 Ra1+ 52 Kb2 Rh1 53
Ka3 Kd6 54 g5 Kc5! (intending ... Kb5! hanging on to the a-pawn) 55 g6 (55
Kxa4 Ra1+ 56 Kb3 Rb1+ 57 Kc3 Rxb4 wins) 55 ... Rh3+ 56 Kxa4 Kd6! 57
Kb5 Rxh5+ 58 Kc4 Rh4 59 Nd3 Rg4 60 Kd4 Rxg6 and the passed h-pawn
wins.
49 ... Kxe7 50 Kb5 Rd2 51 g4 Rxh2 52 Kc5 Ke6
 Exercise (planning/critical decision): White is close to achieving the draw. Up until now, defence
of his kingside pawns represents a seemly impossible barrier. He
can play: a) 53 Nd3, bringing the marginalized knight from the boondocks
back
into play, while securing the f4-pawn. b) 53 Kd4, inching his king closer
to his pawns. One line should draw, while the other loses. What is your
decision?

53 Kd4??
In one move White turns a thing of great value into worthless dross. One
look at White’s king and we think: Candidate for high blood pressure
medication. When we blunder, that terrible moment of recognition of our
action emits the same feeling as when our computer screen jars us with that
ever so hatefully blinking error sign.
Answer:
Navara missed a) 53 Nd3! Rg2 54 Ne5 Rf2 55 Nd3 Rf3 56 Kd4 Rg3 57
Ne5 when White secures both kingside pawns and holds the draw.
53 ... Rh4!
Game over. This move crushes any sense of relief White’s king may have
been feeling from the lull of the last few moves. The g4-pawn falls, giving
Black a passed h-pawn. “I hate to say I told you so”, lectures the rook, whose
favourite words in the English language just happen to be: “I told you so”.
54 f5+
54 Ke4 Rxg4 and the h-pawn wins.
54 ... Kf7! 55 Ke5 Rxg4
The thuggish rook’s blows beat down upon the g4-pawn with that awful
staccato thudding sound of heavy luggage being taken up a flight of stairs.
56 Nd5 Ra4! 57 Nc3 Rb4 58 Ne4 h5 59 Ng5+ Kg8 60 f6 h4 61 Kf5

Exercise (combination alert): Black’s win is a

mathematical certainty, provided we find the combination:

Answer: Simplification. Black returns the exchange to promote his h-pawn.

61 ... Rb5+! 62 Kg6 Rxg5+ 0-1
It’s all over. “From now on, I will be doing the thinking for both of us,”
says Black’s rook, grinning as widely as a clown, as he catches a glimpse of
his now helpless g6-cousin as after 63 Kxg5, Black’s passer runs away with
63 ... h3.
Summary: In such Open Ruy Lopez-like kingside/queenside majority situations, remember the trick
... f5! which effectively establishes a light-square blockade on White’s kingside majority from the light-
squares, at the cost of handing the opponent a connected, passed e-pawn.

Game 6
E.Inarkiev-R.Kasimdzhanov
FIDE Grand Prix, Jermuk 2009

1 e4 e5 2 Nf3 Nf6 3 d4 Nxe4 4 Bd3 d5 5 dxe5 Nc5 6 Be2

White preserves his admittedly important light-squared bishop at the very

high price of a lost tempo. I feel that this concession renders this line too
milquetoast for White to extract an edge.
6 ... Be7
This move cuts out annoying Bg5 possibilities while preparing to castle.
7 0-0 0-0 8 Be3
Inferior to the text move is 8 Nc3 as it blocks future c2-c4 ideas, which
will help chip away at Black’s centre. After 8 ... c6 9 Nd4 Qc7! (Black
provokes the potentially weakening f2-f4 before undermining with ... f6) 10
f4 (after 10 Bf4 Nbd7 11 Nf3, J.Polgar-A.Giri, Hoogeveen 2009, Black looks
slightly better with 11 ... Ne6! 12 Bg3 Ndc5 since his majority is more
mobile than White’s) 10 ... f6 11 exf6 Bxf6 12 Be3 Ne6, Black has at least
equality, V.Popov-L.Ovcharenko, correspondence 2010.
 8 ... c6
Black stabilizes his centre.

Question: What other plan is there for Black?

Answer: Black can also go for a piece play plan to put pressure on the e5-pawn with 8 ... Nc6 9 Nc3
Be6 10 Bb5 Nd7 (handing over the d5-pawn in exchange for the e5-one) 11 Nxd5 Ndxe5 12 Nxe7+
Qxe7 13 Nxe5 Nxe5 14 Qd4 (White picks up the bishop pair, but the symmetrical pawn structure offers
Black very reasonable chances to obtain full equality):
a) 14 ... Rfd8 15 Qxe5 Rd5 16 Qe4 Rxb5 17 b3 c6 18 c4 Rh5, S.Karjakin-
B.Gelfand, Jermuk 2009. Black’s h5-rook is actually out of play and
represents a slight liability. White stands a shade better after 19 Rad1.
b) 14 ... Ng4! may be an improvement as Black holds his own after 15 Qc5
Qf6 16 Bd4 Qf4 (threatening mate) 17 g3 Qh6 18 h4 Rad8 19 Bc3 (the
greedy 19 Qxa7! is met with 19 ... g5! and if 20 Be3?!, 20…Qh5! 21 Bxg5?
Rd5! sets up dual threats on both b5- and g5-bishops) 19 ... Rd5! 20 Qb4 a5
21 Qa4 Ne5 22 Be2 Bd7 23 Qe4 Bc6! 24 Bxe5 Qe6 25 f4 Qh3 26 Qg2
Qxg2+ 27 Kxg2 Rxe5+ 28 Kf2 Re7 with a likely draw.
9 Nbd2
The knight belongs here rather than the c3-square. As previously
mentioned, with Nc3, White’s knight blocks c2-c4 ideas, giving up the option
of eroding Black’s d5-point. Also, 9 Nc3 leaves the knight with nothing to do
and no place to go. On the d2-square, the knight can be transferred to b3,
where it keeps watch over the c5- and d4-squares.
White can also play the immediate 9 c4 dxc4 10 Bxc4 Nbd7 (I like this
move, since it ties White’s f3-knight to defence of the e5-pawn):
a) 11 b4!? is deemed “White’s most principled option” by Sakaev. After
11 ... Ne6 12 Qb3 Nb6 (12 ... a5!? is an unplayed possibility and if 13 Bxe6?!
fxe6 14 Qxe6+ Kh8 15 b5, Black has the shot 15 ... Nxe5! 16 Qxe5 Bf6 17
Qh5 Bxa1 18 Ng5 Bf5 19 Nf7+ Rxf7 20 Qxf7 Bg6! 21 Qxb7 cxb5 22 Nd2
(22 Qxb5?? loses to 22 ... Bd3) 22 ... Rb8 when Black has all the chances,
due to the bishop pair and queenside pawn majority) 13 Rd1 Qc7 14 Nbd2
Rd8 Black stands no worse, A.Savoca-N.Limanskis, correspondence 2010.
b) 11 Nbd2 Nb6 12 Bxc5 (12 Be2 Bf5 is pleasant for Black, who is about
to drop in on the d3-square) 12 ... Bxc5 13 Bb3 Bf5 and Black had nothing to
fear, H.Wunderlich-M.Semrl, correspondence 2009)
9 ... Ne6!

Threatening ... d5-d4 and clearing the pathway for ... c7-c5. Black has
likely attained at least equality.

Question: But doesn’t the e6-knight clog Black’s queenside

development – especially Black’s light-squared bishop?
 Answer: Not at all. Just as in Kramnik’s game earlier in the chapter, Black has the option to
fianchetto the light-squared bishop.
10 Nb3
Covering against the ... d4 threat and fighting for control of the d4-square.
10 ... c5
The logical adjunct to his last move. Black activates his queenside pawn
majority. However, he can also play the passive yet very playable 10 ... Nd7
which allows White his intent with 11 c4 dxc4 12 Bxc4 Nb6 13 Bd3 Nd5,
A.Bokros-G.Kovacs, Hungary 2013. Black’s position looks active after 14
Bd2 a5.
11 Re1
A new move in the position. White provokes ... d5-d4. I sometimes deviate
from theory, simply to induce a kind of bureaucratic spite in a booked-up
opponent, rather than an objective search for an improvement in the position.
I get the feeling that Inarkiev does the same here. His move fails to bother
Black. Let’s have a look at the alternatives:
a) With 11 c4!?, White begs for ... d5-d4, hoping to utilize d3 as a
blockade square. After 11 ... d4 12 Bc1 f5! (Black mirrors White’s pawn
structural strategy) 13 exf6 Bxf6 14 Bd3 Nc6 15 Re1, M.Palac-V.Kunin,
Schwarzach 2010, Black achieves an excellent position after 15 ... Re8,
intending to unravel with ... Nf8.
b) The restraining 11 c3 (clamping down on …d5-d4) can be countered
with 11 ... b6 12 Re1 Nc6 13 Qc2 (clearing the path for Rad1) 13 ... Qc7 14
Rad1 Rd8 15 Bf1 a6 (Black begins to roll forward his queenside pawn
majority) 16 h3 (A sign that White has nothing useful to do) 16 ... b5 17 a3
Rb8 and Black emerges with superior central control and a potential
queenside initiative, G.Berlinger-G.Mueller, correpondence 2010.
11 ... b6 12 c4!?
White seeks to provoke ... d5-d4, reaching a position similar to note a)
from just above.
12 ... Bb7!
Kasimdzhanov correctly prefers central piece play to 12 ... d4 13 Bd2 Bb7
14 Bd3 h6 15 Nc1 Nd7 16 Qe2 Qc7 17 Be4! when White swaps off Black’s
best piece as well as clears the d3-square for his knight, with about even
chances.
13 Bd3
 This move costs White a tempo, but I don’t have any great suggestions
here for a big improvement. Houdini suggests 13 cxd5 Bxd5 14 Qb1 Qd7 15
Bd3 g6 16 Rd1 Qb7 but Black already looks better.
13 ... dxc4!
Rendering White’s last move a total waste of time.
14 Bxc4 Nc6

From the following inferences, we come to the conclusion that Black

stands better:
1. We note a clashing conjunction of opposite visions. Black’s queenside
majority looks more mobile than White’s kingside version, since Black can
play for ... a7-a6 and ... b7-b5, while White is a million miles away from
achieving f2-f4-f5.
2. Black controls the key d4-square.
3. Black has a grip on the central light squares.
15 Qb1
Inarkiev hopes to transfer the queen to the f5-square to generate kingside
play and to help attack the e6-knight. However, Black’s central and
queenside counterplay is up to the challenge.
15 ... Qd7!?
Black provokes Rd1 which seemingly allows White to generate play down
the d-file. I prefer the simpler 15 ... Qc7! 16 Qf5 Rad8! (this pawn sacrifice is
more of a loan than a gift as White cannot hold on to the extra pawn
meaningfully) 17 Bxe6 fxe6 18 Qxe6+ when Black gets excellent
compensation after 18 ... Kh8 19 Bg5 Rde8 20 Bxe7 Rxe7 21 Qh3 Nxe5. The
wonderful thing about poverty is there is no place for your economic status to
go but up. Black regains his previously sacrificed pawn with an edge after 22
Nxe5 Rxe5 23 Qg3 Rxe1+ 24 Rxe1 Qxg3 25 hxg3 Rd8 and if 26 Re7 Rd1+
27 Kh2 Bd5! 28 Re8+ (28 Rxa7? Kg8! (White has no good defence to ... c4,
trapping the knight) 29 Ra6 c4 30 Rxb6 cxb3 31 axb3 Rb1 and White will
lose the ending) 28 ... Bg8 29 Re2 h6 and Black’s bishop is clearly better
than White’s remaining knight, and his majority is superior as well.
16 Qf5
White attackers begin to congregate in the vicinity of Black’s king,
dropping disturbing hints as to their intent.

Question: Why didn’t White gain a tempo with 16 Rd1 before playing Qf5?

Answer: Black still holds the edge after 16 ... Qc8 17 Qf5 Ng5! 18 Qxc8 (18 Nh4?! allows Black to
inflict even deeper structural damage with 18 ... Nh3+! 19 Qxh3 Qxh3 20 gxh3 Bxh4) 18 ... Nxf3+ 19
gxf3 Raxc8 20 Rd7 Ba8 21 f4 Rfd8 22 Rad1 Kf8 and Black’s chances look slightly better due to his
superior structure.
16 ... Rad8 17 a3
Eliminating ... Nb4 tricks.
 17 ... Qc8
Anticipating White’s coming Rd1.
18 Rad1?!
Anticipating Black’s coming combination with 18 Kh1 fails to fully
equalize after 18 ... Nc7 19 Qxc8 Bxc8 20 Kg1 h6 21 Rac1 Be6. Still,
White’s position is preferable to the game continuation.

Exercise (combination alert): White’s natural last move walks

into a little combination which inflicts structural damage. How?

Answer: Discovered attack/zwischenzug.

18 ... Ng5!
White’s would-be kingside initiative evaporates as we speak. Black’s
devious intent is suddenly unmasked, like the villain at the conclusion of
every Scooby Doo episode.
19 Qxc8
White’s queen mirrors her sister’s whims, in that both seek to destroy the
other. “The time I deign to take orders from the likes of you is never – and
not even then!” spits White’s queen to her sister who is spotting a sinister
smile.
19 ... Nxf3+
 Check! Black damages White’s kingside pawns with this neatly-inserted
zwischenzug.
20 gxf3 Bxc8 21 f4 Be6!

Eliminating White’s best piece and insuring that White’s remaining bishop
will be a bad one, with pawns fixed on its colour.
22 Rxd8 Rxd8 23 Bxe6
Not 23 Nd2? as it hangs a pawn to 23 ... Rxd2! (undermining/removal of
the guard) 24 Bxe6 Rxb2 25 Bd5 Nd4 26 Rd1 g6! and Black is up a pawn,
with a winning ending despite White possessing the bishop pair.
23 ... fxe6
Black is virtually up a pawn, since White’s kingside majority is held back by Black’s three pawns. Still,
it won’t be easy to break through White’s attempt at setting up a fortress.
24 Rc1 h6 25 Nd2
For now, the e4-square looks like the best placement for White’s knight.
25 ... Nd4
Heading for a beautiful perch on the f5-square.
26 Kg2 Nf5 27 Kf3 g5!
Restless flutterings take place structurally on the kingside. Black is willing
to undouble White’s pawns to weaken the chap on e5.
28 Ne4
After 28 fxg5? Nxe3 29 Kxe3 Bxg5+ 30 f4 Bxf4+! (White’s king is
somewhat taken aback when the elderly black bishop places dignity aside and
unexpectedly begins to dance the Charleston) 31 Kxf4 Rxd2, White will be
unable to save the ending following 32 b4 cxb4 33 axb4 Rd4+ 34 Ke3 Rxb4
35 Rc8+ Kf7 36 Rc7+ Kg6 37 Rxa7 b5 38 Kd3 Kf5 39 Rh7 Rh4 40 Rb7 b4
41 Rb5 Rh3+ 42 Kc4 Rxh2 and if White captures the b4-pawn with either
piece, he lands in a resignable king and pawn ending.
28 ... gxf4

Exercise (critical decision): Which way should White recapture the f4-pawn?

29 Bxf4?!
White either misses or ignores the position’s sub-context. The natural
recapture allows Black to make serious inroads.
Answer: Correct was the unconventional 29 Kxf4! which limits Black’s edge and prevents his
coming idea.
29 ... Rd3+!
Black’s rook infiltrates down the third rank. Note that had White played 29
Kxf4 instead, 29…Rd3+ can be countered with 30 Rc3! Rd1 31 b4! and if
Black accepts the b-pawn offer, White can cause enough problems with his
rook down the c-file to maintain equal chances.
30 Ke2 Rb3 31 Rb1
 What choice do we have when alternatives are worse? There is no point
trying:
a) 31 Rg1+ Kf7 32 Bc1 b5 when White’s position is completely passive.
b) 31 Rc2?? and White walks into a fork after 31 ... Nd4+.
31 ... c4
Setting up the ... Bxa3 threat.
32 Nd2 Rd3! 33 Be3?!
Possibly an overreaction:
a) Worse is 33 Nxc4?? which drops a piece to 33 ... Rd4.
b) Inarkiev probably didn’t like 33 Rd1 Nd4+ 34 Ke1 c3 35 bxc3 Rxc3 36
Bxh6 Rxa3 when Black gets two connected queenside passed pawns. Still,
this line looks better than the game continuation.
33 ... Nxe3 34 fxe3 b5
Now that the c4-pawn is secure, Black can direct his attention to the weak-
e3-pawn.
35 b3?
He had to try 35 Rg1+ Kf7 36 Rg3 and put up stubborn resistance.

Exercise (combination alert): White’s last move, an attempt to

undermine the c4-pawn is mistimed. How did Black win material?
 Answer: Overloaded defender. White is unable to cover his e3-pawn with Nf1, since the b3-pawn
will hang instead.
35 ... Bg5!
We sense unwholesomely purposeful menace in the bishop’s approach.
The cleric is the direct opposite of Raskolnikov (from Dostoyevsky’s Crime
and Punishment) in that he moves from pious to murderous as the story goes
on. Two against one constitutes an unfair fight and the pincer action wins at
least one pawn.
36 Rg1 Rxe3+ 37 Kf2

Exercise (combination alert): White threatens both h2-h4 and Rxg5+. Black can cover both threats
with 37 ... Rh3, but there is something so much stronger. What is it?

37 ... Rh3
This should still win.
Answer: Far more effective is 37 ... c3!! when White’s dismal options offer little scope for realistic
hopes for survival:
a) 38 Rxg5+ hxg5 39 Kxe3 c2! and White is helpless to halt the c-pawn’s
promotion.
b) 38 h4 cxd2 39 hxg5 Re1! (forcing simplification as Black wins the king
and pawn ending) 40 Rxe1 dxe1R! (I always underpromote if I can get away
with it!) 41 Kxe1 hxg5 42 Ke2 Kg7 43 Kf3 Kg6 44 Kg4 a5 45 b4 a4 leaves
White in zugzwang.
38 Nf3 c3! 39 Nxg5
No choice, since 39 Kg2?? Rxf3! 40 Kxf3 c2 costs White a piece.
39 ... hxg5 40 Rxg5+ Kf7 41 Ke2
Hopeless for White is 41 Rg2 c2 42 Rg1 Rxb3 43 Rc1 Rh3 44 Kg2 Rxa3
45 Rxc2 b4.
41 ... c2!
Clearance. Black lawn-mows White’s queenside pawns.
42 Kd2 Rxb3 43 Kxc2
Also futile is 43 Rg3?? c1Q+ 44 Kxc1 Rxg3 45 hxg3 Kg6 46 Kd2 Kf5 and
Black wins the king and pawn ending.
43 ... Rxa3 44 h4 a5 45 Rg4

Exercise (planning): We all know what Black wants: to end

White’s resistance. But how to bridge the needy gap from
here to there? What is Black’s most efficient winning plan?

Answer: Simplification. White has to decide between a lost king and pawn ending or coughing up
yet another pawn.
45 ... Ra4!
A logical extension to what is already a given: Black dominates the
proceedings no matter what option Black picks.
46 Rxa4
46 Rg1 is the equivalent of resignation after 46 ... Rxh4.
46 ... bxa4 47 Kb2 Kg6 48 Ka3 Kh5 49 Kxa4 Kxh4 50 Kxa5 Kg5 51
Kb4 Kf4

The black king establishes himself as the standard bearer of all which is
good in the world, as he strikes a heroic pose, with one arm curving
heavenward (the same pose as Saddam Hussein’s now fallen statue), to show
others the path. “There is no use in appealing to my ordinary human
compassion and decency, since as you can plainly see, I am extraordinary in
every way,” Black’s king tells his brother, who in his entire life, has always
followed one step behind. Black wins the king and pawn ending by a single
tempo. The position would be drawn if White’s king were on c3, since it
could get in front of Black’s e-pawn and seize opposition.
52 Kc3 Kxe5 53 Kd3
The stricken look White’s king gives his brother resembles that of a
particularly emotional martyr approaching the stake.
53 ... Kf4 0-1
Even my six-year-old students know how to win this king and pawn
ending from this stage.
Summary: 6 Be2, while preserving the light-squared bishop, loses time and is too passive for
White to hope for even an edge.

Game 7
L.Bruzon-V.Ivanchuk
Carlos Torre Memorial, Merida 2007

1 e4 e5 2 Nf3 Nf6 3 d4 Nxe4 4 dxe5

Black’s idea isn’t available if White plays the move order 4 Bd3 d5 5 dxe5
and we reach the position we covered in the first two games.
4 ... Bc5!?

This is not just a fool’s errand. Black, who advances a radical theory with
a move dangerously close to the forbidden, refuses to play the thematic
freeing move ... d7-d5. Instead he immediately attempts to take over the
initiative by going after the f2-pawn. It rarely works when we wrench loose
reality and then attempt to force-fit it into our own ideas. This move may well
be an exception to the rule. “But if Bruzon had had an intention to take his
opponent aback, he failed completely as Ivanchuk responded with a move
that previously only amateurs had employed,” writes GM Maxim Notkin.

Question: Wait a minute! Why is this line so rare for

Black? I don’t see a good way to defend the f2-pawn?

Answer: The f2-pawn will indeed fall. As compensation, White gets an open f-file and a
development lead.
5 Qd5?!
The queen is amiable and good natured, until overcome by one of her fits
of homicidal madness. Lashing out sometimes represents the attempt to repair
an absence of counterplay. Opportunism overpowers prudence and in the
space of a single move, the lion’s share of White’s potential opportunities
vanish. This double attack proves ineffective and the trusty comps say Black
can get away with the crime and avoid losing a piece:
a) Contesting the bishop with 5 Be3?! in a sense gives complete legitimacy
to Black’s idea. The opening of the f-file fails to compensate for the self-
inflicted structural damage after 5 ... Bxe3 6 fxe3 d5 7 exd6 Nxd6 as Black is
about to safely castle and I see no compensation for the weak e-pawn,
S.Keller-S.Turner, correspondence 1997.
b) The critical line is 5 Bc4! where White ignores Black’s f2-threat to open
the f-file for the h1-rook.

Following 5 ... Nxf2 (we grab the pawn and ask White to demonstrate
compensation) 6 Bxf7+! (6 Qd5?! works our poorly for White after 6 ... Qe7
7 Rf1 c6! 8 Qxf7+ [8 Qd2? d5 9 Be2 Ne4 10 Qf4 0-0 and White can resign,
since he is down two pawns for absolutely no compensation, P.Kurumanov-
G.Nagibin, Moscow 2009] 8 ... Qxf7 9 Bxf7+ Kxf7 10 Rxf2 Bxf2+ 11 Kxf2
and White is down a full exchange but Black must still overcome
development hurdles. White would probably have full compensation if
queens remained on the board, but this isn’t the case) and here, Black has a
choice of safety or insanity:
b1) The safer choice is 6 ... Kf8 7 Qd5 Nxh1 8 Bh5 (perhaps White can
improve with 8 Qxc5+! and Black had better know the analysis which
follows, or else ... ! 8 ... Kxf7 9 Nc3 d6 10 Qd5+ Kf8 11 Be3 Nc6 12 0-0-0
Qe7 13 Rf1 Ke8 14 Bg5 Qf7 15 exd6 h6 16 Bd2 Be6 17 Re1 Kd7 18 Qb5
Kc8 19 dxc7 Re8 20 Rxh1 Qxc7 and White has full compensation for the
exchange, with one pawn and an out-of-play a8-rook) 8 ... Qe7 9 Bg5 Bf2+
The only move. 10 Ke2 Qe6 11 Nc3 h6 12 Bd2, E.Podvoysky-L.Ovcharenko,
correspondence 2010. Houdini continues to rate the position at approximately
even after 12 ... Nc6 13 Rxh1 Bb6 14 Rf1 Kg8 15 Bg4 Qxd5 16 Nxd5 d6 17
e6 Kh7 18 Bf5+ g6 19 Bxg6+! Kxg6 20 Nh4+ Kh5 21 Rf5+ Kxh4 22 Be1+
Kg4 23 Rf4+ Kh5 24 Rf5+ Kg6 25 Rf6+ with perpetual check. Of course,
neither we nor our opponents will ever manage to remember 25 moves of
complex analysis over the board. So in reality, the one with the better feel for
the position (us, of course!) will probably emerge victorious.
b2) Black can also go for a high-stakes gamble with 6 ... Kxf7!? 7 Qd5+
Kg6 8 Qxc5 Nxh1 and White is down a full rook. Yet the h1-knight will
eventually be won, so the question is: will White’s ferocious initiative and
development lead compensate Black’s extra exchange? The comps evaluate
this position at dead even. Following 9 Nc3 h6 10 Qc4 Kh7 11 Qe4+ g6 12
Nh4 Qe8 13 Nd5 (this looks pretty scary as f6- and c7-forks are menaced) 13
... Rf8 14 Qe3 (14 Nxc7? Nf2 15 Qd5 Qe7 16 Nf3 Nc6 17 Kxf2! Rb8 18 Kg1
Rf5! leaves White in trouble) 14 ... g5 15 Qd3+ Kg8 16 Ng6 Nf2 17 Qc4!
Qf7 18 Nxf8 Nc6! 19 Nxd7! Bxd7 20 Nf6+ Kg7 21 Qxf7+ Kxf7 22 Nxd7
Ne4 23 Be3 Rd8 24 Rd1 b6 25 a3 Ke6 26 Nf6 Rxd1+ 27 Kxd1 Nxf6 28 exf6
Kxf6, Black’s superior king position more than compensates for White’s
bishop over knight theoretical edge.
5 ... Bxf2+
White’s king gets tagged by an errant bullet. Naturally, 5 ... Nxf2?? loses
after 6 Qxc5 Nxh1 7 Bg5! (even stronger than 7 Qg1) 7 ... f6 8 exf6 gxf6 9
Bh4 Qe7+ 10 Qxe7+ Kxe7 11 Nc3 c6 12 Bd3 d5 13 0-0-0 and the h1-knight
falls and Black is busted as following 13 ... Rg8 14 Rxh1 Rxg2? 15 Bg3 Bg4
16 Nh4, Black snares more material.
6 Kd1
 White’s king awakens to a radically new view of life.

Question: How does Black save his piece if White plays 6 Ke2 instead?

Answer: The geometric anomaly 6 ... f5! saves the day after 7 exf6 (Notkin writes: “Steinitz would
have appreciated 7 Bg5 Nxg5 8 Kxf2 Nxf3 9 Kxf3!? but objectively after 9 ... Nc6, White has no
compensation for the pawn”) 7 ... Nxf6 8 Qe5+, M.Glotov-G.Nagibin, Moscow 2009. Now Black can
play 8 ... Kf7! (less precise is 8…Kf8 which still favours Black) 9 Ng5+ Kf8 and White’s problem is
the f2-bishop remains immune due to a g4-knight fork. White lacks full compensation for the exchange
and pawn, although he may still retain some practical chances.
6 ... f5!

I already prefer Black, whose extra pawn means more than White’s
development lead. 6 ... Nc5?! is inferior as White gets a good position after 7
Bg5 c6 (7 ... f6?? 8 exf6 gxf6 9 Bh4! Bxh4 10 Qxc5 Bg5 11 Nxg5 fxg5 12
Qe5+ Kf7 13 Bc4+ Kg6 14 Bd3+ Kh6 15 h4 gxh4 16 Qf4+ Kg7 17 Qg4+
Kf7 18 Rf1+ and Black must hand over his queen) 8 Qd2 (also, White is
much better after 8 Qd6 Ne4 9 Bxd8 Nxd6 10 exd6 Kxd8 and now both 11
Ng5 and Ne5 appear to give White a powerful initiative) 8 ... Ne4 9 Bxd8
Nxd2 10 Nbxd2 Kxd8 11 Ne4, it is White who has the advantage.
7 Bc4
7 exf6?! Nxf6 8 Qe5+ Qe7 only benefits Black as White has nothing to
show for his pawn deficit.
7 ... Qe7 8 Nc3
 Question: Why does White invite doubled pawns?

Answer: Bruzon recognizes that Black’s e4-knight is his key defender. Without it, White has access
to Bg5 tricks:
a) After 8 Rf1 Nc6 (8 ... c6! looks like a big improvement as 9 Qd3 b5 10
Bb3 Na6 11 Bg5 Nxg5 12 Nxg5 Qxg5 13 Rxf2 Nc5 14 Bf7+ Kd8 15 Qxf5
Qxf5 16 Rxf5 d5! 17 Rf1 Rf8 18 Nd2 Bg4+ 19 Ke1 Ke7 20 h3 Rxf7 21
Rxf7+ Kxf7 22 hxg4 Re8 is a lost ending for White) 9 a3, K.Dragun-
U.Eliseev, Moscow 2010, Black still stands better following 9 ... b5! 10 Bd3
Bb7 when Black’s pieces are more active and White’s king looks less secure
than Black’s.
8 ... c6
Denying White rights to the d5-square. Ivanchuk correctly assesses that
White will be unable to take advantage of the temporary d6-hole.

9 Qd3

Question: It feels to me like White has compensation

for the pawn. How does Black develop his queenside?

Answer: Please see Black’s next move!

9 ... b5! 10 Bb3 a5
 ... a4 is in the air, as is ... Ba6. 10 ... Bb6 11 Nxe4 fxe4 12 Qxe4 d5! is
similar to the game’s continuation.
11 a4 b4 12 Nxe4 fxe4 13 Qe2!?

This already looks like desperation. Bruzon didn’t like 13 Qxe4 d5! 14
Qe2 Bc5 15 Bg5 Qe6 16 Be3 Be7 17 Qe1 0-0 as White is strategically lost
for the following reasons:
1. His king remains insecure.
2. His light-squared bishop remains out of play. If he plays c2-c3, to re-
post the bishop on c2, it allows ... bxc3, which further exposes White’s king.
3. White’s a1-rook is out of play.
4. White’s e5-pawn is a liability.
13 ... d5!
Ivanchuk unerringly finds the way to keep the initiative. A look at the
weaker alternatives:
a) Ivanchuk correctly rejects 13 ... exf3? 14 Qxf2 when suddenly White’s
light-squared bishop returns to life and Black’s massive lag in development
puts his king in danger.
b) 13 ... Bb6 14 Ng5 Qxe5 15 Nf7 Qd4+ 16 Bd2 e3 17 Nxh8 d5 18 c3
Qxd2+ 19 Qxd2 exd2 20 Kxd2 Nd7 21 Rae1+ Kf8 22 Rhf1+ Nf6 23 Bc2
Kg8 24 Rxf6! bxc3+ 25 bxc3 gxf6 26 Re8+ Kg7 27 Bf5! Bxf5 28 Rxa8 Be4
29 Ra6 Bg1 30 Rxa5 Kxh8 31 Ra8+ and White’s passed a-pawn offers
serious counterplay.
14 exd6
Also, 14 Qxf2 exf3 15 Qxf3 Nd7 16 Re1 Nc5 17 Be3 Be6 gives Black a
strategic bind and thus the advantage.
14 ... Qxd6+ 15 Bd2
The alternatives are cheerless:
a) 15 Nd2?? e3 16 Bc4 Kd8! and Black wins a piece.
b) 15 Qd2?? Qxd2+ 16 Nxd2 Bg4+ is crushing.
15 ... e3 16 Rf1
Threat: Rxf2. White is unable to unpin with 16 Kc1?? since Black has 16
... Ba6 17 Bc4 Bxc4 18 Qxc4 exd2+ 19 Nxd2 Kd8! 20 Rd1 Be3 and Black
consolidates.
16 ... Ba6!
The text move is far more ruthless than 16 ... Kd8?! 17 Rxf2 exf2 18 Qxf2
when White’s bishops offer counterplay for the exchange.
17 Bc4 Bxc4 18 Qxc4

Exercise (planning): In my opinion, Ivanchuk is one of the most

refreshingly original strategists in the world. White plans to unravel
with Ke2 next. Can you find Ivanchuk’s deep solution to his needs?
 Answer: Prophylaxis. Black transfers his once sleeping a8-rook to e7, which adds critical
reinforcement to the choking e3-pawn, effectively short circuiting White’s Ke2 defensive idea.
18 ... Ra7!!
This argument will be settled by brinksmanship of ideas, rather than
mathematical number crunching. Creativity and passion in this case rule over
cold rationality. Ivanchuk safeguards his king with the uncanny acuity of a
natural defender/counterattacker, while at the same time, generates attacking
chances.
19 Ke2 Re7 20 Be1
The developing 20 Rad1 (intending Bc1) is met with 20 ... Qg6! and White
should go for 21 Rxf2 (even worse is 21 g3? which weakens too many light
squares and Black is winning after 21 ... Rf8) 21 ... exf2+ 22 Kxf2 Rf8 when
Black has both the extra exchange and the initiative.
20 ... Qg6! 21 g3
White decides that his king’s needs are more pressing than any other
potential project. 21 Bxf2?? is met with the crushing 21 ... Qxg2!.
21 ... Re4!
The white queen’s subtle ear detects nuances of impertinence in the rook’s
tone.
22 Qd3
Alternatives also lose:
a) 22 Qc5?? blows a tempo in a position where White can’t make a single
inaccuracy. After 22 ... Nd7 23 Qxa5 0-0 24 Bxf2 Qe6!! 25 Bg1 Qc4+ 26
Ke1 Qxc2 and mate is inevitable as Black threatens ... Rxf3 and ... Qd2
mate!.
b) 22 Qb3 Nd7 23 Bxf2 Nc5! (a zwischenzug, gaining Black a key tempo)
24 Qa2 b3! 25 cxb3 (White’s queen is buried alive) 25 ... Rf8 26 Be1 Rc4!
(threat: ... Qd3 Mate!) and White can resign.
c) 22 Qa2 Bxe1 23 Raxe1 Nd7 24 Kd1 Rf8 25 Kc1 e2 26 Rf2 Kd8! And
White is in zugzwang. There is no defence to the coming ... Qh5.
22 ... Nd7 23 Rd1
White has better practical chances with 23 Bxf2 Nc5! 24 Qd1 Qh6!! 25
Bg1 0-0 26 Rc1 Rfe8 27 b3 Qh3 (threat: ... Qg2+) 28 Qd6 Qg2+ 29 Ke1 Ne6
30 Nh4 Qd2+ 31 Qxd2 exd2+ and if 32 Kxd2 Rd8+ forces mate.
 Exercise (combination alert): How did Ivanchuk continue his attack?

Answer: Piece sacrifice/pin. Black gets his king out of the centre and activates his last ineffective
piece, the h8-rook.
23 ... 0-0!!
I had forgotten that this move was still legal!
24 Qxd7
Or 24 Qb3+ Kh8 25 Rxd7 Qf5 and the double attack on the f3-knight and
the d7-rook is curtains.
24 ... Qh5
Black’s point. White’s game collapses if the f3-knight falls.
25 g4 Qh3!
 Beautiful chess. Ivanchuk, having calculated a decisive kill, allows the
coming knight fork.
26 Ng5 Qg2!
The g2-overlord (overlady?) orders the village burned down.
27 Bxf2
Also futile is 27 Nxe4 Qf3+! 28 Kd3 e2+ 29 Kc4, leaving Black with the
pleasant choice of taking either rook and promoting to a new queen.
27 ... Rxf2+!
Alternatives are:
a) The weak 27 ... exf2+?? 28 Nxe4 Qxe4+ 29 Kd2 Qf4+ 30 Kd3 and
White wins instead.
b) The comp move 27 ... Re5!! 28 h4 Rxf2+ 29 Rxf2 exf2+ and Black is
poised to win as White has to shed plenty of material just to protect his king.
28 Rxf2
Also, 28 Kd3 Rd2+ 29 Rxd2 Qxd2+ 30 Kxe4 Qxd7 picks up the queen.
 Exercise (combination alert/critical decision): White’s king is the small
fish who swims a part of the lake which is laced with nets. He gets
caught, yet continues to slip through and escape – up until now. Black can
capture on f2 with either his queen or his e-pawn. Only one is correct:

Answer: Capturing with the pawn is correct.

28 ... exf2+! 0-1
Exclams rain down on the white king’s head like blows to a losing boxer
in the ring. Now the black rook and queen work in perfect synchronization as
after any White move, Black will promote the f2-pawn to a queen and mate
will follow soon.
Conversely, 28 ... Qxf2+?? blows it as after 29 Kd3 e2 30 Qd8+ Qf8 31
Qxf8+ Kxf8 32 Re1, it is White, not Black, who wins.
Summary: Shockingly, 4 ... Bc5!?, which almost looks like a kind of Reversed Traxler Attack,
holds up under the computer’s scrutiny (if we can survive the critical 5 Bc4! line). That, plus the fact
that it will almost certainly surprise your opponent, makes it a tempting surprise weapon. Also,
remember, this line can only be played when White plays the 4 dxe5 move order.

Game 8
V.Gashimov-V.Kramnik
Azerbaijan vs. the World match, Baku 2009

1 e4 e5 2 Nf3 Nf6 3 d4 Nxe4 4 Bd3 d5 5 Nxe5

This is White’s main line in the Scotch-Petroff.
 Question: Why is White okay with the pawn symmetry?

Answer: For the simple reason that he leads in development by a single tempo, so Black is still far
from proving equality. Remember, 5 dxe5 transposes to previous games of the chapter.
5 ... Nd7
Our best bet is to immediately challenge White’s e5-outpost.

Question: Will we be covering 5 ... Bd6?

Answer: We won’t be covering this branch in detail due to space concerns (for some odd reason,
the moody folk from Everyman tend to gnash their teeth, utter vile curses and regard your writer with
deep loathing when I exceed the space quota – which is pretty much every single book I write!).
Anyway, here’s some sample lines which well illustrates the themes in the 5…Bd6 line: 6 0-0 0-0 7 c4
Bxe5 8 dxe5 Nc6 9 cxd5 Qxd5 (the e5-pawn is in some danger but White has a good solution with 10
Qc2! Nb4 11 Bxe4 Nxc2 12 Bxd5 and here, Black has:
a) 12…Bf5 13 g4! (this deflection trick insures White an edge) 13 ... Bxg4
14 Be4 Nxa1 15 Nc3 Bh3 16 Re1 f5 17 exf6 Rae8 18 Bd2 Rxe4! (returning
the exchange by giving the a1-knight the c2-escape square) 19 Nxe4 Nc2 20
Rc1 Nd4 21 Rxc7 Rf7 (or 21 ... gxf6 22 Be3 Nf3+ 23 Kh1 Nh4 24 f3!? Rf7
25 Rc3 b6 26 Bf2 Nf5 27 Rc1 with even chances) 22 Rxf7 Nf3+ 23 Kh1
Kxf7 24 fxg7 Kxg7 25 Be3 b6, M.Tal-A.Karpov, Milan 1975. Although
down a pawn, Karpov held the draw without difficulty, since White’s king
was paralyzed and conversion was impossible without an error from Black’s
side.
b) After 12 ... Nxa1, Black can win an exchange but it is temporary since the a1-knight lacks an
escape route. With 13 Be4! Whit seizes control over the c2-square, while cutting off ... Bf5. Following
13 ... Re8 14 Na3! Rxe5 15 Bf3 Ra5 16 Bf4 Rxa3 17 bxa3 Nc2 18 Rd1 (threatening mate on d8) 18 ...
Be6 19 Bxb7, White’s bishop pair offers an edge in the ending.)
6 Nc3!?
We should be aware of this sideline.

Question: Why does White deliberately allow damage to his structure?

Answer: In exchange, White gets open lines, a space edge and some attacking chances. Sixth move
alternatives:
a) Current theory deems 6 Qe2 as no longer considered dangerous for
Black, mainly due to Larsen’s amazing coming idea. The knights get traded
off with 6 ... Nxe5 7 Bxe4 dxe4 8 Qxe4 Be6 9 Qxe5 Qd7 (Black’s bishop pair
and development lead offer full compensation for the pawn) 10 0-0 0-0-0 11
Be3, and here, the mercurial Dane sprung 11… Bb4!!.
 This idea, preventing the tempo gaining Qa5, was a game-change of a
novelty:
a1) 12 Nc3 f6 13 Qg3 Bxc3 14 bxc3 h5 15 h4 g5!, A.Karpov-B.Larsen,
Tilburg 1990.
Karpov’s position already looks quite uncomfortable. White’s damaged
queenside pawns, weak light squares and non-existent attacking chances give
Black the edge. Larsen went on to defeat the reigning world champion in
brilliant style. This game is annotated in Larsen: Move by Move. It would be
contemptible and wilful folly, akin to madness on your part, if you didn’t
order this marvellous book immediately!
 a2) White is probably better off with 12 c3 f6! (White’s queen is denied access to the a5-square) 13
Qg3 Be7 14 Nd2 h5 15 f3 g5 and Black’s bishop pair and coming kingside pawn storm offer full
compensation for the pawn, S.Tiviakov-G.Timoscenko, Ohrid 2001.
b) 6 0-0 offers White nothing. Following 6 ... Nxe5 7 dxe5 Nc5:
b1) 8 Nc3 c6 9 Ne2 Be7 10 f4 f6 11 Be3 Nxd3 12 Qxd3 0-0 13 c4 fxe5 14
fxe5 Be6 15 Nf4 Bf5 16 Qd4 Bg5 17 cxd5 Bxf4 18 Rxf4 Qxd5 19 Qxd5+
cxd5 20 Raf1 g6 21 g4 Be6 22 Rxf8+ Rxf8 23 Rxf8+, a draw was agreed
here, S.Mamedyarov-V.Kramnik, Wijk aan Zee 2008.
b2) 8 Be2 Be7 9 Be3 0-0 10 f4 (10 c4 Be6 is fine for Black, J.Ehlvest-
D.Fridman, Tallinn 2013) 10 ... f6 11 exf6 Rxf6 12 c4 d4! 13 Qxd4 Rd6! 14
Qxc5 Rd3 15 Qa5 Rxe3 and Black’s bishop pair in the open position plus
dark-square control more than compensate for White’s extra pawn,
M.Kauppinen-F.Sage, correspondence 2010.
c) 6 Nxf7 is a drawing trick used by White:
c1) 6 ... Kxf7 7 Qh5+ Ke7 (7 ... Ke6 8 Qe2 Kf7 9 Qh5+ is drawn) 8 Qe2 8
... Kf7 (not 8 ... Kd6? 9 Bf4+ Kc6 10 Bxe4 dxe4 11 Nc3 (threat: Qb5 mate!)
11 ... a6 12 Qc4+ Kb6 as now White has access to a deadly attraction/knight
fork combination with 13 Bxc7+! Qxc7 14 Nd5+ and it’s game over) 9 Qh5+
Ke7 and after White repeats with 10 Qe2, Black has nothing better than to
repeat moves with 10…Kf7 with a draw.

c2) Question: Can Black avoid a draw with 6 ... Qe7?

 Answer: IM Raetsky and FM Chetverik consider this move “risky but playable,” while I would
caution the reader to stay clear of it. White can respond with 7 Qe2! Qxf7 8 f3 Ndf6 9 Nd2 Qh5 10 fxe4
Qxe2+ 11 Kxe2 Bg4+ 12 Kf2 dxe4 13 Nxe4 Nxe4+ 14 Bxe4 0-0-0 15 c3 Bd6 16 Re1!, N.Bensiek-
J.Baudoin, correspondence 1995. The h2-pawn isn’t hanging and Black’s slight development lead fails
to fully compensate the missing pawn.
d) 6 Nxd7 is White’s main line, which we will cover next in the book.
6 ... Nxe5
The other knight trade 6 ... Nxc3 7 bxc3 Nxe5 8 dxe5 transposes to our
game.
7 dxe5 Nxc3 8 bxc3
 In exchange for the structural damage, White gets the following
compensation:
1. An e5-pawn, which insures a space edge and some kingside attacking
chances.
2. A half-open b-file which discourages queenside castling.
According to current theory, the position remains in dynamic balance.
8 ... Bc5
Black’s bishop takes its most aggressive post:
a) The radical 8 ... Qh4!? was tried in a high-level game.

Question: This looks like an amateurish attempt to attack

when lagging in development. What is Black’s idea?

Answer: Black isn’t attempting to attack. Instead, Black’s queen jumps to h4 to eradicate White’s
possible attacking Qh5 foray. With 9 0-0 c6 10 Be3 Be7 11 f4 g6!?, S.Rublevsky-Wang Yue, Ningbo
2010, Black’s last move was an attempt to set up a light-squared blockade. White should take up the
dare and allow an open g-file after 12 f5! gxf5 13 Bxf5 Be6 14 Rb1 b6 (14 ... 0-0-0?! 15 Rf4 Qg5 16
Qf3 Rd7 17 Bxa7 looks quite favourable for White) 15 Qf3 Rf8 16 a4! (intending to loosen up the
queenside with a4-a5 if Black castles long) 16 ... Bxf5 (16 ... Qxa4?! 17 Ra1 Qc4 18 Bxb6 regains the
pawn favourably) 17 Qxf5 Qe4 18 Qxe4 dxe4 19 Rf4 Kd7 20 Rxe4 Ke6 and Black’s active king offers
full compensation for the pawn.
b) The more conservative 8 ... Be7 development scheme is also fine for
Black. Following 9 Qh5 (hoping to induce weakness with ... g6) 9 ... Be6 10
Rb1 Qd7! 11 Bg5 (The trusting 11 Rxb7?? walks into 11 ... Bg4) 11 ... c6 12
0-0 Bxg5 13 Qxg5, A.Grischuk-V.Kramnik, Mexico 2007, The position
remains balanced after 13 ... 0-0 14 f4 f6 15 exf6 Rxf6 16 f5 Bf7 17 Rf3 Kh8.
9 0-0 0-0
Black has also tried 9 ... Be6 to clamp down on any c3-c4 breaks. After 10
Rb1 Bb6 11 Qh5, Black has a difficult time castling kingside with White’s
queen hanging about. The game followed 11 ... Qd7 12 Bg5 h6 13 Rfe1 c6 14
c4?! (this looks like a dubious pawn sacrifice and the position remains in
sharp balance after 14 h3 Ba5 15 Bd2 Qe7 16 f4 g6 17 Qe2 b5!?) 14 ... dxc4
15 Be4, E.Sutovsky-A.Dreev, Internet (blitz) 2002. Black has the superior
chances after 15 ... Bg4 16 Qh4 Bd8! 17 Bxd8 Rxd8 18 h3 Be6 19 c3 0-0 20
Re3 b6 21 Rg3 Kh8 22 Bc2 f5 when White’s attack is at a standstill, while
Black retains his extra pawn.
10 Qh5
I don’t know how the vast sweep of the queen’s domineering personality
can be crammed into a single body. Kramnik once walked into a mate in one
against a comp in a drawn position. So perhaps Gashimov hopes for the same
kind of luck. White’s queen induces pawn weakness.

10 ... g6

Question: Black’s last move weakens the dark squares. Is it

 worth the risk to induce White to sacrifice on h6, with 10 ... h6?

Answer: Nyet! It’s a big mistake. White can easily lift in enough attackers to get the job done after
11 Bxh6! gxh6 12 Qxh6 f5 13 exf6 Rxf6 (13 ... Qxf6 would be great, if not for 14 Qh7 mate!) 14 Qh7+
Kf8 15 Rae1 (Principle: In a mating attack, don’t chase the enemy king. Instead, cut off his escape
routes first) 15 ... Bg4 16 Bg6 (threat: Qh8 mate!) 16 ... Rxg6 17 Qxg6 Qd7 18 Re3! (the rook can’t be
taken and Rg3 is threatened) 18 ... Bd6 (stopping Rg3) and White plays 19 Rg3! anyway! Hey, I said
“stopping Rg3!” but it transpires that the rook is immune. Black can resign here.
11 Qh6 Qe7

A novelty. Black leaves the rook on f8 and attacks the e5-pawn with the
queen instead. Kramnik intends to gain kingside space with ... f7-f5 next. 11
... Re8 is also good enough for equality and after 12 Bg5 Be7 13 Bxe7 Qxe7
14 Rae1 c5 15 c4 dxc4 16 Bxc4 Bf5 17 c3 Rad8 18 Re3 Qf8 (the queen is
hired on a probationary basis to guard Black’s king) 19 Qg5 Qe7 20 Qh6 (the
queens pad back and forth in their cage, like tigers pacing at the zoo) 20 ...
Qf8, a draw by repetition of moves occurred in A.Kosteniuk-H.Koneru,
Batumi 2012.
12 Bf4 f5!
This is Kramnik’s refined idea. If White captures en passant, Black’s
defenders will soon turn into attackers. If he doesn’t, then White’s kingside
attack is slowed down.
13 Rab1
 Gashimov refuses to capture en passant, seeing that it spikes Black’s piece
activity level after 13 exf6 Qxf6 (now the c3-, f4- and f2-points all come
under attack) 14 Bxc7 Qxc3 15 Bg3 Bf5 16 Rad1 Rac8 and Black stands no
worse.
13 ... b6 14 Rfe1 Be6
We saw this same idea in Kramnik’s game against Navara in this chapter.
Black gives White a passed e-pawn. In return, he gets to blockade White’s
kingside pawn majority, while his own majority remains flexible on the
queenside.
15 Bg5
“The deluded of this world are sorely in need of my wise council,”
declares the bishop, whose eyes insultingly remain on Black’s flustered
queen.
15 ... Qf7 16 Qh4 c6 17 a4 Rfe8 18 Qf4

An assassin covertly studies the habits/patterns of her victim-to-be. White

clears the way for the pawn thrust h2-h4-h5.
18 ... Be7
Going after White’s most dangerous attacking piece.
19 Bh6
The h-pawn rush 19 h4 can be met with 19…Bxg5 20 Qxg5 Qg7 21 h5 Qe7! 22 Qe3 g5!. For
Black’s king, his survival is 100% dependent upon the efficiency of his e7-defender. I like Black’s
chances since his king remains safe, while White’s weakened pawns could prove to be an issue in an
ending.
19 ... c5
Activating his queenside pawn majority and following the principle: Meet
a coming wing attack with a central distraction.
20 h4
Gashimov hopes to have time for the rook lift Re3-g3 and the pawn thrust
h4-h5.
20 ... Red8
Black has a risky alternative plan in 20 ... c4!? 21 Be2 Rac8 22 h5 Rc5!?
with the idea of putting pressure on the a-pawn with ... Ra5. This would place
a burden on White’s kingside attack to provide results.
21 h5
A lone thundercloud appears over Black’s king.
21 ... d4
The central counter arrives.
22 c4

Exercise (critical decision): White’s last move is based upon the philosophy: When
you sink capital into a risky investment (White’s already weakened structure),
sometimes you must protect it by throwing in even more money (the c4-pawn),
just to keep it afloat. White just offered us the c4-pawn. Should we take it?
 Answer: White’s last move was a strong one and we should decline the poisoned c4-offer.
22 ... gxh5!
Kramnik grabs a pawn, but not the one Gashimov had hoped for. Black’s
position exudes a patina of sangfroid and understated defensive capability.
The environment, which is just barely on the fringes of civilization, is about
to fall into a state of utter lawlessness. The inclusion of this externally ugly
move adds a vital and necessary dimension to the defence. Silence can either
mean assent or shocked disproval. In this case, Black decides to speak his
mind with the idea of generating counterplay down the g-file, as well as to
win a pawn. Most definitely not the greedy 22 ... Bxc4?? as White’s attacking
chances increase exponentially with 23 hxg6 hxg6 24 Bxc4 Qxc4 25 e6!
(cutting Black’s queen off from the defence when White threatens Qe5 with a
winning attack, no matter how Black responds) 25 ... Rd5 (also, 25 ... Qd5 26
Qg3 Kh7 27 Bf4 Rh8 28 Be5 wins) 26 Qc7 Re8 27 Qg3 Kh7 28 Bf4 (threat:
Qh2+ and Qh6) 28 ... Rh8 29 Be5 d3 30 Bxh8 dxc2 31 Rbc1 Kxh8 32 Qxg6
and White wins.
23 a5
Gashimov rasps away at Kramnik’s defensive barrier with endless
patience. In this way, he eliminates his weak a-pawn.
23 ... Rab8 24 axb6 axb6 25 Ra1
White can try to put pressure on the b6-pawn with 25 Rb3 Qg6! 26 Reb1
(26 Be2 h4 is okay for Black) 26 ... Bd7 27 Ra1 Rb7 28 Rba3! Be6 29 Ra8
Rbd7 30 Rxd8+ Rxd8 31 Ra6 Qg4 32 Rxb6 Kf7 33 f3 Qxf4 34 Bxf4 h4 and
White may have a microbe of an edge in the ending but my feeling is Black
should hold the game.
25 ... Rd7 26 Ra3 Kh8!
Kramnik utilizes the open g-file for counterplay.
27 Rea1 Rg8 28 Be2 h4 29 Ra7 Bd8 30 Rxd7 Qxd7 31 Ra8?!
White maintains the balance with 31 Qf3, controlling the b7-square.
31 ... b5! 32 Ra6??
White has a flashy continuation to hold the game with a series of
deflection sacrifices after 32 cxb5 Bd5 33 e6 Qxe6 34 Rxd8! Rxd8 35 Bg7+!
Kxg7 36 Qg5+ Qg6 37 Qe7+ Kh6 38 Qxh4+ Kg7 39 Qe7+ Kh6 and it is
perpetual check!
Let’s return to the text continuation 32 Ra6??.
 This idea would work, if not for a single consideration.

Exercise (combination alert): Have you ever looked into your confident
opponent’s eyes after making a move, and then produced that awful
thought: “Does he know something about the position I don’t?” White, a
world class player, just blundered big time in what was an even position.
What did he miss? (I’ve already provided the clue just a short while
ago…)

Answer: Double attack. White hangs a full rook to cover the mate threat on g2.
32 ... Qb7! 0-1
It becomes obvious that the queen, if anything, understated her capability
as a hired killer on her resume. “If you are too poor to buy the object of your
desires, the next step is to steal it,” thinks the queen.
Summary: This position remains in dynamic balance in the 6 Nc3!? line. White’s increased piece
activity makes up for the weakened pawn structure.

Game 9
G.Guseinov-V.Kramnik
Azerbaijan vs. the World match, Baku 2009

1 e4 e5 2 Nf3 Nf6 3 d4 Nxe4 4 Bd3 d5 5 Nxe5 Nd7 6 Nxd7

 White’s main line. He rids himself of the unbalancing ... Nxe5 dxe5 issue.

Question: But by taking on d7, didn’t White just increase Black’s development?

Answer: Correct, but it also did something else: It made Black’s d5-pawn less secure, due to Qh5
issues. Stay tuned and we will see this theme played out.
6 ... Bxd7 7 0-0
A few games later we look at early Nc3 lines, on the seventh and eighth
moves.
7 ... Bd6
This move is in essence a sound pawn sacrifice. We don’t have room to
cover in detail the very risky branch 7 ... Qh4!? which in my opinion, favours
White after 8 c4 0-0-0 9 c5! (cutting off ... Bd6, while gaining useful
queenside space which may later aid in an assault upon Black’s king) 9 ... g5
(with the intention of developing the bishop to g7) 10 Nc3 Bg7 11 g3 Qh3 12
Nxe4 dxe4 13 Bxe4 Bb5 (a double attack on the f1-rook and the d4-pawn) 14
Bg2 Qf5 15 Be3! (a promising exchange sacrifice) 15 ... Bxf1 16 Bxf1 Rhe8
(16 ... Bxd4?? is a bad bargain for Black after 17 Bxd4 Qe4 18 Bxh8 Rxd1 19
Rxd1 when White gets way too much material for the queen) 17 Qa4 Kb8 18
Rd1 c6 and White stands better. For the exchange, he gets the bishop pair, a
pawn, and attacking chances against Black’s king, V.Anand-V.Ivanchuk,
Linares 1993.
8 c4
This undermining move is White’s best continuation and the main line.
8 Qh5 is an obvious try:
a) An imprecise continuation is 8 ... Nf6 9 Re1+ Be7 (9…Kf8 which
disrupts Black’s castling and keeps the h8-rook out of play for a while,
leaving White with a strong initiative) 10 Qe2 Be6 11 f4 (intending f4-f5) 11
... g6 12 Nd2 and Black has an unpleasant future since f4-f5 ideas hover,
H.Hunt-R.Griffiths, England 2013.
b) Black should counter with the pawn sacrifice 8 ... Qf6!:
b1) Risky is 9 Qxd5!? Bc6 10 Qc4 (10 Bb5?! wastes time and is met with
10 ... 0-0-0! threatening ... Bxh2+) 10 ... 0-0-0 and Black’s big development
lead easily compensates the pawn. The game continued 11 Be3 Bf4 12 Na3
Bd5 Not 13 Qb4?? (correct was 13 Qa4 Ng5! 14 Bxf4 Qxf4 (Black threatens
mate with ... Nf3+!) 15 f3 Kb8 16 Rae1 Bc6 and Black regains his sacrificed
pawn with an even position) 13 ... Ng5! (awful sacrificial threats loom over
the f3-square) 14 Be2 (covering the f3-square, but this isn’t enough) 14 ...
Bxh2+! 15 Kxh2 Qh6+ 16 Kg1 Nh3+! 17 gxh3 Qxh3 and incredibly, the
players agreed to a draw in C.Callow-H.Garcia Rojas, correspondence 2003.
White is certainly lost here, as 18 f3 runs into 18 ... Qg3+ 19 Kh1 Rd6 and
White must hand over his queen.
b2) White may have been better off with the less greedy 9 Nc3 Qxd4 10
Be3 Qe5 11 Qxe5+ Bxe5 12 Nxd5 Nf6 13 Rae1 Nxd5 14 Bd4 0-0 15 Bxe5
Nb4, Z.Efimenko-Li Chao, Ningbo 2011. White’s extra pawn is pretty much
meaningless for winning purposes after 16 Bxc7 Nxd3 17 cxd3 Bb5 18 Re3
Rae8 19 Rfe1 Rxe3 20 Rxe3 Rc8 21 Re7 Bxd3 (Black regains the lost pawn)
22 Bf4 Ba6 with a very likely draw.
In Game 12, we will look at 8 Nc3.
8 ... c6
Covering the d5-pawn and making space on the c7-square for the bishop,
in case White pushes his c-pawn to c5.

Question: Instead of bolstering the centre with 8 ... c6, can Black
get away with castling here, just audaciously offering the d5-pawn?
 Answer: 8 ... 0-0 is a rarely played idea, which actually looks quite playable for Black after 9 cxd5
(9 c5 Be7 10 Qc2 Re8 also looks perfectly fine for Black) 9 ... f5 10 Nc3 Qf6 11 Qc2 Rae8 12
Ne2.White’s extra pawn doesn’t mean much and it appears to me that White only stands a shade better,
V.Orekhov-C.Sirotti, correspondence 2011.
9 Nc3

White maintains central pawn tension. In the next game, we will look at 9
cxd5.
 9 ... 0-0
9 ... Nxc3 10 bxc3 0-0 11 cxd5 cxd5 12 Qh5 transposes to the game.
10 cxd5

Question: Can White grab a pawn with 10 Nxe4?

Answer: Black immediately regains it with the double attack trick 10 ... dxe4 11 Bxe4 Bxh2+! 12
Kxh2 Qh4+ 13 Kg1 Qxe4 14 Re1 Qg6 15 Bf4 Rfe8 and White got nothing from the opening, B.Fister-
F.Sage, correspondence 2011.
10 ... Nxc3 11 bxc3 cxd5 12 Qh5

This double attack on the h7- and d5-pawns usually means Black will go
one short. But this is not the end of the story since Black gets full
compensation, according to current theory.
12 ... g6!?
This is a sound pawn sacrifice.

Question: Is the sacrifice obligatory?

Answer: No. Black can also simultaneously cover both the h7- and d5-pawns with 12 ... f5.
However, after 13 Qf3, now, his best move here is to offer the d5-pawn anyway with 13 ... Kh8! 14
Bd2! (if 14 Qxd5?? Bc6 15 Qb3 Bxh2+! 16 Kxh2 Qh4+ 17 Kg1 Bxg2! (Emanuel Lasker came up with
this double bishop sacrifice idea in a simul game, and ever since then, it has become a stock tactical
theme) 18 Kxg2 Qg4+ 19 Kh2 Rf6! 20 Qxb7 Qh4+ 21 Kg1 Rg6+ 22 Qg2 Rxg2+ 23 Kxg2 Qg4+ 24
Kh2 Qf3 (White has loads of material for the queen, with two rooks and a bishop but his king remains
deathly insecure) 25 Ba6 Rb8!. Black prepares to swing the rook over to the kingside via b6 and White
is curiously helpless to do anything about it) 14 ... Qh4 (threatening mate) 15 h3 Bc6 (finally, Black
deigns to protect the d5-pawn) 16 Rfe1 Rf6 17 Rab1 Raf8. The players agreed to a draw in this equal
position, P.Schuster-T.Turgut, correspondence 2009.
13 Qxd5
So White won a pawn, but at a high cost.
13 ... Qc7 14 Bh6

In this position, White has also tried:

a) 14 c4 Bc6 15 Qg5 Bxh2+ 16 Kh1 Bd6 17 d5 (17 Bb2 Rae8 18 Rae1 f6!
19 Qh6, D.Korze-J.Helbich, correspondence 2008. Black equalizes after 19 ...
Rxe1! (exchanges benefit Black as they help to nullify White’s slight spatial
edge) 20 Rxe1 Re8 21 Rxe8+ Bxe8 (White’s central pawns are no longer so
scary and the position remains dynamically balanced) and 17 ... Rae8! solves
all of Black’s difficulties:
a1) 18 f4 (18 dxc6? doesn’t give White enough for the queen after 18 ...
Re5 19 Qh4 Rh5 20 Qxh5 gxh5 21 Bh6 Rd8 22 cxb7 Qxb7 23 Rae1 Qb2
when Black has all the chances in this position, since White’s rook and
bishop fail to generate enough attacking counterplay for Black’s queen) 18 ...
f6 19 Qg3 Bd7 and now, White bailed out with perpetual check after 20 Bxg6
hxg6 21 Qxg6+ Kh8 22 Qh6+, A.Motylev-V.Kramnik, Moscow 2005.
a2) 18 Bb2 Be5 (challenging White’s control of the a1-h8 diagonal ensures
full equality) 19 Qd2 Bd7 20 Rfe1, Y.Dothan-T.Tiits, correspondence 2009.
The position is balanced after 20 ... b6.
b) White can give up his c3-pawn with 14 Qf3 to try and sneak his queen
into the f6-square, in conjunction with Bh6. After 14 ... Qxc3:
b1) 15 Qf6 is met with 15…Qc6!, not giving White time for Bh6 due to
the threat to win White’s queen with ... Bxh2+. After 16 Bg5 Rfe8 17 Rac1
Re6! (a perfect answer, which equalizes completely) 18 Rxc6 Rxf6 19 Rc3
Re6 and Black stands no worse, S.Grigoriants-A.Motylev, Krasnoyarsk 2007.
b2) 15 Bh6

15… Qxd4! (I like this exchange sacrifice better than 15 ... Rfe8 16 Qf6
Bf8 17 Bxf8 Rxf8 18 Be4 when I prefer White by just a microbe) 16 Bxf8
Bc6! (this clever zwischenzug protects Black from Qxb7 tricks) 17 Be4 Bxe4
18 Rad1 Bxf3 19 Rxd4 Be5 20 Rd2 Bf4 21 Rd4 Be5 22 Rd2 Bf4 23 Rd4 Be5
24 Rd2 and a draw was agreed in R.Edouard-L.Fressinet, Caleta 2010.
c) White can also remove the threat of …Bxh2+ with 14 g3 Be6 (this
move regains the lost pawn) 15 Qf3 Qxc3 16 Rb1 Qxd4 17 Be4 Qc4 18 Rxb7
Rab8 19 Qf6 (19 Rxa7 Bc5 is awkward for White) 19 ... Bh3 (threatening
mate on f1) 20 Re1 Qe6 21 Qxe6 fxe6! 22 Be3 Rxb7 23 Bxb7 a5 with an
even ending, F.Vallejo Pons-B.Gelfand, Leon 2010.
14 ... Rfd8
Vastly inferior is 14 ... Bxh2+? as not all zwischenzugs work. This is a bad
bargain for Black after 15 Kh1 Rfe8 16 f4 Re3 17 Qg5! (threatening Qf6) 17
... Qd8 18 Rf3 Qxg5 19 Bxg5 Rxf3 20 gxf3 f6 21 Kxh2 fxg5 22 fxg5 with an
extra pawn and a winning position for White.
15 Qg5
Anand has essayed 15 h4 in a blindfold game against Kramnik. After 15…
Be6 16 Qg5 Qxc3 17 Rfd1 Bf8 18 Bxf8:
a) In his game with Anand, Kramnik played 18 ... Rd5?! (this zwischenzug
allows White to take over a mild initiative) 19 Qf6! Rxf8 20 Be4 Rh5 21
Rac1 Qa5 22 f3 Rb5 23 a4 Rb6?! 24 d5! Bf5 25 d6 Bxe4 26 fxe4 Qxa4 27 h5
Qxe4 28 Qe7! (28 h6?? is met with 28 ... Qe3+ and the h-pawn falls) 28 ...
Qb4 29 h6! and Black was unable to meet the dual threats of 30 Qf6, when ...
Qb2 was met with Rc3, and also the secondary threat 30 Qxf8+!, followed by
d6-d7, V.Anand-V.Kramnik, Amber Blindfold (rapid), Nice 2009.
b) 18…Kxf8! (this is the move Kramnik should have played, and the one
which I am certain he had ready for Guseinov if given a chance) 19 Qh6+
Kg8 20 h5 Rxd4 21 hxg6 hxg6 22 Rac1 Qb2 23 Rb1 Qxa2!? (23 ... Qc3 is a
repetition draw) 24 Bxg6 Rxd1+ 25 Rxd1 fxg6 26 Qxg6+ Kf8 27 Qf6+ Bf7
28 Qd6+ Kg7 29 Qe5+ and a draw is reached.
15 ... Bxh2+
A new move and an attempt to improve upon 15 ... Qxc3 when White
holds the cards after 16 Rfd1 Ba4?! (16 ... Bf8! may be better as 17 Bxf8
Rxf8 18 Rac1 Qa3 19 Bc4 Rae8 looks fine for Black) 17 Rac1 Qa3 18 Rc7
Rd7 19 Rxd7 Bxd7 20 Bc4 (Black’s king finds himself without adequate
defenders) 20 ... Bf8 21 Qd5! Bxh6 (21 ... Qe7 22 Qxb7 Rc8 23 Bxf8 Kxf8
24 Bb3 leaves Black down a pawn without compensation) 22 Qxf7+ Kh8 23
Qxd7 and Black was down a pawn and busted, S.Rublevsky-A.Volokitin,
Poikovsky 2008.
16 Kh1 Bd6 17 c4
White finally protects his c-pawn while activating his centre.
17 ... Bf8
White threatened Qf6 and moreover, every swap helps the side with fewer
pawn islands (Black has two compared to White’s three). Also, White’s
attacking chances recede with each trade.
18 Bxf8 Rxf8
 Question: Does White stand better? After all, he is the one who controls the centre.

Answer: I think the game is dynamically balanced, White’s central pawns are visually impressive,
but on the flip side, he must also protect them. White’s passed d4-pawn may later become a target, and
if the pawn is induced to the d5-square, then both central pawns have been fixed on the same colour of
White’s remaining bishop – a liability in any ending.
19 c5!?
This move is exceedingly risky, since it chronically weakens d4. On the
plus side, the move opens a path for Bc4.
19 ... Rad8 20 Be4?!
The bishop seizes the wrong diagonal. Correct was 20 Bc4!, adding
pressure to the f7-pawn. The game remains balanced after 20 ... Be6 21 Bxe6
fxe6 22 Qe3 Rf4! and if 23 Qxe6+ Qf7 24 Qxf7+ Kxf7 25 Rab1 Rd7 26 d5
Rc4 27 Rfc1 Rxc1+ 28 Rxc1 Rxd5 29 c6 bxc6 30 Rxc6 Ra5 31 Rc2 with a
likely draw, although Black does have the more active rook and king.
 Exercise (combination alert): White’s last move was a misjudgement.
Find Kramnik’s powerful response which wins a pawn:

Answer: Challenge White’s bishop without fear of d4-d5, since Black has a tactical resource.
20 ... Bc6!
Black wins a pawn, no matter how White plays it. I have a feeling that
Guseinov didn’t see this one coming. Kramnik induces the d-pawn to
advance as the alternative was to allow Black to start a siege on the backward
pawn.
21 d5
This leads to a tough ending. Maybe he should try 21 Bxc6 Qxc6 22 Qg4
Qa4! 23 Rfd1 Rd5 24 Rab1 Rxc5 25 Rxb7 Rc2 26 Kg1 Rxa2 27 Rbb1 though
White remains down a pawn.
21 ... f6!
This move may be the one which White missed or misevaluated. White’s
harmony falls out of synch.
22 d6
No better is 22 Qd2 as 22 ... Qe5 23 Rfe1 Bxd5 picks off a key central
pawn for no compensation.
22 ... fxg5?!
 The ending isn’t so certain after this inaccuracy. More decisive is 22 ...
Qxd6! 23 cxd6 fxg5 24 Bxc6 bxc6 25 Rfd1 Rf6! (25 ... Rxf2 26 Rac1 Rxa2
27 Rxc6 and White’s passed d-pawn offers serious counterplay) 26 Rab1
Rfxd6 27 Rxd6 Rxd6 28 Rb8+ Kf7 29 Rb7+ Ke6! when the h- and a-pawns
are inconsequential as Black’s king is required to help his c-pawn down the
board. White is busted after 30 Rxh7 c5! since ... Rc6 is threatened and
Black’s king can help shepherd the passed c-pawn down the board while
White’s king can only view the action from afar.
23 dxc7 Rc8
The c7-pawn is doomed.
24 Bxc6 bxc6 25 f3 Rxc7 26 Rfd1
White intends Rd6, shackling Black’s c7-rook to the c6-pawn.
26 ... Rf5 27 Rac1
Perhaps White should activate his rook with 27 Rd8+! Kf7 28 Rad1 a5! 29
Rc1! Rf4 30 Kh2 g4 31 Kg3 Ra4 32 Rc2 gxf3 33 Kxf3 Kf6 34 Rd6+ Kf5 35
Kg3 with better drawing chances than in the game, since Black’s c7-rook
remains chained to its c-pawn.
27 ... Rd5!

Kramnik blocks White’s intended Rd6.

28 Re1?!
Instead, 28 Rxd5! is not so easy for Black to convert after 28 ... cxd5 29
Kg1 Kf7 30 Kf2 d4! (this cuts off Ke3 and is stronger than 30 ... Ke6? 31
Ke3 Ke5 32 Kd3 and Black is unable to make progress) 31 Rc4 Ke6 32 Rxd4
Rxc5 33 Re4+ Kd6 34 Rd4+ Rd5 35 Ra4 a5 36 Ke3 when Black’s win
remains in question.
28 ... Kf7 29 Re4 a5 30 Ra4 Ra7 31 Kh2?!
This move is Houdini’s first choice but looks dubious to me. The move is
inexplicable especially for a human to play, since it is encoded into our DNA
to never allow an opponent access to our seventh rank. It would have been
more prudent to cover the second rank with 31 Rc2.
31 ... Rd2
Of course, Black annexes the second rank.
32 a3 h5
Black is improving his position on the kingside.
33 Re4 Re7!
The once passive black rook is activated with gain of tempo.
34 Ra4

Exercise (critical decision): Black has two plans: a) 34 ... Ree2, doubling rooks
on the seventh rank. But then comes 35 Rg1 and our a5-pawn falls. b) 34
... Ra7,
protecting the a5-pawn, but this is in reality a non-plan, since Black hasn’t
 made progress. It feels like Black is faced with two troubled choices, but
this
isn’t the case. One plan is superior to the other. Which one should he play?

Answer: Black should double on the seventh rank and sacrifice his a-pawn.
34 ... Ree2! 35 Rg1 Rd3!
The point. Black threatens both ... Rxf3 and also ... Ra2.
36 Rxa5?
This move allows Black to create a super pawn majority on the kingside.
36 Rf1 was necessary to prevent its formation. After 36 ... Ra2 37 Rxa5
Rdxa3 38 Rxa3 Rxa3 39 Rd1 Rc3 40 Rd7+ Kg8 41 Rd6 Rxc5 42 Rxg6+ Kh7
43 Rd6, Black is still far from winning.
36 ... Rxf3 37 Ra7+ Kf6 38 Rc7 g4!

The c6-pawn isn’t important. Kramnik plays for mate in the ending!
39 Rxc6+ Kg5 40 Kh1
Most definitely not 40 a4?? walking into 40 ... Rh3 mate!
40 ... g3!
The f3-rook can’t be touched due to the h2-mating threat. Kramnik doesn’t
even bother with the a3-pawn. Black threatens ... Rf4 and ... Rh4 mate!
41 Rd6
A blunder in a losing position. 41 Rb6! (intending to meet ... Rf4 with ...
Rb4) is met with 41 ... Rxa3 42 Rb4 Rc3 43 Rd4 Rxc5 when conversion is
just a matter of time.
41 ... Rf4!

Threatening mate on the move, but more importantly, forcing the win of
the g2-pawn.
42 Rc1 Rh4+ 43 Kg1 Rh2 44 Rd5+ Kh6 45 c6 Rexg2+ 46 Kf1 Rc2 0-1
Summary: Black’s (temporary) sacrifice of the d5-pawn offers full compensation thanks to the
increased piece activity.

Game 10
B.Macieja-E.Rozentalis
US Open Championship, Reno 1999

1 e4 e5 2 Nf3 Nf6 3 d4 Nxe4 4 Bd3 d5 5 Nxe5 Nd7 6 Nxd7 Bxd7 7 0-0 Bd6 8 c4 c6 9 cxd5
 This time, White exchanges on d5 before playing Nc3.

Question: How does this alter the position from last game?

Answer: White plans Qh5 and Qxd5, just as he did in the previous game. But here, he hopes to
undermine Black’s knight, and this time, he does not have an easy way out with ... Nxc3.

Question: Does this mean that this line is inferior

for Black when compared with last game?

Answer: I don’t think so. Black’s knight may actually represent a strength more than a weakness,
since he is the one who seizes the initiative when White grabs the d5-pawn. The knight will be covered
by ... Bc6, and it can also be useful in harassing White’s queen with a timely ... Ng5, as seen later in
this game.
9 ... cxd5 10 Qh5
Note that 10 Nc3 Nxc3 11 bxc3 0-0 12 Qh5 g6 13 Qxd5 Qc7 transposes to
the previous game.
10 ... 0-0!
We offer the d5-pawn in any case.

Question: What is wrong with 10 ... Nf6?

Answer: It’s the same old story. Black falls behind in development after 11 Re1+ Kf8 (11 ... Be7 12
Qe2 Be6 13 Bb5+ Nd7 14 Nc3 (Nxd5 is in the air) 14 ... a6 15 Bd3 Nf6 (15 ... 0-0? hangs a pawn to the
16 Nxd5! trick) 16 Bg5 0-0 17 Bf5! Bxf5 18 Qxe7 Qxe7 19 Rxe7 Ne4 20 Nxe4 dxe4 21 Rxb7 leaves
Black down a pawn in the ending) 12 Qh4 and Black’s h8-rook remains out of play, P.Chacon-
J.Renteria, correspondence 2003.
11 Qxd5 Bc6 12 Qh5

12…g6
Once again, we eliminate the white queen/rook battery tricks, at the cost of
weakening the dark squares around our king.

Question: What compensation does Black have for the pawn?

Answer: The position resembles a Ruy Lopez Marshall Gambit, where Black has the following
compensation for the missing pawn:
1. Black leads in development, aided by the open nature of the game.
2. Black’s super-active pieces aim ominously at White’s king.
13 Qh3

Question: Why did the White queen back off when the h6-square is available?

Answer: After 13 Qh6 Re8 14 Be3, Black has access to 14 ... Bf8 and White just took a repetition
draw after 15 Qf4 Bd6 16 Qh6 Bf8 17 Qf4 Bd6 18 Qh6 Bf8 in M.Bluvshtein-D.Tyomkin, Toronto
2004. If White tries to win with 19 Qh3, then things can rebound on him after 19 ... h5! (threat: ... Bd7
and ... Bg4, trapping White’s queen) 20 Bc4 Bg7 21 Nc3 Bf6! (renewing the ... Bd7 and ... Bg4 threat)
22 Rac1 Bd7 23 g4? (White should just allow his queen to be trapped after 23 Qf3 Bg4 24 Qf4 g5 25
Qxe4 Rxe4 26 Nxe4 Be7 27 f3 Bf5 28 d5 when White’s rook, piece and pawn offer him reasonable
compensation for the queen) 23 ... Bxg4 24 Qg2 Nxc3 25 bxc3 and White’s kingside pawns and
weakened light squares have Black standing better, M.Luise-P.Bujdak, correspondence 2004.
13 ... Ng5!

A key move to remember. Black generates full compensation for the pawn.
White must now be on a constant lookout for ... Nf3+ sacrificial ideas.
14 Bxg5
Handing over the bishop pair offers Black obvious compensation for the
pawn. Alternatives:
a) 14 Qg4 Ne6 (the knight takes aim at the d4-pawn) 15 Bh6! Bf4 16 Bxf4
(16 Bxf8 runs into Qxd4!, setting up the quadruple threats ... Bxh2+, ... Qxd3,
... Rxf8 and ... Qxb2) 17 Qh3 Ng5 18 Qh4 Qxd3 19 Qxf4 Bxg2! 20 Kxg2
Qh3+ 21 Kg1 Nf3+ and White, having lost his queen, resigned in a few
moves, D.Fengier-V.Shubin, correspondence 2008.).
b) 14 Qh6 Be7 15 Be3 Ne6 16 Nc3 Bf6 17 Rad1 Bxd4 18 Bxd4 Nxd4 19
Ne4 Bxe4 20 Bxe4 Qb6, A.Mikhalchishin-D.Pavasovic, Celje 2003. Black
has equalized and now after 21 Rd3 Qxb2 22 Rh3 Ne2+ 23 Kh1 Qg7
(Black’s queen returns in time to assist her king) 24 Qd2 Nd4 25 Rd3 Nc6 26
Rd7 Rab8, Black is fine.
14 ... Qxg5 15 Nc3 Rae8
Cutting off both Be4 and Ne4 ideas.
16 Rad1
White prepares to cut off the Bc6’s excellent long white diagonal with d4-
d5. Instead, 16 Rae1 Qf4 17 Rd1 Bd7! 18 Nd5 Bxh3, W.Bauer-H.Koch,
correspondence 2004, has White possessing a symbolic extra pawn after 19
Nxf4 Bxf4 20 gxh3 Re7 with an equal position.
16 ... Re7!
Enabling ... Bd7.

17 d5?!
This is White’s main move in the position but I feel it’s inaccurate. 17
Rfe1 looks like the correct way to maintain the balance after 17 ... Bd7 18
Ne4 Rxe4 19 Qxd7 Qd5 (threat: ... Bxh2+, winning White’s queen, while
covering the b7-pawn) 20 g3 Rxe1+ 21 Rxe1 Qxd4, J.Pugh-T.Bueno,
correspondence 2001.
17 ... Bd7 18 Ne4
The only move. If 18 Qf3?? then 18…Bg4 19 Ne4 Bxh2+! 20 Kh1 Rxe4!
21 Qxe4 Bd6 22 Kg1 Qh5! wins a full rook, due to the dual threat of
capturing the d1-rook and delivering mate on the h2-square.
18 ... Rxe4 19 Qxd7 Rd8! 20 Qxb7?!
White should play 20 Qxd8+ Qxd8 21 Bxe4 Qh4 22 Rfe1 Bxh2+ 23 Kf1
Bd6. I feel that White should hold the game here.
 Exercise (critical decision): Black has two candidate moves: a) The tempo gaining 20 ... Bxh2+,
intending ... Rh4+, followed by ... Qh6. b) The immediate 20 ... Rh4.
One of the lines is clearly superior to the other. Which one would you
play?

20 ... Bxh2+?!
After this move, Black’s initiative moves as sluggishly as blood pumped in
a clogged artery. Black’s attack loses a large chunk of its authority, and this
sacrifice should just lead to a draw.
Answer: White is under heavy pressure after 20 ... Rh4! 21 g3 Qg4! (intending to slip into the h3-
square) 22 Rd2 Qh3 23 f4 Bc5+ 24 Kh1 Qxg3 with an obvious advantage for Black.
21 Kxh2 Rh4+ 22 Kg1 Qh6 23 g3
An indication that White continues to play for the full point. Conversely,
23 f3 Qe3+ 24 Rf2 Qh6 25 Kf1 Qe3 26 Kg1 Qh6 is a repetition draw.
23 ... Rh1+ 24 Kg2 Rh2+ 25 Kf3!?
There are some moves which defy all attempts at psychoanalysis of
motivation, but I will try. I sense that White plays this move based on the
philosophy: If I try a risky plan and fail, then I lose something. But if I don’t
try it at all, then haven’t I failed already by default? Boldness, like its
opposite, caution, can be carried too far. The king, casting about for a
potential escape route, hopes to remove himself from the unpleasantness with
ostentatious displeasure. Macieja plays a high-stakes gamble, hoping to win
with his extra piece. Have you ever made a very risky move and then inside
your head, jerk with shock, aghast at your own bravado? You feel like the
meek, tongue-tied office drone, who finally side-swipes and knocks down the
arrogant boss with a right hook in a spasm of rage, and then looms over his
fallen and dazed foe (like Ali did with Sonny Liston). Your somewhat huffy
writer did just that during a summer job 33 years ago, and promptly got fired!
Instead, 25 Kg1 Rh1+ 26 Kg2 Rh2+ is drawn, unless White’s king wants to
take his chances with Kf3.
25 ... Re8!

Principle: In mating attacks, seal off escape routes, rather than give chase.
The counterattack revitalizes and grows, charged with a brand new energy.
26 Qd7??
Placed under pressure, our pieces sometimes become hypnotized by patterns of invalidity. This
move loses. In positions where there is but a single correct answer, we must suppress creative thought
and simply play the role of an accountant, whose sole priority is to make the numbers fit. White had to
try 26 Be4! f5 27 Rd4 fxe4+ 28 Rxe4 Qh5+ 29 Ke3 Rxe4+ 30 Kxe4 Qe2+ 31 Kd4 Qxf1 (Black wins a
rook but his king is prey to perpetual check) 32 Qc8+ Kg7 33 Qc7+! Kh6 34 Qf4+ g5 35 Qd6+ with a
draw by perpetual check, since Black must avoid 35 ... Kh5?? 36 g4+! (double attack/clearance) 36 ...
Kxg4 37 Qxh2 and White suddenly has very real chances to win, due to his passed d-pawn.
26 ... Qh5+! 27 Qg4
Other options lead to an early demise:
a) 27 g4 Qh3+ 28 Kf4 Rxf2+! (removal of the guard) 29 Rxf2 Qe3 mate!
 b) 27 Kf4 g5+ 28 Kf5 Qf3+ 29 Kxg5 Rh5 mate!

Exercise (combination alert): Black to play and win:

Answer: Overloaded defenders.

27 ... Rxf2+!
Sometimes a trap can take us by surprise, not by its complexity, but by its
simplicity. Perhaps White expected 27 ... Qxd5+? 28 Be4 Qb5 29 Qf4 when
he has excellent chances to consolidate.
28 Rxf2 Qxd5+ 29 Kf4
The alternatives are futile too:
a) 29 Qe4 Rxe4 30 Bxe4 Qxd1+ and Black wins.
b) 29 Be4 Qxd1+ 30 Re2 f5 wins as 31 Qf4 Qf1+ 32 Rf2 Qh1+! 33 Rg2
fxe4+ 34 Kf2 e3+ 35 Kf3 Qd1+ 36 Re2 Qd5+ 37 Kg4 Qh5 is mate!
29 ... Qd4+!
The details begin to emerge in sharp focus.
30 Kg5
When defending, the game always feels like it drags on interminably,
minute by minute, second by second. Doesn’t it strike you as strange that in
positions where our king is out in the open and on the run, we suffer a feeling
of agoraphobia, yet our flesh-and-blood body is safe, secure and indoors as
we play the game? It is only the mind which suffers the encroaching fear. In
any case, White loses the queen after 30 Kf3 Re3+ 31 Kg2 Qxg4.
30 ... Qxf2
White is still up a piece but it doesn’t do him a bit of good, since his king
is floundering on g5.
31 Qf4 Qc5+! 32 Bf5
The bishop swallows the offence, giving Black’s queen a glacial stare in
response to the provocation. White had no choice but to hand over a piece as
he gets mated after the king moves:
a) 32 Kg4?? Qh5 mate.
b) 32 Kh4?? Qh5 mate.
c) 32 Kf6?? Re6 mate.
32 ... Qe7+!
The queen regards White’s king with an emotion more complicated than
simple physical attraction, as she continues to enter his thoughts, almost
against his will.
33 Kh6
Running back with 33 Kg4 is useless as after 33…Qe2+ 34 Qf3 gxf5+ 35
Kf4 Re4+, mate is inevitable.
 Exercise (combination alert): White’s defenders are guilty
of dereliction of duty to their king. Continue Black’s attack:

Answer: Cut off the king’s g5-escape route and threaten ... Qg7 mate.
33 ... f6!
Most definitely not 33 ... gxf5?? 34 Qg5+ Qxg5+ 35 Kxg5 Re3 36 Kf4 and
White holds the game.
34 Bxg6 Qg7+ 0-1
The queen enters triumphantly, with a pair of her guards tossing rose
petals before her feet as 35 Kh5 Qxg6+ 36 Kh4 Re4 wins the queen.
Summary: The critical position after 13 ... Ng5! gives Black full Ruy Lopez Marshall Gambit-like
compensation for the pawn.

Game 11
Ni Hua-C.Sandipan
Fujairah Masters, United Arab Emirates 2012

1 e4 e5 2 Nf3 Nf6 3 d4 Nxe4 4 Bd3 d5 5 Nxe5 Nd7 6 Nxd7 Bxd7 7 Nc3

Question: Why does White deliberately allow structural damage?

Answer: White does incur some structural damage. For it, he gets the following:
1. He rids himself of Black’s best piece.
2. He opens the b-file and strengthens the d4-pawn.
 GM Sakaev even adds that “Black must play accurately to equalize”.
White can also go for the same idea a move later with 7 0-0 Bd6 8 Nc3.

This variation is also not dangerous for Black. Following 8 ... Nxc3 9 bxc3
0-0 (we cover the risky 9 ... Qh4!? next in Game 12) 10 Qh5 f5
(simultaneously covering the h7- and d5-pawns), note that White is unable to
take advantage of the e5-hole created, mainly since his knights are off the
board. After 11 Rb1 b6 12 Re1 c6 13 Bg5 Qc7 14 c4 Be8 (14 ... Rae8 is
completely equal for Black) 15 Qh3 dxc4 16 Bxc4+ (16 Bxf5 h6 17 Be4 Bd7
18 Qh5 Be8 19 Qh3 is a repetition draw) 16 ... Bf7 17 Be6 Bxe6 18 Rxe6
Rae8! 19 Qb3 (19 Rxe8 Rxe8 20 Qxf5 Bxh2+ 21 Kf1 g6 22 Qf3 Bd6 also
looks perfectly playable for Black) 19 ... Rxe6 20 Qxe6+ Rf7 21 Qe8+ Rf8
22 Qe6+ Rf7 23 Qe8+ Rf8, a draw was agreed in A.Shirov-V.Kramnik, WCC
challenger match (Game 2), Cazorla 1998.
7 ... Nxc3 8 bxc3 Qe7+!
 I think this disruptive move is a complete answer to Black’s opening
problems in the 7 Nc3 variation.

Question: Why does Black block the f8-bishop?

Answer: Black intends to fianchetto the bishop, which in turn keeps his king safer after castling
short. This move, played in only three games in the database, is an improvement over the book move 8
... Bd6 which allows 9 Qh5 (attacking the d5-pawn and making it difficult for Black to castle kingside)
9 ... Qe7+ 10 Be3 c6 (10 ... Be6 11 Bg5! Qd7 12 0-0 and Black experiences difficulty castling) 11 0-0
g6 (I also prefer White’s chances after 11 ... 0-0-0 12 c4) 12 Qh6 0-0-0 13 c4 dxc4 14 Bxc4 f6 15 Bb3
when c3-c4 is coming and I don’t believe Black has proven equality yet, R.Ponomariov-V.Ivanchuk,
Eilat 2012.
9 Be3 g6 10 Qf3
A new move and perhaps an attempt to improve upon 10 Rb1 b6 11 0-0
Bg7 12 Qc1 0-0 13 Bg5 Qd6 14 Bf4 Qf6 15 Qd2 (15 Bxc7 Qc6 16 Be5 Bxe5
17 dxe5 Qxc3 18 Qg5 Rac8 19 Rfe1 Rfe8 20 h4 Qd4 21 h5 Qg4 is at least
equal for Black, since the c2-pawn looks weaker than the d5-one in the
coming ending) 15 ... c5 16 Be5 Qd8 17 Bxg7 Kxg7 18 c4 dxc4 19 Bxc4
cxd4 20 Qxd4+ f6 21 Bd5 Be6 22 c4 Rc8 23 Rfe1 Re8 24 Rbd1 Bf7 with a
comfortable position for Black, M.Abarca Silva-G.Bronts, correspondence
2010.
10 ... Bc6
Simultaneously covering both the d5- and b7-pawns, while discouraging
c3-c4.
11 0-0 Bg7 12 Qg3 0-0
Black has comfortably equalized. He is structurally solid and not behind
on development.
13 h4
Ni Hua hopes to stir up something on the kingside.
13 ... Qd7
I think Black’s last move is slightly more accurate than loosening his
king’s pawn front with 13 ... h5 14 Rfe1 Rfe8 15 Bd2 Qd7 16 Rab1 a6 17
Bf4 Rxe1+ 18 Rxe1 Rc8, although I don’t see a way for White to make
progress. After 19 Qg5, perhaps White has a minuscule edge, though even
here, I don’t think Black experiences any real trouble holding the position.
14 Bf4
This attack on the c7-pawn is more an annoyance than a real worry for
Black.
14 ... Rac8
14 ... Ba4!? is also possible, when White should probably refrain from 15
Bxc7?! Rfc8 16 Ba5 b6 17 Bb4 a5 18 Ba3 Rxc3 19 Bb2 Rcc8 20 c3 Bb5 and
all White achieved is to earn himself a bad b2-bishop.
15 Rae1 Bb5!

Dual purpose:
 1. Black removes a potentially strong white attacker.
2. Black clears the path for a future ... c7-c5 break, generating queenside
counterplay.
16 Bxb5

Question: Isn’t this a bit obliging on White’s part?

Why not allow Black to trade on d3?

Answer: White’s move is admittedly more strategically risky, since now Black’s structure is clearly
superior. However, he gets nothing by not trading: 16 h5 Rfe8 17 Be5 Bxd3 18 cxd3 Re6 19 f4 f6 20 f5
gxf5 21 Bf4 Rxe1 22 Rxe1 Kf7 (dodging White’s h6-pin cheapo) 23 h6 Bf8 24 Qh3 c5 and Black
doesn’t stand worse.
16 ... Qxb5 17 h5
Conspirators gather in the vicinity of Black’s king, speaking in soft tones
out of the sides of their mouths. White pursues his kingside agenda, while
Black’s king feels the uncomfortable sensation of unseen eyes around him.
After 17 Bxc7 Rfe8 18 Be5 Qd7, Black’s pressure on the doubled and
isolated c-pawns easily compensates the pawn sacrificed.
17 ... Rfe8 18 Be5
White’s plan is to eliminate the g7-defender and then launch a dark square
kingside attack with the queen and the h-pawn.
18 ... Qd7
Covering the c7-pawn and returning to help in kingside defence.
19 Bxg7 Kxg7 20 Qf4
Threatening h5-h6+, followed by Qf6.
20 ... f6 21 f3
After 21 h6+ Kf7 22 g4?! Re6, White risks overextension.
21 ... c6 22 Qg3 Kf7 23 Qh4
Hoping to sneak into the h7-square, which of course, Black won’t allow.
23 ... g5!?
Safer is 23 ... Kg7 24 g4 Rxe1 25 Rxe1 Re8.
24 Qg3 Qf5
 Reminding White that all endings will be tough, due to the long term
queenside pawn weaknesses.
25 Qd6
The line opening idea 25 Rxe8? Rxe8 26 f4 fails miserably to 26 ... g4
which leaves White strategically busted.
25 ... Rcd8
Effectively ejecting the intruder.
26 Qb4
26 Qc7+ is easily repelled by 26 ... Rd7.
26 ... Rd7 27 Rxe8 Kxe8 28 Re1+ Kf7
All the infiltration squares are covered.
29 c4!
 White decides not to fight over low-rent property, like the c2-pawn.
Principle: Activity supersedes material in major piece endings. In such
positions where we attempt to stake out an initiative, it’s generally considered
a mistake to impose an upper numeric limit to our generosity.
29 ... Qxc2 30 cxd5 cxd5 31 Qb5 Qc7
Everything is covered. Having ingested the c2-gift, Black concentrates on
consolidation before making his extra pawn on the queenside count.
32 Qd3
Threatening the h7-pawn.
32 ... Kg7 33 Qf5
The queen, drawing closer, salivates with longing for the black king’s
touch.
33 ... Rd6
 Covering the d5- and f6-pawns as well as the e6-square, all vulnerable
points in Black’s position.

Exercise (planning/critical decision): For White, the data provokes simultaneous

anxiety and reassurance. He is down a pawn but seems to have enough
counterplay to hold the game. He has two paths: a) 34 f4, hoping to strip
Black’s king of his pawn cover. b) 34 Re6, to trade queens based on the
principle: Queen endings are notoriously difficult for the material-up side,
due to the looming threat of perpetual check. One of these hypotheses
contains a hidden flaw. Which one would you play?

34 Re6?!
Principles are not a 100% substitute for analytical thought, and some
should be rejected as exceptions, like this one. The position twists into
strands of perplexity, concerning the correct plan. Black may escape
perpetual check in this version, while winning pawns.
Answer: White should have enough to hold the game after 34 f4! Qd7 35 h6+ Kf8 36 Qd3 gxf4 37
Rf1 Rc6 38 Rxf4 when White’s active queen and rook, coupled with Black’s slightly exposed king,
give him excellent chances to hold the game.
34 ... Rxe6 35 Qxe6 Qc1+
The queen’s sudden entry curdles the white king’s joyful mood.
36 Kh2!
 Weaker is 36 Kf2 as Black can help himself to dinner with 36…Qd2+ 37
Kf1! Qd1+ 38 Kf2 Qxd4+ 39 Kf1 Qa1+ 40 Kf2 Qb2+ 41 Kf1 Qb1+ 42 Ke2
Qxa2+ 43 Ke1 Qa5+ 44 Kd1 Qa1+ 45 Kd2 Qd4+ 46 Kc2 Qc5+ 47 Kd2.
After that buffet, Black simply plays 47…d4 and White lacks perpetual
check.
36 ... Qf4+ 37 g3!
The best try. There is no point playing:
a) 37 Kh1 Qh4+ 38 Kg1 and 38…Qxd4+ picks off the d4-pawn with
check.
b) 37 Kh3?? Qh4 mate is even worse.
37 ... Qxf3
There is still a chance to blow it all away with 37 ... Qxd4?? 38 Qe7+ Kh6
39 Qf8+ Kxh5 40 g4+ Kg6 41 Qg8+ with perpetual check.
38 g4!
Threatening perpetual check.
38 ... Kh6! 39 Qc8!
With the sneaky idea of Qf8 mate while continuing to cover the g4-pawn.
39 ... Qe2+ 40 Kg3 Qe7
How frustrating. Black covers everything.
41 a4 b6!
Halting the a-pawn’s march.
42 Kf2 Kg7
 43 Qb8?!
It’s critical for White’s queen to continue coverage of the f5-square. Black
has a much harder time converting after 43 Kf1! as now if Black continues 43
... Qe3?, White has 44 h6+! Kf7 45 Qd7+ Qe7 46 Qf5! Ke8 47 Qxd5, the
Black’s king is too exposed and White should draw.
43 ... Kh6
The king tells his b8-sister exactly what he thinks about her – but only
when he is safely at an appreciable distance.
44 Kf1
44 Qc8! is still begging to be played.
44 ... Kg7?!
The king, pretending to be deaf to the white queen’s complaints, goes
about his business in surly silence, aware of the white queen’s unwanted
attentions. Instead, 44 ... f5! 45 gxf5 Qf6 is stronger. For example, White
loses the king and pawn ending after 46 Qe5? Qxe5 47 dxe5 Kg7! when
Black’s numerous passers decide.
45 Qh2?
Again, 45 Qc8! is necessary and it’s not clear at all how Black makes
progress.
 The queen trips and falls on her lavish royal rear end. The last move
follows the philosophy: If we are unable to implement a plan, then the next
best thing is an analogue facsimile. The trouble with this idea is White did
indeed have a plan – which was to do nothing and ask Black just how he
plans to make progress.

Exercise (planning): Sometimes we lose games because we

believe we are lost, when in actuality, we may not be. White hopes
he has time for h5-h6+. He doesn’t. Work out a winning plan for Black:

Answer: Pawn breakthrough/luft. Once Black’s king gets air, the win becomes clear.
45 ... f5!
Black untangles the webbed fibrosis of the kingside and his former
difficulties recede into the past. Now his position is winning since his king is
given luft. At this point, I’m certain Ni Hua arrived at one of those awful
“OMG!-what-have-I-done?” moments.
46 h6+ Kf6 47 gxf5 Kxf5
Now Black’s king is inured to his odious h2-sister’s presence.
48 Qc2+ 0-1
White resigns as the position after 48…Qe4 is just a lost position for him.
Black’s dominant queen saunters in and declares, “In the absence of
established authority, the most powerful and wise among us – namely, myself
– will rule”.
Summary: The odd, almost unplayed 8 ... Qe7+! is a complete answer to the 7 Nc3 line.

Game 12
G.Kamsky-A.Karpov
World Championship (Game 6), Elista 1996

1 e4 e5 2 Nf3 Nf6 3 d4 Nxe4 4 Bd3 d5 5 Nxe5 Nd7 6 Nxd7 Bxd7 7 0-0 Bd6 8 Nc3

8 ... Qh4!?
The queen carries herself with a silent arrogance of a person acutely aware
of her own importance. Karpov attempts a theoretical ambush with a rarely
played, risky yet dynamic idea, which resembles Black’s ... Qh4 sweep in the
Ruy Lopez Marshall Gambit.

Question: What is so risky about it?

Answer: In a few moves, White’s rook will give check on e1, when Black will either be forced into
a pin or lose castling rights.
9 g3
A slight weakening of the dark squares is induced in White’s camp. The
text move is superior to 9 h3?! Nxc3 10 bxc3 0-0 when Black is at least equal
and possibly stands a shade better.
 9 ... Nxc3
This important zwischenzug inflicts some degree of damage to White’s
structure.

10 bxc3

Question: Can White meet our zwischenzug with the counter-zwischenzug 10 Re1+?

Answer: I don’t see a single game in the database with it. Your idea looks quite playable, but on the
flip side, I don’t believe Black experiences any difficulties after 10 ... Be6 11 bxc3 Qh3:
a) The position 12 Bf1 was reached by Mamedyarov and Kramnik via a
slightly altered opening move order. After 12 ... Qf5 13 Rb1 0-0-0 14 Qe2 c6
15 Rb3 Rhe8 16 Be3 Re7 17 Reb1 Rdd7 18 a4 Qg4 19 Qxg4 Bxg4 20 c4
dxc4 21 Bxc4 Bf5 22 Bd3 Bxd3 23 Rxd3 Bc7 24 c4 Bb6 25 Rbd1 Re4 26
Kg2 f5 27 Kf3 g6, the game is balanced and the players agreed to a draw,
S.Mamedyarov-V.Kramnik, Nice 2008.
b) Houdini also points out the line 12 Rb1 0-0-0 13 Qe2 (maybe
contemplating future Rxb7 or Ba6 sacrificial ideas) 13 ... Rde8 14 Be3 h5!?
15 Rxb7! Bg4! (not 15 ... Kxb7?? 16 Ba6+ Ka8 17 Qb5 and Black, unable to
cover both the b7- and c6-squares, gets mated) 16 f3 Rxe3! (deflection) 17
Qxe3 Bxg3 18 Rxc7+! Bxc7 19 Ba6+ Kb8 20 Rb1+ Ka8 (certainly not 20 ...
Bb6?? 21 Qe5+ Ka8 22 Qxd5+ Kb8 23 Qb7 mate!) 21 Bb7+ Kb8 22 Ba6+
Ka8 with a draw by perpetual check.
10 ... Qg4!?
When we try a high-risk plan like this, we must first put common sense on
hold. This move costs Black castling rights, and maybe actually deserves a
‘?!’ mark instead. By allowing this, Karpov fragments his defensive
resources. Much safer is 10 ... Qh3! 11 Re1+ Be6 12 Rb1 0-0-0 which
reaches the Mamedyarov-Kramnik note from above.
11 Re1+ Kd8?!
This plan should collapse under the sheer weight of its own improbability.
Now Black’s king will be insecure for a long time to come. Karpov, who
refuses to allow his creativity to be constrained by mere convention, certainly
seemed to be in a reckless mood. But then the question arises: Why then did
he play Petroff’s, an opening antithetical to a reckless nature? Black’s king
looks safer on the other side of the board after 11 ... Kf8! 12 Be2 Qf5 13 Rb1
b6 14 c4 Re8 15 Bd3 Rxe1+ 16 Qxe1 Qe6 17 Qc3 dxc4 18 Bxc4 Qf6 19 h4
h6 20 Bb2 g5 when Black’s position looks satisfactory to me, in this very
sharp position.
12 Be2 Qf5 13 Rb1 b6
Not 13 ... Bc6? 14 Bd3 Qf6 15 Qh5 (threat: Bg5) 15 ... h6 16 c4 dxc4 17
Bxc4 Re8 18 Bb2 as Black is in deep trouble, since his a8-rook is unable to
participate.
14 c4
Dual purpose: Kamsky undoubles his c-pawns and also opens the centre,
which enhances the danger for Black’s king.
14 ... dxc4 15 Bxc4 Re8
Now we see the plus side of Karpov’s earlier decision to slide his king to
d8: It enables Black’s h8-rook to be activated via the e8-square.
16 Be3 Bc6?!
More risky play!
 Question: Doesn’t Karpov’s last move just lose a tempo?

Answer: It does, but by provoking d4-d5, Black removes the flexibility out of White’s once mobile
central pawns. Black had a better plan in 16 ... Ke7! (intending ... Kf8, since I don’t see a useful
discovery for White) 17 Bd3 Qh3 and after 18 Bf4+ Kf8 19 Bxd6+ cxd6 20 Qf3 Rab8 21 Qd5 Re6 22
Re4 Rh6 23 Rh4! Qe6! 24 Qxe6 Bxe6 25 Rxh6 gxh6 26 a4 a5 27 Kf1 (White loses the initiative after
the greedy 27 Bxh7? Bd7 28 Ra1 f5 29 Bg6 Rc8) 27 ... Bd7 28 Bb5 Bf5 29 Rb2 Rc8 30 Bd3 Bxd3+ 31
cxd3 Rc1+ 32 Ke2 Ra1 33 Rxb6 Ke7, Black should hold the ending.
17 d5 Bd7 18 Bf1
Clearing the way for c2-c4-c5. Another way of putting Black under
pressure is 18 Qd4 f6 19 Bd3 Qh5 20 Qc4.
18 ... h6 19 c4 Re7!
 Up until now, the black king’s defenders have a haphazard look. With his
last move, Karpov deals with a pair of long neglected tasks.

Question: Why did Karpov’s last move get an exclamation mark?

Answer: Black has two issues to resolve:

1. His king’s safety in the centre of the board.
2. His out-of-play a8-rook.
In one move, Karpov intends to solve both issues, by uniting the
disconnected pair with a single idea:
Step 1: Transfer his king out of the danger zone, with ... Ke8 and ... Kf8.
Step 2: Activate his sleeping a8-rook with ... Re8.
Now having said this, I still don’t believe Black can fully equalize.
20 Bd3
This move is incommensurable with White’s dreams of victory. A good
way of putting Black under pressure is 20 Qc1!, preparing to force open the
kingside with c4-c5 next. For example 20…Ke8 21 c5 bxc5 22 Bxc5 Rxe1 23
Qxe1+ Kf8 24 Bxd6+ cxd6 25 Rb7 and not only is Black tied down, his a-
pawn is also weak.
20 ... Qf6 21 Kg2
It’s dangerous to keep the king on a light square. White’s advantage
dissipates somewhat after 21 Bf1 Bf5 22 Rc1 Ba3 23 Bd4 Qg6 24 Rxe7 Kxe7
25 Rc3 Bb4 26 Re3+ Kf8 27 Bg2 Bd6 when Black is very close to equality.
21 ... Ke8 22 Bc2
Worth considering is 22 Qc2, covering the c3-square from the black
queen.
22 ... Qc3
Targeting the c4-pawn. Black has equalized.
23 Bb3 Kf8 24 Rc1 Qf6

25 Bc2

Question: Why didn’t White go for the c5-break

when he had the chance to do so at this point?

Answer: Kamsky probably feared the line 25 c5 Rxe3! 26 Rxe3 Bxc5 27 Rf3 Qd6 28 Qd3 Kg8
when Black gets full compensation for the exchange with one pawn, the bishop pair, dark square
control and a weakness-free position.
25 ... Rae8
Now it appears as if Black castled kingside all along.
26 Qd3?
Perhaps thinking about an h7-infiltration, White allows the light squares
around his king to be weakened irretrievably.
26 ... Bg4!
 Speaking of infiltration, Karpov eyes the invasion of the f3- and h3-
squares.
27 Bd2??
Mistakes usually come in pairs and the game goes downhill from here:
a) 27 Qh7? also does nothing to bother Black after 27 ... g5 28 Kg1 Bb4 29
Rf1 Be2 and Black picks off the exchange.
b) It is imperative to contest the light squares with 27 Bd1 though Black
still holds the edge after 27…Bf5 28 Qb3 Be4+ 29 Kg1 Kg8.

Question: Telling White’s oblivious bishop that his last decision

is a misjudgement, is a bit like telling a skydiver that it may be a good
idea to first put on a parachute before jumping from the plane. White’s
last move was a monster blunder. How did Karpov force the win?

Answer: Step 1: Infiltrate the second rank and press on the f2-pawn.
27 ... Re2
A new social dynamic comes into play, with the roles of attacker and
defender reversing.
28 Rxe2 Rxe2 29 Rf1 Rxd2! 0-1
Step 2: Deflection/removal of a key defender. White’s most pessimistic
fears are now officially confirmed, and the events couldn’t have worked out
better if Karpov had been in charge of the arrangements himself. The players
transact business on d2 – a deal which only favours one party. Recapture is
impossible, as after 30 Qxd2 Qf3+ (Black’s queen just foreclosed on the
white king’s final defensive option) 31 Kg1 Bh3, White’s king sees the
malicious, grinning face of his own worst nightmare. “Our g1-cousin’s
childlike attempts to cling to power are refreshingly artless,” comments the
bishop. This is that awkward moment when two sets of hands reach for the
last Oreo cookie on the plate.
Summary: The risky 8 ... Qh4!? can be used as an ambush weapon. At the
club level, it may take your opponents completely by surprise. I think Black’s
best line is 10 ... Qh3!, which leads to the Mamedyarov-Kramnik note, rather
than Karpov’s riskier 10 ... Qg4!?, which may even deserve a dubious mark.
Chapter Three
The Old Main Line
To gain something of value, normally one must first make an investment
outlay. In the case of this chapter, our investment is time. We must study our
lines deeply, since one slip could mean turning a happy ‘=’ evaluation into
the dreaded ‘+-’. We arrive at the incredibly theoretically dense Old Main
Line, where we face three battlegrounds which shimmer in reflection in our
thoughts. Sometimes I sigh wistfully about the Golden Age (pre-comp; pre-
database) days, where we weren’t required to learn reams of theory, in
contrast to the contemporary squalor of having to be omniscient for the first
18 or so moves.

Firstly, White can try to exploit the knight’s e4-placement by playing to

undermine it with c2-c4, destabilizing Black’s centre in the process.
 White can also play quietly, with c2-c3 and Qb3 to follow. In the position
below, we sometimes sacrifice our b7-pawn to generate a kingside attack.

In the third diagram, White plays an early Nc3, allowing us ... Nxc3, damaging the opponent’s
structure. In exchange, White gets kingside attacking chances.

Game 13
A.Naiditsch-V.Kramnik
Dresden Olympiad 2008

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7

We don’t have the space to cover 7 ... Bg4 in detail. This pin used to be
Black’s main move in the 1980s and part of the 1990s. After 8 c4 Nf6 9 Nc3
Bxf3 (My long-time buddy IM Jack Peters just sent me the following game,
so I thought I would sneak a portion of it into this book! The continuation
went 9 ... Be7 10 cxd5 Nxd5, J.Peters-V.Kudryavtsev, Orange 2014, and
White stands slightly better after 11 Re1 0-0 12 Be4 Be6 13 a3 Re8 14 Qd3)
10 Qxf3 Nxd4 11 Qe3+ Ne6 12 cxd5 Nxd5 13 Nxd5 Qxd5 14 Be4 Qb5 15 a4
Qa6 16 Rd1, White’s bishop pair and development lead offered him plenty of
compensation for the pawn, G.Kasparov-A.Karpov, World Championship
(Game 6), London 1986.
8 c4
This is White’s main line, which undermines Black’s e4-knight. Later in
the chapter, we also look at key side lines 8 Re1 (Games 19 to 22) and 8 Nc3
(Games 23 and 24).
8 ... Nb4
We go after White’s powerful light-squared bishop, but without gain of
tempo, since if White retreats the bishop to e2 (the main line), we later lose
the tempo back when White plays a2-a3. We won’t be covering the 8 ... Nf6
or 8 ... Bg4 lines.
9 cxd5!?
 Naiditsch is not heedful of the conservative point of view in the argument
and reacts to Black’s last move with a rather surly response. This is an
involved, forcing variation which we need to know well.

Question: It seems to me that White’s last move was strategically suspect. Didn’t
he just hand us the bishop pair, light squares and also take on a d4-isolani?

Answer: All true. At first glance, it feels as if White’s strategic concessions fail to even meet
rudimentary standards of soundness. When we examine the position deeper, we notice that White gets
the following in return:
1. White gains a development lead, since we used up three full moves in
an open position to pick off the d3-bishop. Our development lag is further
aggravated by the fact that our queen sits on the d5-square, which may lead to
tempo loss with a future Nc3 or Re5 from White.
2. The position is wide open, enhancing White’s development lead, with
both e- and c-files open for use.
Our resources are fully adequate, but let’s not underestimate White’s
coming initiative. 9 Be2 is White’s main line, which we look at in the next
four examples (Games 14 to 18) in this chapter.
9 ... Nxd3
One set of events forms the shape of the next in the sequence. If we are
going to suffer, then at least we get paid for it in strategic coin.
 10 Qxd3

10…Qxd5

Question: Is it possible to play the consolidating try 10 ... Nf6?

Answer: The move is playable but I prefer White’s position after 11 Qb5+ Bd7 (11 ... Qd7 12
Qxd7+ Bxd7 13 Nc3 gives White just a shade of an edge in the ending) 12 Qb3 (12 Qxb7 Rb8 and
White must take a repetition draw, since the a6-square isn’t available for the queen due to ... Bb5. After
13 Qxa7 Ra8 14 Qb7, a draw is reached) 12 ... 0-0 13 Nc3 when the extra d5-pawn exerts a cramping
influence on Black’s position, J.Timman-V.Kovacevic, Zagreb 1985. Also, it isn’t clear to me just how
Black plans to regain the pawn.
11 Re1 Bf5
Developing a piece, protecting the e4-knight, while getting White’s queen
nervous about potential knight discoveries in the future.
12 Nc3
Let’s look at other options:
a) White has a more violent option with 12 g4!? Bg6 13 Nc3 Nxc3 14
Qxc3 f6! 15 Bf4 (15 Qxc7 is met with 15 ... 0-0! 16 Rxe7 Qxf3 17 Rxg7+
Kh8 18 Bh6 Qxg4+ 19 Qg3 Qxd4 when Black stands no worse) 15 ... Kf7!
(the c7-pawn is low priority as Black’s king safety comes first) 16 h4 (16
Qxc7 Rhe8 17 Nh4 Be4 and Black’s powerful grip on the light squares,
coupled with White’s loose king-position, easily compensate for White’s
extra pawn) 16 ... h5 17 Qxc7 Rhe8 and here, White has 18 Ng5+! when the
knight goes cartwheeling off the cliff, as if unafraid of gravity. If not for this
move, then Black stands better. After 18 ... fxg5 19 Bxg5 Kg8 (19 ... Qf3 20
Bxe7 Be4 21 Bg5+ Kg8 22 Qh2 Qxg4+ 23 Qg3 Qd7 and Black is easily
compensated for the missing pawn, since his bishop is more potent than
White’s) 20 Bxe7 Qxd4 21 gxh5 Bxh5 22 Qe5 Qg4+ 23 Qg3 Rac8 24 Qxg4
Bxg4 25 Bb4 Be2 26 Rac1, White’s extra pawn is meaningless and the
players agreed to a draw here, T.Radjabov-V.Kramnik, Nice 2009.
b) 12 Ne5 0-0-0 13 Qf3 g6 14 g4!?

Question: Why are top GMs so attracted to the strategically repulsive g4-thrust?

Answer: Well, it’s not the sort of move I would even consider, yet both Radjabov (with 12 g4) and
Adams (in this variation) sensed their initiative compensated for the long term weakness. Houdini
claims g2-g4 is playable in both variations, but fails to earn White an advantage. After 14 ... Bb4:
b1) 15 Re2? 1Bxg4 16 Nxg4 Nc3! 17 Qxd5 (White has better chances to
hold the game after 17 Re3 Qxf3 18 Rxf3 Ne2+ 19 Kg2 Nxc1 20 Nc3 Rxd4
21 Ne5 Re8 22 Nb5 Ne2) 17 ... Nxe2+ 18 Kf1 Rxd5 19 Kxe2 Rxd4 20 Ne3
Rh4 21 Nf1 Bd6 22 Be3 Bxh2, Black’s rook and three pawns proved to be
too much for White’s two minor pieces, M.Adams-J.Smeets, London 2009.
b2) 15 Re3! Qxd4 16 gxf5! (16 Nxf7? Qd1+ 17 Kg2 Qxc1 18 Nxd8 Rxd8
19 gxf5 Nd6! 20 fxg6 hxg6 (I think White struggles as his extra exchange is
quite useless) and now if 21 Qg4+?, 21…Nf5 22 Qxb4 Nxe3+ 23 fxe3 Qc6+
24 Kg3 Qh1 and White plays without his rook and knight, and his king is
dangerously exposed) 16 ... Ng5 17 Qe2 Qd1+ 18 Qxd1 Rxd1+ 19 Kg2 Rxc1
20 h4 Bd2 21 Nxd2 Rxa1 22 hxg5 gxf5 seems like a sharp and approximately
balanced ending.
12 ... Nxc3 13 Qxc3

Exercise (critical decision): Black’s c7-pawn is hanging, and he is also behind in

development. His choices: a) 13 ... c6, furthering his development deficit,
while
protecting his c-pawn. b) 13 ... Be6, Black offers the c7-pawn to catch up
in development. One of the variations is clearly superior. Which one would
you play?

Answer: The position is critical enough to give up the c-pawn in favour of development.
13 ... Be6!
Kramnik decides to answer the challenge without evasion. The c7-pawn is
a far lower priority than getting his king out of the centre. 13 ... c6?! walks
into the shot 14 Bh6! and after 14…gxh6 15 Re5 Qd7 16 Rae1 Be6 17 d5 0-
0-0 18 dxe6 fxe6 19 Rxe6 Bd6, A.Baert-M.Dutreeuw Belgium 2000, I don’t
see Black’s compensation for the missing pawn after 20 Rxh6.
14 Re5
 The rook considers the black queen’s occupation of the d5-square as an
affront. Once again, initiative takes a front seat over strategic or material
concerns. 14 Qxc7 allows Black to unravel after 14 ... Bd6 15 Qc2 0-0 16
Bd2 Bf5 17 Qb3 (17 Qd1 Rac8 18 Bc3 Rfe8 19 Rxe8+ Rxe8 has Black’s
ultra-active pieces easily making up for White’s extra pawn) 17 ... Qxb3 18
axb3, R.Huebner-V.Smyslov, Velden 1983. Black’s bishop pair and superior
structure more than compensate for White’s rather worthless extra pawn.
14 ... Qc6
The queen makes her daily “As-God-as-my-witness!” vows, regarding her
hated e5-cousin.
15 Qa5
Menacing the d4-d5 pawn fork.
15 ... Rd8!

Kramnik decides to scamp his duty to the a7-pawn. Once again, he is

willing to hand over his pawn to further his development, while covering
White’s d4-d5 threat.
16 Bf4
Let’s examine the alternatives:
a) Nobody has tried to grab a pawn with 16 Qxa7 yet. After 16…0-0 17
Qa5 Bf6 18 Rc5 Qe4 19 Rxc7 Rxd4! (threatening a back rank cheapo) 20
Be3 (20 Nxd4?? Bxd4 (threat: ... Bd5) 21 Re7 b6 22 Qd2 Bc5 and Black
wins, since 23 Ra7 Bd5 forces mate) 20 ... Ra4 21 Qc5 Rc4 22 Qd6 Rd8 23
Rxc4 Qxc4 24 Qb6 Qxa2! (once again playing on White’s weak back rank)
25 Qxd8+ Bxd8 26 Rxa2 Bxa2, Black’s bishop pair offers a tiny edge.
b) 16 Bg5 is attempt to rid Black of his bishop pair, at the cost of time.
Following 16 ... Bxg5 17 Nxg5 0-0 18 Nxe6 fxe6 19 Qc3, Black equalized
and the players agreed to a draw, F.Vallejo Pons-R.Kasimdzhanov, San
Sebastian 2009.
16 ... 0-0
After the immediate queen trade offer 16 ... Qb6, 17 Rb5 Qxa5 18 Rxa5 a6
19 Bxc7 Rd5 20 Bb6 Rxa5 21 Bxa5 gives Black full compensation for his
pawn, thanks to his bishop pair, centralized king and light square control,
D.Solak-I.Karkanaque, Plovdiv 2010.
17 Rc1 Qb6 18 Rb5 Qxa5
Now the atmosphere for Black feels more light-hearted with the absence of
White’s queen.
19 Rxa5

An irksome issue arises for Black, whose c7- and a7- pawns hang
simultaneously.
19 ... Ra8!
Black equalizes with this counterintuitive, multi-purpose move:
1. He covers the a7-pawn.
 2. Rxc7 is not possible, due to a ... Bd8! trick.
3. Bxc7 isn’t possible due to the awkward ... Rfc8 pin.
20 d5!
The only way to sustain the initiative.
20 ... Bd7!
Gifted liars tend to add as much of the truth as possible to their fictional
story to disguise and offer cover to their falsehood. Black’s bishop falls under
this category. Frustratingly for White, the c7-pawn again remains
mysteriously covered by unseen, powerful allies.
21 Ne5
21 Bxc7? Bb4 snags the exchange, while 21 Rxc7? Bb4 wins the exchange
all the same. The bishop, who considers himself a fusion of Gandhi, Mother
Teresa, the Dalai Lama and Mr. Rogers, isn’t quite as virtuous as outer
appearances indicate.
21 ... Bd6!
This pin trick fully equalizes. After leaving the c7-pawn hanging for what
seemed like an eternity, Kramnik finally defends it.

22 Bg3 Bxe5
The simplest solution. The opposite-coloured bishops ending is very likely
drawn.
 23 Bxe5 c6
The heroic c-pawn finally survives his long ordeal. He conducted his life
under the assumption that he was just a little bit better than everyone else. As
it turns out, he was right.
24 dxc6
The entire game was a homework contest and Kramnik’s grade gets an
‘A+’ on his assignment. 24 d6 is tempting on the surface, but the pawn will
go nowhere, since it is eternally blockaded. Black stands fully equal after 24
... Rfe8 25 Bd4 b6 26 Rg5 g6 27 Re5 c5 28 Bc3 Rxe5 29 Bxe5 Re8 30 f4
Re6 31 Rd1 f6 32 Bc3 Kf7 and If anyone stands better, it’s Black.
24 ... Bxc6
Now all is as it should be for Black and the position obviously heads for a
draw.
25 Bd4 a6 26 Re5 Rfe8 27 Rxe8+ Rxe8 28 Be3 Rd8 29 f3 f6 30 Kf2 Kf7
31 h4 ½-½
Summary: 9 cxd5 is a long, forcing, homework contest, which we need to know cold. Don’t
underestimate White’s development lead, despite our bishop pair, light squares, control over d5 and
White’s potentially weak d4-isolani. We must remain on high alert, realizing that White hopes to
intercept and take advantage of even a single carelessly unguarded action on our part.

Game 14
V.Anand-V.Kramnik
Corus A, Wijk aan Zee 2010

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 c4 Nb4 9 Be2
 White’s main move. He takes time out to preserve his important light-
squared bishop.
9 ... 0-0 10 Nc3
This move is played much more often than the immediate 10 a3 Nc6 11
cxd5 Qxd5 12 Nc3 Nxc3 13 bxc3 Bf5 14 Re1 Rfe8 15 Bf4 Rac8 and we
transpose back to this game.
10 ... Bf5
Black backs up the e4-knight, which is the goal of White’s undermining
project.
11 a3
With this move, White forces clarification of the centre.

Question: Can’t White save a tempo with 11 Re1?

Answer: It’s a double question mark move, which hangs an exchange to 11 ... Nxc3 12 bxc3 Nc2.
11 ... Nxc3
Amazingly, I see five games in the database where Black plays the rather
baffling 11 ... Nc6?? when White wins a key central pawn for zero
compensation after 12 Nxd5.
12 bxc3 Nc6 13 Re1
Principle: Don’t rush to resolve pawn tension without good reason:
 a) 13 cxd5 Qxd5 14 Re1 Rfe8 is considered dynamically even.
b) 13 Be3 (White’s bishop isn’t as active as it is on the f4-square) 13 ...
Na5! (challenging White for control of the c4 square) 14 cxd5 Qxd5 (now ...
Nc4 is in the air) 15 Nd2 Qc6 (threatening the c3-pawn) 16 c4 Rad8 17 d5
Qg6 18 Qa4 b6 19 Bf3 Bc2 20 Qb5 Nb7! (Black seizes control over the c5-
hole) 21 Rfe1 (21 Qa6? Nc5 22 Qxa7 Bd6 23 Bxc5 bxc5 24 Qa5 Qf6! ( ...
Ra8 is coming) 25 Ra2 Ra8 26 Qb5 Qf4 27 g3 Qxd2 28 Bd1 Qxd1 29 Rxd1
Bxd1 and Black gets too much for the queen) 21 ... Nc5 with advantage to
Black, who had seized control over the central dark squares, J.Morgado-
A.Bormida, correspondence 1995.
c) 13 Bf4 is slightly premature, since Black benefits from an early ... Bd6
challenge. After 13 ... dxc4 14 Bxc4 Bd6 15 Bg5 Qd7 16 Nh4 Na5 17 Bd3
Bxd3 18 Qxd3 Qg4 19 Nf3 Qe6! (fighting for control over the c4-blockade
square) 20 Rfe1 Qd5 21 Be7 Rfe8 22 Ng5 f5 23 Bxd6 cxd6 24 Qh3 h6 25
Qh5 Rxe1+ 26 Rxe1 hxg5 27 Re8+ Rxe8 28 Qxe8+ Kh7 29 Qh5+, the game
ends in perpetual check, R.Ponomariov-L.Dominguez Perez, Elancourt 2013.
13 ... Re8
We prepare to challenge White along the e-file.
14 cxd5
White has a sharp alternative with 14 Bf4 dxc4 15 Bxc4 Bd6 16 Rxe8+
Qxe8 17 Ng5 (going after the f7-pawn) 17 ... Bg6 (the attempt to get three
minor pieces for a queen with17 ... Bxf4?? fails after 18 Bxf7+ Qxf7 19 Nxf7
Kxf7 20 Qb3+ Kg6 21 Qxb7 when the double attack on the a8-rook and c6-
knight is devastating) 18 Bxd6 cxd6 19 h4 Qe7 20 Qg4 h6 21 Nh3 and with
this, White menaces both Qxg6 and the attacking Nf4.
 Here, the strong 21 ... Qe4! equalises for Black:
a) 22 Nf4 is met with the tricky 22 ... Ne5! 23 dxe5 Qxc4 24 exd6 (24 e6
Re8 25 Qg3 Bf5 26 exf7+ Qxf7 is also just fine for Black) 24 ... Qxc3 25 Rf1
Be4 26 Ne2 Qe5 27 Rd1 Rd8 28 Qf4 Qxf4 29 Nxf4 g5 30 hxg5 hxg5 31 Nh3,
R.Sevecek-A.Zilberberg, correspondence 2002. If anyone stands better in this
ending, it is Black as after 31 ... f6, the d6- pawn may be a target when
compared to Black’s healthy queenside pawn majority, and Black’s bishop
looks superior to White’s knight.
b) 22 Qg3 Na5 23 Ba2 Qc2 24 Bd5 Qd3 25 Qxd6 Qxc3 26 Rf1 Bd3 27
Rd1 Bc2 28 Rf1 Bd3 29 Rd1 Bc2 30 Rf1 Bd3 and a draw was agreed, Zhou
Weiqi-Bu Xiangzhi, Danzhou 2013.
14 ... Qxd5 15 Bf4
Luring Black’s rook to the c8-square, when it prefers to play to the d-file
instead. Less accurate is 15 Be3 since Black’s a-rook reaches a superior post
after 15 ... Bf6 16 Nd2 Rad8. Black’s position is slightly better here than in
our tabiya position, since his queenside rook sits on d8, rather than the more
passive c8-square, V.Bologan-J.Degraeve, Belfort 2002.
15 ... Rac8
 We reach one of the most critical tabiya positions in the entire book. These
are scary computerized times when theory essentially begins on move 15.
The general consensus has shifted to 15 ... Rac8, over the older 15 ...
Bd6.The point of placing the rook to c8 is to play an eventual ... Na5, and
then activate with ... c5, with Black’s rook backing up the c-pawn.

Question: Black’s last move, 15 ... Rac8 looks awfully passive.

Why not challenge immediately with 15 ... Bd6?

Answer: This is still playable and was Black’s main continuation about two decades ago, but has
since been superseded by 15 ... Rac8. The trouble with moving the bishop to the d6-square is that the
bishop is better placed on f6 instead, where it keeps an eye on the d4-pawn, and this in turn discourages
White from expansion with c3-c4. After 15…Bd6 16 c4 Qe4 17 Be3 Rad8 and now 18 Ra2! cuts off ...
Qc2 ideas and prepares to bolster support of the d4-pawn if Rd2 is needed. I think White secured a tiny
edge here, although Black’s game is obviously very playable. Following 18 ... Bg6 19 Qc1 Na5 20 c5!
(a committal yet powerful move, not surprising as Kasparov is not a man known for his patience on the
chess board) 20 ... Be7 21 Bb5 Qd5, G.Kasparov-A.Karpov, New York (rapid) 2002. At this point
White can secure a clear advantage with 22 Bxe8! Qxa2 23 Ba4 Bf8 24 Bf4! which leaves Black in
obvious trouble, since there is no good way to protect the c7-pawn. Note that instead of Karpov’s 21…
Qd5, 21 ... c6 would still lead to a White plus after 22 Bf1 Qd5 23 Rb2 b5 24 cxb6 Bxa3 25 bxa7 with
huge compensation for the exchange, thanks to the very strong passed a-pawn.
16 h3

Question: Is there any point to White’s last move?

 Answer: White’s last move serves two functions:
1. White creates air for his king, negating future back rank issues.
2. White can meet Black’s ... Bf6 ideas with Nh2!, intending Ng4 to harass
Black’s f6-bishop. We will look at 16 Bd3 in Game 17 and 16 c4 in Game
18.
16 ... Be4
A look at other lines:
a) After 16 ... Bf6, 17 Nh2! (intending Ng4) is slightly unpleasant for
Black. Following 17 ... Qa5 18 Bd2 Rcd8 19 Bf3 h6 20 Ng4 Bxg4 21 hxg4
Bg5 22 Bxg5 hxg5, V.Kramnik-V.Anand, Wijk aan Zee 2003, White can
secure a clear initiative after 23 Qc1! Rxe1+ 24 Qxe1 Kf8 25 Qc1 Re8 26
Rb1 when Black is quite uncomfortable since he experiences difficulties
defending the b7-pawn.
b) Black’s major alternative is 16 ... h6 which is fully playable for Black.
After 17 Nd2 (clearing the path for Bf3) 17 ... Na5 18 Bf3 Qd7 19 Ne4 Bf8
20 Ra2 Kh8 (eliminating all future Nf6+ tactics though 20 ... Nc4 21 Rae2 c6
22 a4 Rcd8 also looks close to equal) 21 Rae2 b6, S.Karjakin-Wang Yue,
Beijing 2013. The position looks balanced after 22 a4 Be6 (threat: ... Bb3) 23
Nd2 Bd6.
17 Qc1!?

A new move at the time, the idea of which is to play c3-c4, and then Qc3,
keeping the queen off the d-file and supporting the mobile pawn centre. One
is hard pressed to call it a distinguished refinement. Whenever I think I
discovered an original opening idea or plan in one of my games, I always
wonder if it really is an original idea, or is it perhaps my subconscious
regurgitating someone else’s idea, which I had seen in the distant past? My
sinking feeling is: It’s nearly always the latter, and then my ‘original’ idea is
in reality, inadvertent intellectual plagiarism, from the mind of some past
great player, into my own. Leko and Shirov have also experimented with the
complex 17 Nd2!? which is risky but probably still playable for White. 17 ...
Bxg2! 18 Bg4 Bh1 (The unplayed 18 ... f5!? looks interesting and play may
continue 19 Be2 (threatening a cheapo with c3-c4) 19 ... Na5. Now if White
continues with 20 c4, 20…Qxd4 21 Be3 Qf6 22 Kxg2 f4 23 Bxa7 Ra8
regains the piece but may not secure equality. White’s pressure continues
after 24 Bf3 Rxa7 25 Re6! (preparing to treble the major pieces on the e-file
with tempo) 25 ... Qf5 26 Qe2 Ra6 27 Re5 Qd7 28 Re1) 19 f3 Bh4! 20 Re4
f5 21 Kxh1 fxg4 22 hxg4, V.Anand-V.Kramnik, Wijk aan Zee 2010. GM
Mikhail Golubev gives White a slight plus here, while I think Black looks
fine after 22 ... Ne7 23 Bh2 Ng6 24 Qe2 Re7.
In Games 15 and 16, we will look at 17 Be3 and 17 Qa4 respectively.
17 ... Na5
Threatening a b3-fork.
18 Qe3
An untested try is 18 c4 and it may actually be an improvement. White
looks better after 18 ... Qf5 19 g4 Qf6 20 Ne5 h6 (not 20 ... Nb3? 21 Qe3
Nxa1 22 Qxe4 Nb3 23 g5 Qb6 24 Bd3 g6 25 c5 when Black is forced to
return the piece, with a probably lost position after 25 ... Bxc5 26 dxc5 Nxc5
27 Qc4) 21 Qe3! with a comfortable spatial edge.
18 ... Bf8
18 ... Bf5 is perhaps a superior alternative. After 19 Ne5 c5, Black’s gloom
begins to dissipate somewhat, since he achieves the ... c5 break before White
has a chance to activate with c3-c4. Now we see one of the benefits of
Black’s 15 ... Rac8. Following 20 Bf3 Qb3 (this is possible, since Black
controls the b1-square) 21 Qd2, K.Jones-A.Herrera, correspondence 2009.
White’s chances still appear slightly better to me after 21 ... cxd4 22 cxd4
Qa4 23 Qa2 Be6 24 d5 Qxf4 25 dxe6 f6 26 Nf7 Qc4.
 Exercise combination alert/critical decision: Can White get away with 19 c4?

Answer: White can indeed play 19 c4 and take over the initiative.
19 c4!
Kramnik may have been labouring under the illusion that he had control
over the c4-square. Perhaps he missed the tactical possibility of this strong
move.
19 ... Qd8
White’s point is that 19 ... Nxc4?? is unplayable, due to 20 Bxc4 Qxc4 21
Nd2 and now Black has no way to regain the lost piece. For example:
a) 21 ... Qd5 22 Nxe4 f5 23 Nc3 and White retains his extra piece.
b) 21 ... Qa6 22 Nxe4 f5 23 Qb3+ and once again, White escapes with his
extra piece.
c) 21 ... Qa4 22 Nxe4 f5 23 Nc3 wins for White.
d) 21 ... Qc2 22 Nxe4 f5 23 Qd2! and White absconds with the piece
again.
20 Ne5 Bf5 21 Qc3!
“More or less the maximum White can get from this opening,” said
Anand.
21 ... b6 22 Rad1
 Anand allows a chunk of his advantage slip. White continues to add
pressure after 22 c5! Qd5 23 Rad1 Rcd8 24 Bh5 g6 25 Be2 f6 26 Ng4 Qf7 27
cxb6 cxb6 28 d5 Bxg4 29 Bxg4 Rxe1+ 30 Qxe1 f5 31 Bf3 and the d-pawn
ties Black down.
22 ... Qf6
Kramnik returns the favour. Stronger was 22 ... f6! and if 23 Ng4 Bg6 24
Ne3 Bd6 25 Bxd6 cxd6 26 Bg4 Rc7 27 Bf3, White has only a slight edge.
23 Qg3
Both sides continue to underestimate the force of 23 c5! when 23 ... Bxh3?
isn’t playable, due to 24 Qd2! Bf5 25 Bg5 Qe6 26 Bh5! g6 27 Nxg6 with a
winning position for White.
23 ... Nc6
Principle: The cramped side should seek swaps. Let’s look at other
choices:
a) 23 ... c5 24 d5 Bd6 and according to GM Landa (ChessPro website), 25
Bh5 g6 26 Re3 is advantageous for White.
b) 23 ... Bd6 24 Bh5 g6 25 Bf3 c6 26 Re3 and the pressure mounts for
Black in the centre and kingside.
24 Ng4 Qg6 25 d5
25 Bxc7 “was not clear to me,” said Anand. Play may continue 25 ... Re4
26 d5 Nd4 27 Bf1 Bxg4 28 Rxe4 Qxe4 29 hxg4 Ne2+ 30 Bxe2 Qxe2 31 Rf1
Re8 32 Qf3 Qxc4 33 d6 Qb5 34 Rd1 Qd7 35 Qd when the passed d-pawn is a
source of concern to Black, who remains tied down.
25 ... Na5 26 Bxc7 Bc2
Golubev preferred 26 ... Bc5 intending ... h7-h5, which does look superior
to Kramnik’s move in the game.
27 Rc1! Nb3?!
Necessary was 27 ... Nb7, with Black a pawn down but having a playable
position.
 Exercise (combination alert): When great changes are about to occur,
the player who tends to profit is the one who anticipates the new
order of things first. So Kramnik decides to act first and worry about
consequences later. Black’s position was under heavy pressure and his
last move, although tempting, was a blunder. How did White exploit it?

Answer: Deflection. White’s minor pieces and queen are in a position to do great mischief to
Black’s king if his queen is diverted from her g6-defensive duties.
28 Rxc2!
Now Black’s position grows decidedly worse.
28 ... Qxc2 29 Nh6+
The f7-pawn falls with check and Black’s king begins to feel icy fingers of
apprehension tighten around his throat.
29 ... Kh8
The king is forced to tolerate the knight’s insolence with forced grace.
30 Nxf7+
Now the familiar contours of the black king’s old life recede rapidly.
30 ... Kg8 31 Nh6+
“Your Grace, I lament that such vulgar fisticuffery is necessary in the ‘re-
education’ program,” the knight informs Black’s king, who by now is beaten
black and blue.
31 ... Kh8 32 Nf7+ Kg8 33 Nh6+ Kh8 34 Be5!
No draw. White’s bishop conveys the intensity of his hate to the g7-pawn,
while White’s knight mulishly refuses to budge.
34 ... Qg6 35 Bg4!
Anand manages to find the continuation to sustain the initiative.

35…Rxc4?
A difficult position is that place where rationality dies and desperation is
born. We see a rare case of Kramnik losing the thread of the game. After this
move, White’s pieces run amok. Black’s last defensive hope lay in 35 ...
Rcd8 36 Qxb3 Qxh6 37 Be6 Bd6 38 Bxd6 Rxd6 39 Re4 when White still
dominates with two pawns for the exchange and a bind, but at least Black can
continue to play on. One thing to note is that Black is unable to return the
exchange with 39 ... Rexe6?? 40 dxe6 Rxe6 41 Qd3! (Black’s back rank is
fatally weak) 41 ... Kg8 42 Qd8+ Kf7 43 Qd7+ and the rook falls.
36 Qxb3 Rxe5
It’s time to resign after 36 ... Rc5 37 Nf5.
37 Rxe5 Rc1+ 38 Kh2 Bd6
 Exercise (critical decision): White must cover his rook. He can
play 39 f4 or 39 Qe3. Both should win, yet one of the moves
forces Black’s immediate resignation. Which one should he play?

39 f4
It’s dangerous to select move choices based on what fits our
preconceptions, yet we all do it. The natural move in this instance is the
wrong one, and prolongs the game.
Answer: Some might claim that frosting on a cake is redundant, but I certainly wouldn’t. More
decisive is 39 Qe3! when the queen seeks to fulfil a fundamental purpose: her own self-interest. This
move is an instant game-ender, due to Black’s weak back rank. After 39 ... Bxe5+ 40 Qxe5, oddly
enough, there is no way for Black to cover against the coming Qb8 back-rank check.
39 ... Bxe5 40 fxe5 gxh6 41 Qe3 Qb1 42 d6 Rh1+
This move is made for catharsis, but little over-the-board benefit, since
White’s king has nothing to fear.
43 Kg3 Re1 44 Qf4
The queen attempts to bring up the rather ticklish subject of the black
king’s upcoming execution date with as much tact as she can muster with the
threat of Qf8 mate!
44 ... Rf1 45 Bf3 1-0
Summary: The tabiya position after 15 ... Rac8 is one of the most critical ones for us to study in the
entire book. Black’s full equality is difficult to obtain against a knowledgeable opponent who knows
the theory.

Game 15
P.Leko-V.Kramnik
World Championship (Game 1), Brissago 2004

1 e4 e5 2 Nf3 Nf6
A World Championship match is that terrifying place where lifetime
reputations are either lifted or demolished in the span of just a few chess
games. To hold White to a draw in a World Championship Petroff’s game is
an achievement. To win with Black is an unexpected bonus.
3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 c4 Nb4 9 Be2 0-0
10 Nc3 Bf5 11 a3 Nxc3 12 bxc3 Nc6 13 Re1 Re8 14 cxd5 Qxd5 15 Bf4
Rac8 16 h3 Be4 17 Be3

Question: Didn’t White just lose a tempo, since

he moved his bishop twice to reach the e3-square?

Answer: Well, it depends on whether you think Black’s ... Rac8 is a useful move. I happen to think
so, since it enables a future ... Na5 and ... c5. In any case, White’s bishop, having ‘lured’ Black’s rook
to c8, now retreats to e3 to bolster defence of the d4-pawn and enable c3-c4. There is one other factor
to White’s decision: On the f4-square, White’s bishop is loose and vulnerable to future ... Qf5 tricks.
The bishop is more securely placed on the e3-square. This factor will become clearer in the notes to
White’s 18th move.
17 ... Na5
 Threatening Nb3, while preventing c3-c4.
18 c4!?

Question: Didn’t you just say “preventing c3-c4”?

Answer: I may have been a little hasty there. White can still play this move due to a tactical
element.
Let’s examine the positional treatment of mobilising the pawn centre with
18 Nd2:
a) 18 ... Bxg2? isn’t possible in this version, since White’s bishop is no longer on f4. For example,
after 19 c4 Qc6 20 d5 Qg6 21 Bh5! (this trick severs the black queen’s connection to her g2-bishop) 21
... Bf3+ 22 Bxg6 Bxd1 23 Bf5! (Black pieces hang simultaneously on d1 and c8) 23 ... Ba4 24 Bxc8
Rxc8 25 Bxa7! Bf6 26 Rac1 Bb2 27 Rb1 Bc3 28 Be3, White is up a full exchange.
b) 18…Bf5 19 c4 Qd7 20 Nf3 Bf6 21 Rc1 c5! 22 dxc5 (22 d5? Bb2 wins
the exchange) 22 ... Qxd1 23 Rcxd1 Nb3) going after White’s c5-pawn) 24
c6 Rxc6 25 Bxa7 Bc3 26 Rf1 Ra6 (this double attack regains the lost pawn
and ensures equality) 27 Be3 h6 (there is no rush as the a3-pawn isn’t going
anywhere) 28 c5 Rxa3 29 Bc4 Rc8 30 Nh4 Be6 31 Bxe6 fxe6 32 Rd7 Nxc5
33 Bxc5 Rxc5 34 Rxb7 and the players agreed to the draw in a few moves,
V.Ivanchuk-V.Kramnik, Wijk aan Zee 2008.
18 ... Nxc4!
The logical follow up. Houdini prefers 18 ... Qd6 when it’s difficult for
White to squeeze anything from the position. Following 19 Bd3 Bxd3 20
Qxd3 Qg6 21 Qc3 b6 22 Rac1 Bf6 23 Qb4 h6 24 Qa4 Qd3 25 Red1 Qb3 26
Qxb3 Nxb3 27 Rb1 Na5 28 Rdc1 Nc6 29 d5 Ne5 30 Nd4 c6 31 Nxc6 Nxc6,
Black equalized and the players agreed to a draw, J.Uusitalo-J.Weber,
correspondence 2011.
19 Bxc4 Qxc4 20 Nd2
This trick costs Black material, yet I believe he remains dynamically even.
20 ... Qd5 21 Nxe4 Qxe4
 Exercise (combination alert): How did White win material?

Answer: Double attack.

22 Bg5!
The crooked accountant bishop sells off Black’s assets and absconds with
the proceeds.
22 ... Qxe1+
Also possible is 22 ... Bxg5 23 Rxe4 Rxe4 24 Qb1 (with a double attack
on the e4-rook and also the b7-pawn) 24 ... Rce8 25 Qxb7 Bf6 26 Qxc7 Bxd4
27 Rf1 Re2 when White looks too tied down to make progress, despite his
material surplus.
23 Qxe1 Bxg5
 In the aftermath, Black has rook, bishop, pawn and a rock-solid formation
for the queen – a pretty good deal although White soon wins another pawn
with a tactic.

Question: How would you assess this position?

Answer: Dead even, despite Houdini’s ‘+=’ evaluation. Black’s position feels solid and I can’t
imagine him losing, barring a blunder.
24 Qa5!?
An ambitious attempt. White’s apparent initiative has a distinctively
transitory feel to it. Leko simultaneously attacks the g5-bishop and the a7-
pawn. The trouble with this move is that it allows Black a dangerous passed
c-pawn. Far safer is 24 Qb4 b6 25 Qa4 Bf6 26 Rd1 a5 27 Qc6 Rcd8 28 g4 h6
29 Kg2 Re7 30 d5 Rd6 with a very likely draw, since Black’s fortress looks
impenetrable.
24 ... Bf6 25 Qxa7 c5
This pin trick insures a dangerous passed c-pawn.
26 Qxb7 Bxd4 27 Ra2
Covering the second rank against a future ... Re2.
27 ... c4
The initiative changed hands, but without altering the overall evaluation.
Black’s passed c-pawn is more dangerous than White’s a-pawn.
28 Re2
A clever trick based on removing the protection of Black’s c8-rook.
28 ... Red8 29 a4
Leko continues to play for the win. Instead, 29 Rd2 (threat: Rxd4) 29 ...
Bf6 30 Rxd8+ Rxd8 31 Qc7 c3 32 a4 h6 33 a5 Rd2 34 a6 Be5! (Black is
willing to pay any price to get the c-pawn to c2) 35 Qxe5 c2 36 Qe8+ Kh7 37
Qe4+ g6 38 Qc4 Rd1+ 39 Kh2 c1Q 40 Qxf7+ and the game ends with
perpetual check.
29 ... c3
The c-pawn is deeper and therefore more dangerous than its a-file
counterpart.
30 Qe4
Seizing control over the c2-square.
30 ... Bb6
Halting the a-pawn’s climb.
31 Qc2 g6
Finally, Black eliminates back rank issues.
32 Qb3 Rd6
Alternatively, 32 ... Ba5 33 Rc2 Rb8 34 Qc4 Rb4 35 Qa6 Rb1+ 36 Kh2
Bc7+ 37 g3 Rb2! 38 Rxc3 Rxf2+ 39 Kg1 Ra2 (threatening a Rd1 back rank
cheapo) 40 Rd3 Rb8 41 Rd1 Rbb2 42 Qc8+ Kg7 43 Qxc7 Rg2+ 44 Kf1
Raf2+ leads to perpetual check.
33 Rc2 Ba5 34 g4
Offering his king more air.

Question: Why not challenge Black’s loose a5-bishop with 34 Qb5?

 Answer: White is unable to make the slightest progress in this line after 34 ... Rd1+ 35 Kh2 Rd2!
and White is forced back with 36 Qb3.
34 ... Rd2
It becomes obvious that White doesn’t stand even an iota better.
35 Kg2 Rcd8 36 Rxc3?!
I’m not so certain that divergence from the norm is wise here. At last, the
nettlesome c-pawn is eliminated but at a high cost. Only Black has chances in
the two rooks versus queen and pawn ending. The safer path to draw is to
maintain the c2-blockade with 36 Rc1.
 36 ... Bxc3 37 Qxc3 R2d5 38 Qc6
The rash 38 a5 allows 38 ... Ra8 and the pawn is doomed.
38 ... Ra5
The a-pawn is doomed.
39 Kg3 Rda8 40 h4 R5a6
GM Raymond Keene points out that Kramnik should play 40 ... h6! to
meet 41 h5 with 41 ... g5!.
41 Qc1 Ra5 42 Qh6! Rxa4 43 h5
White’s best chance is to somehow expose Black’s king and deliver
perpetual check, but this is obviously not so easy to achieve.
43 ... R4a5 44 Qf4?
I feel like White’s best drawing hope lies in the line 44 f4! Ra1 45 f5!
R8a3+ 46 Kf4 Rf1+ 47 Kg5 Rh1 48 Kf6 Ra6+ 49 Ke7 Re1+ 50 Kd7 Re4 51
hxg6 hxg6 52 Qh3 Kg7 53 Qc3+ f6 54 g5 gxf5 55 Qh3 fxg5 56 Qxf5 Rd4+
57 Ke7 Kh6 58 Qf8+ Kh7 59 Qf7+ with perpetual.
 Exercise (planning): White’s last move was a strategic
blunder. Construct a clear plan for Black to play for the win:

Answer: Fix White’s pawns and then go after the f-pawn.

44 ... g5!
It is in Black’s favour to retain pawns on the board, since in doing so his
king is kept safer from perpetual check attempts.
45 Qf6 h6!
Only with this tactic does Black manage to stabilize his kingside pawns.
46 f3
Not 46 Qxh6?? R8a6 as the queen finds herself caught in a web of her own
greed.
46 ... R5a6 47 Qc3 Ra4
Preventing f3-f4. Black needs to place a rook on the f4- or f6-square, tuck
his king on the g7-square, and then find a way for his other rook to attack
White’s f-pawn, all without walking into perpetual check. Not an easy feat!
48 Qc6 R8a6 49 Qe8+ Kg7 50 Qb5 R4a5 51 Qb4 Rd5 52 Qb3 Rad6 53
Qc4?!
Instead, 53 Qb2+ Rf6 54 Qc3 Rd8 55 f4!? gxf4+ 56 Kf3 Rdd6 57 Qc2
Rde6 58 Qd2 Re3+ 59 Kf2 Rg3 60 Qd7 f3 61 Kxg3! f2 62 g5! f1Q (after 62
... hxg5 63 h6+ Kxh6 64 Qh3+ Kg6 65 Qf1, White should hold) 63 gxf6+
Qxf6 64 Qd1 won’t be so easy for White to convert.
53 ... Rd3!

When you are slightly better and your opponent has no way to improve his
position, making him await your intent indefinitely can be difficult to
withstand psychologically.

Exercise (critical decision): It’s obvious that White’s position hangs

by the barest of threads. He can play 54 Qb5, or 54 Kf2. One line offers
drawing chances, while the other loses. Which one would you play?

54 Kf2?
This move allows the f3-pawn to fall, and with it, all of White’s hopes
nosedive as well.
Answer: 54 Qb5! puts up maximum resistance after 54 ... Re3 55 Kf2 Rdd3 56 f4! gxf4 57 Qc5
Rf3+ 58 Kg2 Rg3+ 59 Kf2 and the win isn’t so easy for Black since 59 ... Rxg4? allows a perpetual
check after 60 Qe5+ Kf8 61 Qc5+ Kg7 62 Qe5+.
54 ... Ra3!
Threat: ... Rf6.
55 Qc5 Ra2+ 56 Kg3 Rf6
 Kramnik reaches his dream setup. Now White’s ownership-in-perpetuity
of the f3-pawn comes into question. Black’s king is completely safe and
White’s queen is unable to prevent the a2-rook from launching a secondary
attack on the f3-pawn.
57 Qb4 Raa6!
Intending ... Rff4, followed by ... Raf6. A constellation of malevolent
forces array themselves against White’s lonely f-pawn.
58 Kg2
Also, 58 Qc3 allows 58 ... Ra4 and ... Raf4 follows.
58 ... Rf4 59 Qb2+ Raf6
 The f3-pawn falls and White has no chance to save the game.
60 Qe5 Rxf3 61 Qa1 Rf1 62 Qc3
After 62 Qxf1 Rxf1 63 Kxf1 Kf6 64 Ke2 Ke5 65 Ke3 f5 66 gxf5 Kxf5 67
Kf3 g4+ 68 Kg3 Kg5, the h5-pawn falls as well.
62 ... R1f2+! 63 Kg3
Also insufficient is 63 Kg1 R2f4 64 Qc8 (64 Qg3 Rf1+ 65 Kg2 R6f2+ 66
Qxf2 Rxf2+ 67 Kxf2 Kf6 leads to a lost king and pawn ending) 64 ... Re6 65
Qc3+ Kh7 66 Qd3+ Ree4 and the g4-pawn is next on the rook’s hit list.
63 ... R2f3+
 The king and pawn ending is an easy win for Black.
64 Qxf3 Rxf3+ 65 Kxf3 Kf6 0-1
Summary: After 18 c4, we should deliberately ‘fall’ for White’s combination with 18 ... Nxc4!. In
the aftermath, I don’t think Black stands even an iota worse, since we easily have enough to set up a
fortress, or as we saw in this game, even win if White overestimates his chances.

Game 16
P.Svidler-V.Anand
Linares-Morelia 2007

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 c4 Nb4 9 Be2 0-0 10 a3 Nc6
11 cxd5
Remember, we can enter the main line position via this move order as
well.
11 ... Qxd5 12 Nc3 Nxc3 13 bxc3 Bf5 14 Re1 Rfe8 15 Bf4 Rac8 16 h3
Be4 17 Qa4
 Question: What are White’s ideas behind his last move?

Answer:
1. White gets his queen off the d-file, away from a future …Rd8.
2. White seizes control over the c4-square, disallowing Black the ... Na5
blockade plan which we saw in the previous game.
3. White enables c3-c4 and d4-d5, chasing away Black’s c6-knight while
threatening Qxa7.
4. White’s light-squared bishop may later be activated on the c4-square, or
White may later play Bb5, applying pressure to Black’s knight.
17 ... Qf5
The queen moves out of the path of White’s coming c3-c4, while she gains
a tempo on the f4-bishop. Also played is 17 ... a6, denying White’s bishop of
a Bb5-pin option. After 18 Bg3 Qa5 19 Qb3 Bd5 20 Bc4 (20 c4 Bxf3 21
Bxf3 Nxd4 22 Qxb7 Bf6 23 Rxe8+ Rxe8 24 Rd1 g6 25 Qxc7 Qxc7 26 Bxc7
Rc8 27 Bg3 Rxc4! 28 Bb7 a5 29 Ba6 Rc3 30 Rd3 Rc2 and Black’s powerful
knight negates White’s hopes for an advantage based on his bishop pair) 20
... Bxc4 21 Qxc4 Bd6 22 Rab1 Rxe1+ 23 Rxe1 h6! (23 ... Qxa3?? walks into
24 Ng5 Rf8 25 Nxf7! Rxf7 26 Bxd6 Qxd6 27 Re8+ Qf8 28 Rxf8+ Kxf8 29
d5 Nd8 30 Qb4+ Re7 31 c4 and White can try for a win), chances are even,
E.Kondratyev-D.Minelga, correspondence 2008.
18 Bg3 Bc2
Harassing White’s queen. This is considered slightly more accurate than
18 ... a6 19 Nd2 Bc2 20 Qc4 (White threatens Bg4, which induces Black to
weaken on his next move) 20 ... h5 21 Qa2 (threat: Bxh5) 21 ... h4 22 Bg4
Qg6 23 Bh2 Bf5 24 Bxf5 Qxf5 25 Nf3 Qd7, J.Stephan-A.Ponomarev,
correspondence 2008. Black is under some pressure after 26 Rab1 b6 27 a4
Bf6 28 Qc4 a5 29 Be5 Bxe5 30 dxe5 Nd8 31 Re4 when the overextended h4-
pawn falls.
19 Qb5

Question: I’m not sure White gets anything from the ending. Can White try 19 Qc4?

Answer: Black looks equal after 19 ... Na5 20 Qa2 Bb3 21 Qd2 Qc2 (every swap helps Black, who
owns the more compact pawn structure) 22 Bd3 Qxd2 23 Nxd2 Be6. With queens off the board,
White’s central pawns may prove to be more of a liability than an asset.
19 ... Qxb5 20 Bxb5
The bishop’s gaze rests angrily upon Black’s knight.
20 ... a6!
In a chess game, we often take on the guise of a crafty merchant
attempting to out-haggle the customer, who is our equally crafty opponent. In
this case, we dare White to chop the c6-knight, hoping we get enough
compensation for the damage to our structure through the bishop pair and
enhanced control over the light squares.
21 Bxc6
This is more consistent than 21 Re2 Bb3 (21 ... axb5 22 Rxc2 Na5 23 a4
bxa4 24 Rxa4 b6 25 c4 Bf6 also looks fine for Black) 22 Bd3 Bd6 23 Rxe8+
Rxe8 24 Bxd6 cxd6 25 Rb1 Na5 26 Nd2 Ba2 27 Rb2 Be6 28 c4 Kf8 29 d5
Bc8 30 Rb1 h6 when the ending is even since White’s extra space balances
out with Black’s slightly superior structure, R.Scherer-H.Van Unen,
correspondence 2011.
21 ... bxc6

Question: Does our bishop pair provide us with

full compensation for our inferior structure?

Answer: When players follow such mutually contradictory programmes, it feels as if only one can
come out on top. But this isn’t always the case. Black’s bishop pair, enhanced light square control, and
future ... c6-c5 undoubling prospects should provide Black dynamic equality. But a cautionary warning:
we achieve equality only after undergoing a bit of defensive suffering. This means that the position may
be ‘+=’ now, but with correct play, will eventually turn equal – if that makes any sense to you!
22 Re5
Clamping down on Black’s potential ... c6-c5 break, while thinking about
transferring the rook to the a5-square. Let’s look at the alternatives:
 a) 22 Ra2 Ba4 23 c4 Bb3 (Black can also try 23 ... Bf6! 24 Rxe8+ Rxe8 25
Rd2 Bb3 26 c5 Bd5 27 Bxc7 h5 when Black’s bishop pair and light square
control give him full compensation for the pawn) 24 Rae2 Kf8 25 c5 (White
gives up rights to the d5-square, while denying Black a future ... c6-c5 break
and fixing the c7-pawn as a perpetual target) 25 ... Bd5 26 Ne5 (threat: Nd7+,
followed by Rxe7) 26 ... Rcd8 27 Re3 a5 28 h4 f6 29 Nd3 Bf7 (after 29 ...
Bc4 30 Bxc7 Rxd4 31 Nf4 Kf7 32 Ne6 Bxe6 33 Rxe6 Rd1! 34 Rxe7+ Rxe7
35 Rxd1 Rxc7 36 Rd6 h5 37 Kf1 Kg6 38 Ke2 Kf5, Black’s super active king
should make up for his passive rook) 30 Bxc7 Rxd4 31 Nf4 occurred in
A.Grischuk-A.Volokitin, Foros 2006. Black should continue with 31 ... f5! 32
Ne6+ Bxe6 33 Rxe6 Re4! 34 R1xe4 fxe4 35 Bxa5 (35 Rxc6?? Rc8 walks
into a deadly pin) 35 ... Bxc5 36 Rxc6 Bxa3 with a near certain draw.
b) 22 Ne5 Ba4 23 Nd3 (hoping to stop ... c6-c5) 23 ... Kf8 24 Nc5 Bxc5 25
dxc5 a5 26 Rxe8+ Kxe8 27 Rb1 Bb5 28 Re1+ Kd7 and the game is dead
even, E.Tsygankov-V.Yartsev, correspondence 2010.
22 ... f6
Black can also speculate with 22 ... c5!? 23 Rae1 Kf8 24 dxc5 f6 25 Rd5
Bb3 26 Rd7 Ba4 27 Rd4 (27 Rxc7 Rxc7 28 Bxc7 Bxc5 29 Ra1 and White is
much too passively placed to even think about converting his extra pawn) 27
... Bc6 28 Nd2 Bxc5 29 Rxe8+ Bxe8 30 Rd5 Bxa3 31 Ra5 Bb2 32 Ne4 Bc6
when White stood worse and is the one fighting for the draw, A.Shirov-
B.Gelfand, Moscow 2007.
 The text move seizes control over the e5-square and theory ends here.

Question: How am I going to remember all of this!?

Answer: It’s easy for us to forget our lines, especially in Petroff’s, where one line looks much the
same as the next. When we mis-remember a variation we studied diligently at home, it’s like quoting a
person out of context. The only way to entrench these complexities into our heads is to just go over the
lines again and again, mixed with playing them – especially in online blitz, where you can test them
out.
23 Ra5
The rook obediently veers left, as if in response to a command only he
hears.
23 ... Bd3
Covering the a6-pawn.
24 Ne1 Bb5
Preparing an immediate …c6-c5 in anticipation of 25 Nc2, but the idea is
soon nixed in the bud.
25 a4
After 25 Rc1 with the idea of c3-c4, Black can simply ignore it with 25 ...
Bd6!. All of White’s coercive powers fail to intimidate the light-squared
bishop. Thus 26 c4 is met with 26 ... Bxc4! which is possible since White’s
e1-knight hangs if White takes the bishop. White’s c1-rook shakes his head in
grim deprecation of the presumptuous black bishop’s open lack of respect for
authority.
25 ... Be2
The pesky bishop refuses to leave the premises of his diagonal.
26 Nc2 c5?!

Black’s original postulations begin to squirm and wiggle a bit, and the
simple first step in the plan is followed by, in my opinion, an unnecessarily
elaborate corollary. Black must weigh the following factors:
1. He risks all by sacrificing a pawn and fighting back from a position of
slight inferiority.
2. He risks all by not fighting back in a position of slight inferiority.
So Anand decides to go with the former choice, to spend his wealth, rather
than horde it. Defending a slightly inferior position is dreary work, similar to
the unlucky soldier, assigned to guard duty on a cold, rain-soaked night. A
hidden anxiety presses down upon Anand’s mind, about the long-term
weakness of his structure, so he gives away a pawn in the hopes of activating
his pieces. Still, I’m not sure Black should give away this pawn. I think he
gains full equality with 26 ... Rb8! (offering the c7-pawn instead) 27 Bxc7
Rb2 28 Ne3 Bf8 when Black’s active pieces, bishop pair and seventh rank
rook insure full compensation for the pawn.
27 dxc5 Kf7 28 c6
 White has excellent winning chances in the line 28 Re1! Bc4 29 Ne3 Bd8
30 c6 Re6 31 Rd1 Rxc6 32 Rd7+ Kf8 33 Nxc4 Rxc4 34 Rxa6 Rxc3 35 a5
Be7 36 Re6 Bc5 (threat: ... Rxg3) 37 Kh2 Ra3 38 Rc6 Be7 39 Rdxc7 Rxc7
40 Bxc7 with an extra pawn and reasonably good chances to convert to a win.
28 ... Red8
Planning to invade the seventh rank.
29 Ne3
Once again Svidler missed the correct plan: 29 Re1! and if 29 ... Rd2 30
Ne3 Bf8 31 Nd5, the c7-pawn falls.
29 ... Rd2 30 Rb1
Once again, 30 Re1! is White’s best plan.
30 ... Bd6! 31 Bxd6
Black has good saving chances after 31 Kh2 g5 32 Rb7 Ke6 33 Bxd6
Rxd6 34 Ra7 h5 35 Nf5 Rd2.
31 ... Rxd6 32 Nf5?!
This is a tempting tactic, but I think White would have been better off
hanging on to the pawn with 32 Re1! Bd3 33 Nd5 Bc4 34 Nb4 Ra8 when
Black remained under pressure.

After this move, White’s initiative continues, but by diminishing margins.

Exercise (critical decision/combination alert): Black would love to

 play 32 ... Rxc6, but then comes 33 Nd4, forking the rook and the
bishop. Can Black play into this line anyway and get away with it?

Answer: Yes. Black can deliberately allow the fork due to a combination on his next move.
32 ... Rxc6!
“The law is irrelevant to my needs,” declares the lawless rook. His
philosophy is to take without asking, since if he asks, he will be likely to hear
a resounding “No!” to his chagrin. I’m wondering if Svidler discounted this
move in his analysis.
33 Nd4
This fork looks like doom for Black, but he has a strong rejoinder.
33 ... Rb6!
“It’s not easy being me, yet I manage to make it look so effortless,” boasts
the conceited rook. The curtain which separated Black from his objective
suddenly parts. After this trick, White’s initiative dwindles to zero.
34 Re1
After 34 Rxb6 cxb6, White doesn’t have time to take the e2-bishop, since
his own Ra5 hangs after the trade.
34 ... Bd3 35 Rc5 Rd6 36 Re3 Bb1
At long last, the bishop is allowed to retire from his tedious defensive
duties of covering the a6-pawn.
37 Nc6
Threat: Na7, followed by Rxc7+.
37 ... Re8 38 Nd4 Rc8
Also equal is 38 ... Rxe3 39 fxe3 Rd7 40 Rc6 Bd3.
39 Re2 Bd3 40 Re1 Rd7 41 f3 ½-½
Summary: Don’t fear the ending after 20 ... a6! 21 Bxc6. Our bishop pair and light square
domination make up for the damage to our queenside pawn structure.

Game 17
P.Leko-B.Gelfand
Tal Memorial, Moscow 2009

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 c4 Nb4 9 Be2 0-0 10 Nc3
Bf5 11 a3 Nxc3 12 bxc3 Nc6 13 Re1 Re8 14 cxd5 Qxd5 15 Bf4 Rac8 16 Bd3
 Leko’s specialty. He has also used this line to beat Anand as well as draw
against Bacrot and Kramnik.

Question: Why is White willing to swap pieces, which surely helps Black?

Answer: You are correct. White avoids h2-h3 and instead, offers to swap bishops, which violates
the principle: The side with extra space should avoid trades. However, White does remove one of
Black’s best pieces, and in doing so, seizes control over the b1-square for a rook thus paving the way to
push forward his c- and d-pawns. 16th move alternatives:
a) 16 Nd2 (White clears the path for Bf3 before Black gets a chance to
play ... Be4) 16 ... Na5 17 Bf3 (17 Nf1 intending Ne3, is well met with 17 ...
Qb3 with an equal position) 17 ... Qd7 18 Nf1 Bd6 19 Rxe8+ Rxe8 20 Bxd6
cxd6 21 Ne3 Be4 22 h3 and Black solved his opening problems and the
players agreed to a draw, M.Adams-V.Kramnik, Wijk aan Zee 2005.
b) 16 Qc1.

Question: What is the point of this artificial-looking move?

Answer: I agree it’s a convoluted idea. White’s queen protects the f4-bishop, so that he can play c3-
c4, and if Black plays ... Qe4, then he doesn’t gain a tempo on the f4-bishop:
b1) Kramnik experienced one of the biggest opening disasters in his chess
career after 16 ... Na5?! 17 c4 Qe4?? (necessary was 17 ... Qd8 18 Qc3 when
White had only an edge)
 Exercise (combination alert): White to play and win decisive material:

Answer: Double attack. All of a sudden, Black is lost after 18 Bd1! Qd3 19 Re3! Qxc4 20 Re5!
with a killing double attack. Kramnik resigns, since Anand wins a full piece, V.Anand-V.Kramnik,
Sofia 2005.
b2) Black equalizes with 16 ... Bd6! 17 Be3 (17 c4 Qe4 18 Be3 was
H.Prokopp-M.Dos Santos, correspondence 2006 and Black can equalize by
removing queens from the board with 18 ... Qc2) 17 ... Qe4 18 Ra2 (covering
against ... Qc2) 18 ... Na5 19 c4 c5 20 Qc3 b6 (20 ... cxd4 21 Nxd4 b6 22
Nxf5 Qxf5 also looks even. White’s weak pawns are offset by his bishop
pair) 21 d5 Bg6 22 h3 Qf5 23 Bf1 Re7 24 Rd2 Rce8 and the game is even
since White’s protected passed d5-pawn is firmly blockaded. After 25 Nh4
Qf6!? 26 Qxf6 gxf6 27 g3 Nb3 28 Ra2 Nd4 29 Rd1 Rxe3! 30 fxe3 Rxe3 31
Nxg6 Rxg3+ 32 Rg2 fxg6 (Black has more than enough compensation for the
exchange) 16 Qa4 a6 (ending Qb5 or Bb5 ideas) 17 Ne5 Nxe5 18 Bxe5,
A.Brkic-Z.Jovanovic, Zagreb 2004, White threatens Bf3, so simplest is 18 ...
c6 with a balanced position.
16 ... Qd7!
The most accurate continuation. Black prevents c3-c4 since it will render
the d4-pawn en prise.
 Question: What is wrong with immediately swapping on bishops on d3?

Answer: I think that is also playable. But why give White’s queen a free lift to the d3-square? Play
may go 16 ... Bxd3 17 Qxd3 h6 18 c4 Qd7 19 d5 Bf6 20 Rad1 Na5 21 Be5 Bxe5 22 Nxe5 Qd6 23 Qf5
Qf6 24 Qd7 Qe7 25 Qa4 b6 26 Re3 f6 27 Ng4 Qf7 28 h3 Rxe3 29 Nxe3 f5 30 c5 and Black still
struggles to equalize, L.Hyldkrog-L.Lamarche Rodriguez, correspondence 2003.
17 Rb1
White hopes to induce a weakening of the light squares. Let’s examine
other options:
a) 17 Bxf5 Qxf5 18 Bg3 Bd6 19 Qb3 b6 20 c4 Qf6 21 d5 Nd4 22 Nxd4
Qxd4 and White had nothing from the opening, V.Ivanchuk-B.Gelfand,
Mallorca 2004.
b) 17 c4? fails tactically to 17 ... Nxd4! 18 Nxd4 Bxd3! 19 Qxd3 c5 when
Black regains the piece with an extra pawn as interest.
17 ... b6 18 d5!?

The pawn can’t be taken, and if the knight moves, then Black is vulnerable
to Bb5 and Ne5 ideas. White can also try a speculative exchange sacrifice
with 18 Bb5 Bxb1:
a) 19 Qxb1 Bf6 (covering against Ne5) 20 Rd1 (threat: d4-d5)
A.Timofeev-Wang Yue, Ningbo 2010. Black equalizes with 20 ... Qd5 21
Bg3 (White threatens c3-c4). Now, simplest is 21 ... Ne7! 22 Bxe8 Rxe8 23
Re1 Qc6! 24 Qb3 b5 when Black successfully unravels.
 b) 19 d5 is met with 19 ... Rcd8! 20 Nd4 Qxd5 21 Bxc6 Qxc6! 22 Nxc6 Rxd1 23 Nxe7+ Rxe7 24
Rxd1 when things look bad for Black, since White threatens both a back rank mate and also the
hanging b1-bishop, but Black has an all-purpose answer with 24 ... Bd3! which plays on White’s own
weak back rank. Following 25 f3 Rd7 26 Kf2 c6 27 Ke3 Bf5 28 Rxd7 Bxd7 29 Bb8 a6 30 Ba7 b5 31
Kd4, a drawn ending is reached.
18 ... Bxd3
18 ... Na5?? hangs a piece to 19 Ne5 when Black’s queen is disconnected
from her f5-bishop.
19 Qxd3
19 dxc6? Qf5 leaves Black with both the bishop pair and superior
structure.
19 ... Bxa3
Do we not all occasionally yield to avarice? This pawn grab is a novelty
by Kramnik. Instead, 19 ... Bf6 20 c4 Ne7 21 Rbd1 Rcd8 22 h3 Ng6 23 Bg3
h6 24 Qb3 Re7 25 a4 Rde8 26 Rxe7 (26 Qb5 Rxe1+ 27 Nxe1 Qxb5 28 cxb5
Be5 is fine for Black) 26 ... Nxe7 27 Re1 c6 (at last, Black strikes back at
White’s d5-point) 28 Be5 cxd5! 29 Bxf6 dxc4 30 Qxc4 gxf6, P.Leko-
V.Kramnik, Dortmund 2007, has Black standing at least equal though his
extra pawn is well….extra!
20 Ng5
The rainy season begins over Black’s kingside. Not very subtle, but it
induces dark square weaknesses in Black’s camp. White’s queen gets the
feeling that the headstrong black king is not a person who is swayed by either
kindness or logic so she decides to apply the lash, ordering her knight to do
her dirty work.
20 ... g6 21 Ne4
Threatening a mega-fork on f6.
21 ... Qf5!
The queen’s delightful wind chime tinkle of a laugh betrays her inner
ruthlessness. “An improvement upon Kramnik’s 21 ... Be7. In fact this move
was tested in a correspondence game. We’ll not know what Leko prepared in
this line, unless Black is curious in the future,” wrote GM Victor
Mikhalevski. After 21 ... Be7 22 Rbd1 (Mikhalevski speculates that Leko had
prepared 22 Qb5 as an attempted improvement on the Jakovenko-Kramnik
game) 22 ... f5! 23 d6 cxd6 24 Qd5+ Kh8 25 Nxd6 Bxd6 26 Bxd6,
D.Jakovenko-V.Kramnik, Moscow 2007, only Black has winning chances
after 26 ... h5!. I don’t believe White’s piece activity fully compensates his
missing pawn.
22 Bxc7 Rxc7!
Gelfand’s move is an improvement over 22 ... Ne5 23 Bxe5 Rxe5 24 f3
which is slightly better for White, since his powerful d5-pawn is a cause for
concern for Black, R.Markus-J.Deforel, correspondence 2008.

A plan can be innocent of ornamentation, yet still be effective. This

theoretical novelty is now the main line. Gelfand ignores White’s threats with
disarming simplicity and gets full compensation for the exchange.

Exercise (combination alert): White can win the exchange

(but not the advantage) with his next move. How?

Answer: Discovered attack.

23 Nf6+
The knight decides to circumvent the law to mete out his own brand of
justice.
23 ... Qxf6 24 Rxe8+
The rook eliminates the only witness to the crime.
24 ... Kg7 25 dxc6 Bc5 26 Rb2 Rxc6
 The once murky position focuses into greater detail. White won the
exchange, but in a sense his combination misfired, the way a boxer swings at
the opponent’s jaw-line, missing by a centimetre, only to connect with thin
air. Black’s compensation for the exchange:
1. He owns a passed a-pawn.
2. Black controls the dark squares.
3. Black’s ironclad defensive setup offers his opponent no real targets.
Conclusion: White’s goal of a full point is so close, yet so unavailable. For
the remainder of the game, it feels like White plays with only a half-formed
intent to win.
27 Qd8
If White refuses to swap queens, Black may even take over the initiative
with ... Rd6 next.
27 ... Qxd8 28 Rxd8 a5
Threat: ... Bb4! The a-pawn completely ties White down and negates all
winning chances for him.
29 Kf1
Covering his back rank. 29 Ra8? allows 29 ... Bb4 when only White can
lose.
29 ... a4 30 Ke2 a3
The a-pawn impounds the square it coveted for a long time.
 31 Ra2

“When I heal, I will kill you all,” threatens the humiliated a2-rook to his
tormentors. The passive rook requires long convalescence before full
recovery.
31 ... b5!?
The b-pawn weaves its way up the board with feline competency and
delicacy. The goal is to swap itself for White’s c3-pawn.

Question: Why is Black taking action?

Answer: In any potential confrontation, we must remember that a feasible option is to do absolutely
nothing and refuse to respond to the provocation. I don’t really see how White makes progress if Black
just shuffles back and forth, doing nothing. Instead, Gelfand decides to pick one of his limited options
to take action. He may have worried about a plan where White returns the exchange to achieve a
favourable rook and pawn ending, mainly since White’s king is already on the queenside. But I don’t
see how White achieves this. Let’s take a look: 31 ... h5 32 Kd3 Rc7 33 Kc2 Ra7 34 Kb3 Kf6 35 Re2
Kg7 36 g3 Kf6 37 f4 Kg7 38 h3 Kf6 39 Rd5 Kg7 40 Ka2 Kf6 41 g4 hxg4 42 hxg4 Kg7 43 g5 Ra8! (43
... Ra4? is met with 44 Rd8! with the threat of Ree8) 44 Rd7 Rc8 45 Kb3 Ra8 when White’s threat to
make progress carries no great conviction and Black’s pieces roam where they please.
32 Rb8
Leko attempts to halt the ... b4 pawn exchange idea. However, Gelfand
found a way around the barrier. Instead, 32 Kd3 b4 33 cxb4 Bxb4 is equal.
32 ... Bd6!
 A pair of pawns gets swapped after all, despite the white rook’s series of
surly complaints.
33 Rxb5 Rxc3 34 Kd2 Rc6 35 g3 h5
Preparing to whittle down the pawns further with ... h4.
36 Kd1 h4 37 Ra5
37 gxh4!? Kh6 38 Ra5 Rc4 39 h5 Rd4+ 40 Kc2 gxh5 41 Ra1 f6 is still
about even.
37 ... hxg3 38 hxg3 Rb6 39 Rc2 Rb1+ 40 Ke2 Rb2!?
Another way to ensure the draw is to essentially do nothing with 40 ... Rb3
41 Rd5 Bf8 42 Rd8 Be7 43 Re8 Bf6 44 Kd1 Bb2 45 Ra8 Rb7 and White
can’t do a thing.
41 Kd3
41 Rxb2 axb2 42 Rb5 Be5 43 f4 Bf6 44 g4 Kf8 White is too tied down to
make progress. Still, this line may be White’s only prayer at a win.
41 ... Bb4! 42 Ra4 Be1!

The bishop’s most pious hopes are realized. “My divine authority cannot
be invalidated by a few disgruntled ‘nay’ votes,” he declares, while eyeing
white rooks and king, meaningfully. A draw is certain if White drops a
kingside pawn.
43 Rxa3 Rxc2 44 Kxc2 Bxf2 ½-½
Summary: A model game by Gelfand, which pretty much neutralizes White’s ambitions in the 16
Bd3 variation.

Game 18
G.Kasparov-A.Karpov
X3D Match (Game 2), New York (rapid) 2002

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 c4

In the next game, we will look at the 8 Re1 line.
8 ... Nb4 9 Be2 0-0 10 Nc3 Bf5 11 a3 Nxc3 12 bxc3 Nc6 13 cxd5 Qxd5
14 Re1 Rfe8 15 Bf4 Rac8 16 c4

No goofing around or prep necessary. White simply pushes his c-pawn

forward at the earliest given opportunity. Other choices are:
a) 16 Qc1 Qd7 17 Qb2 Bd6 18 Be3 Na5 19 c4 b6 20 Rad1, V.Ivanchuk-
Wang Hao, Beijing 2013 when the game looks balanced after 20 ... c6.
b) 16 Qa4 Bd7 (Black immediately challenges White’s queen position,
while bolstering support of the c6-knight, in case of Bb5 ideas) 17 Qc2 (17
Rab1 Qf5 18 Bg3 Bf8 (the b7-pawn is tactically covered) 19 Rb5, A.Shirov-
B.Gelfand, Monte Carlo 2003 and Black can play 19 ... Nxd4! 20 Nxd4 Bxb5
21 Qxb5 Qxb5 22 Nxb5 c6 23 Nd4 Bxa3 when Black’s rook and two healthy
pawns are at least worth White’s two minor pieces) 17 ... Qf5 18 Qxf5 Bxf5
19 Bb5 Bd7 (19 ... a6!? reminiscent of last game, also looks playable as 20
Bxc6 bxc6 21 Re5 Bd3 22 Rae1 Kf8 23 Ra5 Bd6 looks okay for Black) 20 d5
Ne5 21 Bxd7 Nxd7 22 Bxc7 Rxc7 23 d6 Rxc3 24 dxe7 f6 25 Rad1 Rc7 26
Nd4 Ne5 27 f4 Nc6 28 Nxc6 bxc6 29 Rd6 c5 occurred in V.Anand-
V.Kramnik, Mexico City 2007. The e7-pawn is doomed and only Black can
win.
16 ... Qe4 17 Be3 Bf6
Black is ambitious, going after the newly weakened d4-pawn with this
sharp move. White’s 16 c4 line went out of favour after Kramnik played 17
... Qc2 (the queen exchange took most of the fun out of White’s position) 18
d5 Na5 19 Nd4 Qxd1 20 Rexd1 Bd7 21 Bd2 Bf6 22 Bxa5 Bxd4 23 Rxd4
Rxe2 when Black equalized effortlessly and Kramnik scored a quick draw
with Black, an important feat in a World Championship match, P.Leko-
V.Kramnik, World Championships (Game 3), Brissago 2004.
18 Rc1
Kasparov deviates from his game against Adams, which goes 18 Ra2 b6
19 h3 Na5 20 g4 Bg6 21 g5!? (White will stand worse if his initiative dies)
21 ... Be7 22 Ne5 Bd6 23 Nxg6 hxg6 24 c5 Bf8 25 Bg4 Rcd8 26 Rae2 Qd5
27 Bf4 Rxe2 28 Qxe2, G.Kasparov-M.Adams, Linares 2002. At this point 28
... Qc6 retains balanced chances for Black.
18 ... b6

Question: What is the point of b7-b6, which weakens Black’s queenside?

 Answer: It’s a stabilizing move which reduces the effectiveness of d4-d5 ideas from White. Also it
gives Black’s knight a semi-stable post on a5. The following year, Karpov tried 18 ... h6 19 d5 Ne5 20
Bxa7 Nd3 21 Bxd3 Qxd3 22 Rxe8+ Rxe8 23 Bd4 Qxa3 24 Bxf6 gxf6 25 Qd2 Kg7 A.Shirov-A.Karpov,
Cap d’Agde 2003. White’s position is only cosmetically better after 26 Nd4 Bg6 27 h3 Re4 when all of
Black’s pieces are very active and White has no way to exploit the breakup of Black’s kingside pawns.
19 h3
More energetic is 19 c5 and it leads to enormous complications after 19 ...
Bd7 20 Ba6 Rcd8 21 Qd2 Qg6 22 d5 Ne7 23 d6 Bg4 24 Nd4 Bh3 25 g3 bxc5
26 dxc7 (26 dxe7 Rxe7 27 Bd3 Qh5 28 Be2 Qg6 29 Bd3 is a draw by
repetition) 26 ... Rd5 27 Bd3 Qg4 28 f3 Qc8 29 Ne2 Bf5 30 Nf4 Rd7 31
Bxc5 Bxd3 32 Nxd3 Rxc7 and Black equalized, “ZackS”-“Pandini” Internet
2007.
19 ... Bg6?!
“Too slow,” writes Kasparov:
a) Kasparov suggested 19 ... Na5!? which Karpov had avoided, fearing 20
Bd2 but then Kasparov gave 20 ... Qb7!, undaunted by 21 Bxa5 bxa5 which
he assessed as “double-edged”. Houdini agrees with this evaluation.
b) Safer though is 19 ... h6 20 c5 Qd5! 21 Bc4 Qd7 22 cxb6 cxb6 23 Ba6
Rcd8 24 Bb5 Be4 25 Qa4, O.Zenker-P.Stadler, correspondence 2008. Black
retains equality after 25 ... Rc8 26 Bf4 Bxf3 27 Rxe8+ Rxe8 28 gxf3 Qf5 29
Bg3 Nxd4! 30 Bxe8 Ne2+ 31 Kg2 Nxc1 32 Qxa7 Nd3 33 Bxf7+ Kh7 (White
runs out of checks and Black menaces ... Ne1+) 34 Qxb6 Ne1+ 35 Kf1
Qd3+! 36 Kg1 Nxf3+ 37 Kg2 Ne1+ 38 Kg1 and the game ends in a draw.
20 c5!
 A powerful move, after which Black experiences difficulties along his
queenside light squares.
20 ... Ne7
Karpov rushes his knight away from White’s coming Bb5 pin.

Question: The e7-square feels like a contorted one.

Is the a5-square a superior option for the knight?

Answer: Both squares have their problems. The virtue of …Ne7 is that it overprotects the d5-square
and thus blockades the d4-pawn from advancing. After 20 ... Na5, White can punt 21 d5! and now if 21
... Bb2 22 cxb6 Bxc1 23 Bxc1! (White threatens to win heavy material with Bb5! next) 23 ... Qb1 24
Ba6 axb6 25 Bxc8 Rxc8 26 Qa4 (threatening a Qe8 mating cheapo) 26 ... Rf8 (26 ... h6 is met with 27
Bxh6) 27 Bf4 Qd3 28 Bxc7 Qxd5 29 Ne5! f6 30 Nxg6 hxg6 31 Bxb6, White has an extra pawn.
21 Ba6
The comp likes 21 g4!? h6 22 Bb5! and now maybe Black should consider
sacrificing the exchange for the light squares with 22 ... Qa8!? 23 Bxe8 Rxe8
with some, but probably not enough light square compensation for Black. I’m
not sure a line like this would appeal to Kasparov, who is not inclined to give
up his precious initiative for mere material gain.
21 ... Rcd8 22 Bg5
This time 22 g4 can be met with 22 ... Qd5! 23 Bc4 Qa8 24 g5 Bxd4 25
Nxd4 bxc5 when Black is doing just fine.
 22 ... Qc6 23 cxb6
Now Kasparov begins to inflict some damage to Black’s structure as well.
23 ... Qxb6 24 Bxf6! gxf6!
Karpov correctly avoids 24 ... Qxf6?! 25 Qa4 c6 26 Bb7 when black
pawns begin to fall.
25 Qa4 c6

“Here, with an obvious positional advantage and significantly more time

than my opponent, I subconsciously began losing concentration,” writes
Kasparov. As we all know, the hardest thing in the world is to bring home the
point from a position of advantage, against a grimly determined opponent.
26 Bf1
Kasparov criticized this move, claiming he played planlessly from this
point on. 26 Re3 intending to double on the e-file looks like a better way to
make progress.
26 ... Kf8
Just in case White attempts to double on the e-file, Karpov bolsters
defence of the e8-rook.
27 Re3 Rb8 28 g3?!
Kasparov also criticized this move, claiming the bishop shifts to the wrong
diagonal. He suggested 28 Bc4 after which Black’s defensive options feel
constrained.
 28 ... Red8 29 Bg2 Nf5!?

I have often told people that Karpov’s style is the oddest of all the World
Champions. So strange are his strategic algorithms, that I just can’t ever
predict his moves! I think this brilliant double pawn sacrifice, which
exponentially enhances Black’s piece activity level, is also unnecessary, since
29 ... Rd5! causes White to worry about ... Ra5, without assuming the risks of
this double-pawn sacrifice.
30 Rxc6 Qb2!?
Karpov refuses to keep ambition under a tight reign. Is he playing for the
win!? White has trouble winning after 30 ... Qb5! 31 Qxb5 Rxb5 32 Re1 Ra5
33 Ra1 Nxd4 34 Nxd4 Rxd4 35 Rxf6 Bf5! 36 g4 Be6 37 Rf3 when White
looks too tied down to convert the extra pawn.
31 Rec3 Kg7?!
31 ... Qe2! intending a check on b1, and Black stands no worse. For
example: 32 Qc4 (32 Bf1 Rb1 33 Qb4+! Rxb4 34 Bxe2 Rb1+ 35 Rc1 Rxc1+
36 Rxc1 Nxd4 37 Nxd4 Rxd4 Black stands no worse) 32 ... Rb1+ 33 Rc1
Qxc4 34 R6xc4 Rxc1+ 35 Rxc1 Nxd4 is equal.
32 Qxa7
Kasparov continues to ingest pawns to the point of repletion. Now Black is
in deep trouble. It’s hard to believe Kasparov lost this position, two pawns up
and with more time on the clock!
 32 ... Qa1+
I prefer 32 ... Re8 thinking about ... Re2.
33 Rc1 Rb1 34 Rxb1 Qxb1+ 35 Kh2 Qa2?!
Black puts up greater resistance with 35 ... Qb2! 36 Qb6 (36 d5 Rxd5 37
Rc8 Rd3 38 Ra8 Rxf3! 39 Bxf3 Nd4 40 Kg2 Nxf3 41 Qe7 (41 Kxf3?? Qb3+
42 Qe3 Qd5+ picks up the a8-rook) 41 ... Bc2 42 Kxf3 Qb3+ 43 Kg4 Qd5 44
Qf8+ Kg6 45 Qg8+ Kh6 and White must take perpetual check, since his own
king is exposed.) 36 ... Rb8 37 Qxb2 Rxb2 38 g4 Ne3 39 fxe3 Be4 40 Rc1
Bxf3 41 Rg1 Re2 42 a4 and I don’t think Black will be able to hold the draw.

One of the problems with playing under time pressure is that our normal
sense of nuance evaporates, and we regress to the state of cavemen and
cavewomen.

Exercise (combination alert): Karpov, seriously low on the clock,

blunders. White’s tribulations come to an end, but only if you find
the right line. How can Kasparov put his great opponent away?

36 Qc7
White is still winning after this move though he could have settled the
issue more decisively.
Answer: Kasparov missed the immediate win with 36 g4! which ends the game, no matter where
the knight plays:
a) 36 ... Nh6 37 Qe7 Rf8 38 Qxf6+ Kg8 39 Ne5 and if 39 ... Qxa3, 40 Bd5
forces mate.
b) 36 ... Nd6 37 Qe7 and the woebegone rook and knight can only sigh
miserably and take their medicine. This double attack ends the game.
c) 36 ... Rd6 37 Rc7 Nh6 38 Rc8! (threat: Qe7!) 38 ... Qe6 39 Qc5 Ng8 40
Nh4 Ne7 41 Rc7 when Black is pretty close to zugzwang.
d) 36 ... Ne3 37 fxe3 Be4 38 Qc7! Rd7! 39 Qc8 Rd6 40 Rc1! Bxf3 41 Qc2
and White wins.
36 ... Re8
36 ... Rd5! puts up greater resistance.
37 Rc2?!
37 g4! is once again strong.
37 ... Qxa3 38 Rd2
Kasparov consolidated somewhat, being a pawn up, but the win is no
longer a trivial matter.
38 ... Nd6 39 Qc5 Qa6!

Amazing defensive intuition. Karpov’s motives only become apparent

several moves later. Karpov had one minute to Kasparov’s five on the clock,
and intuitively he played the best defence. White’s game is too easy to play
after 39 ... Qxc5 40 dxc5 Ne4 41 Rc2 Nxg3 42 Rc4 Nf5 43 c6 when the c-
pawn ties Black down and I don’t believe he can save the game.
40 Nh4 Rc8 41 Qd5 Rc1 42 Rb2
With hindsight, it was better to fix Black’s structure and play 42 Nxg6
hxg6 when the win was going to be very difficult for White.
42 ... Be4!

“Know this presumptuous fool: I do not lose power simply because you
don’t believe in it,” the wizard scolds White’s queen. An amazing shot,
especially with near-zero time on the clock, playing on the increments.
Karpov, by some mysterious telepathic sensation, tends to have an exquisite
sense of timing of just when to rattle the position. So Black’s bishop, weary
of dealing with low level functionaries, decides he needs access to the highest
ranking white official: the queen.

Exercise: (critical decision): White puts out one fire, only to see another blaze up in the middle of
the board. What a nightmare to have to calculate this with the clock running down! Kasparov’s choices:
a) 43 Bxe4, accepting Black’s piece offer. Now factor in 43 ... Qf1. b) 43 Qa2, essentially ignoring the
piece offer. c) 43 Qxe4, declining Black’s offer and offering a queen for his own attack. One line gives
White good winning chances, another line should be drawn, while in the third White fights for the
draw. We are pulled in three different directions, yet can only choose one:

43 Qxe4?
The district delegates meet to vote on a new leader, and it’s not White’s
queen, as she incorrectly believes. Hmm. This move exudes the feel of
overkill. If you see a lone spider in your home, there is no need to fumigate
and squish it. Just do one or the other, or better yet, find a cup and put it
outside in the yard. One senses that Kasparov is most certainly not a man of
sunny or equable temperament, and that neither “tolerance” nor
“compromise” are words used very often in his vocabulary. Time pressure is
that strange place where irrationality prevails (or at least gets its way!) over
its virtuous twin, logic. This is the worst of the three, although White maybe
can still build a fortress draw. Karpov writes: “At first he simply wanted to
take the bishop. He had 14 seconds left – he stretched out his hand towards
his bishop, but he did not touch it. An instant later his hand hovered over the
e4-square, while with his eyes he traced the path of my queen from a6 to f1,
and then he cast a glance at his knight. What is surprising is that the black
king is exposed, but it can be checked only from one square – f5. Garry again
looked at the position, then at the clock, on which he had two seconds left,
and – he captured my bishop with his queen!”
Answer: White should decline all gifts with 43 Qa2! Qc6 44 Qb3. White remains up a pawn and
still has chances to win. On the other hand, allowing the queen access to the back rank with 43 Bxe4
Qf1, requires some defensive acumen with 44 Nf5+! Nxf5 45 h4! (giving White’s king air) 45 ... Nh6
46 Bf3 Qg1+ 47 Kh3 Rf1 48 Qe4 Rxf2 49 Rxf2 Qxf2 50 g4 f5 51 gxf5 Qf1+ 52 Kg3 Qg1+ 53 Bg2 f6!
and the knight will be brought back into play and White’s exposed king gives the position a trip-to-
nowhere feel, where a win is highly unlikely.
43 ... Nxe4 44 Nf5+ Kf8
White has no back rank mate.
45 Bxe4 Rc8 46 d5!?
This move puts the fortress at risk. 46 Kg2! suggested by Karpov. After 46
... Qe6 47 f3 Qc4 48 h4 Qc1 49 Rb6, Black’s win is still a million miles
away.
46 ... Qa4 47 Bf3??
Correct is 47 Bg2, when White may well erect a fortress draw.
 Exercise (combination alert): When you depart to reach a destination of choice,
general consensus deems that it is preferable to arrive alive, rather than
dead.
White plans to play Kg2, after which he should draw with his fortress.
However, his last move, a monster blunder, does exactly the opposite. Black
to play and win:
 Answer: Double attack. The b2-rook and f3-bishop hang, causing White to drop more material.
47 ... Qa3! 0-1
“In the absence of respect, I suffer hurt feelings. So to balance matters,
you both will soon experience another variety of hurt,” the queen warns
White’s nervous rook and bishop.
Summary: 16 c4 is too simple a move to give Black real problems. The trouble is by pushing c3-c4
too early is that White essentially loosens the d4-pawn, which will be subject to attack from Black’s
pieces.

Game 19
A.Morozevich-Wang Yue
Amber Blindfold, Nice (rapid) 2009

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 Re1

We arrive at another branch of the main line.

8 ... Bg4
We concentrate on this aggressive continuation, which is Black’s most
popular choice and skip coverage of 8 ... Bf5 and 8 ... Nf6.
9 c3

Question: Why can’t White simply win the e4-pawn?

Answer: White loses the initiative after 9 Bxe4?! dxe4 10 Rxe4 Bxf3 11 Qxf3 Nxd4 12 Qd3 Ne6
and White got nothing from the opening. We will cover 9 c4 in Game 22.
9 ... f5
Black logically backs up his formidable e4-knight at the long-term
strategic cost of creating an e5-hole. Let’s see the alternatives:
a) 9 ... 0-0?! is a dubious pawn sacrifice. Black doesn’t generate enough
compensation after 10 Bxe4 dxe4 11 Rxe4 f5 12 Re1 Bd6, K.Skinke-
S.Vledouts,Vaujany 2012. I don’t see enough compensation for Black after
13 Nbd2.

Question: Why must we weaken with ... f7-f5, when we can simply retreat the knight?

Answer: The retreat of the knight is passive.

b) Black gets a slightly unpleasant looking Exchange French-like position
after 9…Nf6 10 h3 Bh5 11 Bg5 (White scores 91.7% from this position,
which gives us a hint of the unpleasantness to come for Black) 11 ... 0-0 12
Nbd2 Re8 13 Qb3 (both the b7- and d5-pawns are under pressure) 13 ... Na5
14 Qa4 Nc6 15 Bb5 Qd6 (15 ... Bxf3 16 Nxf3 Ne4 17 Bxe7 Rxe7 18 Bxc6
bxc6 19 Ne5 Qe8 20 Nd3 also looks bad for Black, since his queenside
pawns have been severely weakened, but it is still better than what happened
in the game) 16 Ne5 and Black is busted, D.Baramidze-G.Pavlik, Tallinn
1997.
10 Qb3
White unpins and goes after the b7-pawn. We will look at 10 Nbd2 in
Game 21 but this move order doesn’t really change anything, since we
normally always transpose to the 10 Qb3 lines.
10 ... 0-0
Black correctly offers the b7-pawn. Let’s examine the lines which Black
keeps the pawn intact.

Question: Can Black cover the b7-pawn with 10…Na5?

 Answer: Castling is pretty much obligatory, since White wins by force here. After 11 Qa4+! c6 (if
11 ... Nc6 there follows 12 Ba6! and Black can already consider resigning) 12 Ne5 Bh5 13 f3 Nd6 14
Bg5! b5, Black resigned after 15 Qb4!, A.Morozevich-L.Kritz, Mainz 2006. Black’s game collapses
following 15 ... Nc8 16 Bxe7 Nxe7 17 a4! 0-0 18 axb5 Nb7 19 bxc6 Nd6 20 Nd7.
The passive 10 ... Rb8 is met with 11 Nfd2! Bh5 12 f3 Nxd2 13 Bxd2 0-0
14 Bf4 and Black, who basically just reached a drab inferior Exchange
French position, lost all three games in the database from this position.
11 Nbd2
White already has to exercise considerable care here as you can see from
the alternative 11th move options:

Question: Can White get away with 11 Qxb7?

 Answer: The immediate pawn grab allows Black a dangerous attack:
a) After 11 Qxb7 Rf6! (the rook considers the loss of b7 a trivially small
investment to acquire the infinitely larger treasure of White’s king) 12 Qb3
(12 Nbd2?? Rb8 13 Qa6 Nb4! (this uncovers the f6-rook on White’s queen
and wins) 12 ... Bxf3 13 gxf3 Rg6+ 14 Kf1 Rb8 15 Qc2 Qd6, Black commits
with a dangerous piece sacrifice. White scores a rather dismal 0% from this
position! Following 16 fxe4 Qxh2:
a1) 17 Ke2 Qh5+ 18 Kf1 fxe4 White has no hope of surviving Black’s
withering attack, J.Griffiths-A.Van Nierop, correspondence 2003.
a2)17 exf5 Rg1+ 18 Ke2 Qh5+ 19 f3 Qh2+ 20 Kd1 Rxe1+ 21 Kxe1 Bh4+
22 Kd1 Qg1+ 23 Ke2 when Black’s king proclaims, “I find your boasts to
take my life far-fetched and infinitely improbable,” and soon discovers a
distortion in own theory after 23 ... Qe1 mate! Apparently the pre-planned
expression of casual condescension the queen had prepared for White’s king
is wasted, since he is already dead.
b) The artificial 11 Nfd2? walks into 11 ... Nxf2!
 12 Kxf2 Bh4+ 13 g3 f4! 14 Kg2 fxg3 15 Be4 (15 hxg3 Bxg3! 16 Kxg3
Qd6+! 17 Kg2 Rf2+! 18 Kxf2 Qh2+ 19 Ke3 Kh8! 20 Qxd5 Qg3+ 21 Nf3
Bxf3 22 Qxf3 Qxe1+ 23 Be2 Qxc1+ 24 Kd3 Qxb2 is also curtains for White)
15 ... Bh3+! (White’s king is met with one windmilling blow after another)
16 Kg1 (16 Kxh3 Qd7+ 17 Kg2 Rf2+ 18 Kg1 Qh3 forces mate) 16 ... gxh2+
17 Kxh2 Qd6+ 18 Kh1 Bxe1 19 Qxd5+ Qxd5 20 Bxd5+ Kh8 left White
hopelessly busted, L.Ljubojevic-S.Makarichev, Amsterdam 1975.
11 ... Na5
The risky pawn sacrifice 11 ... Kh8!? “is in disfavour currently,” writes
GM Victor Mikhalevski. 12 h3 was played in perhaps the most famous
Petroff’s game ever, when Korchnoi challenged Karpov with it in their
Candidates’ final match:
a) The Karpov-Korchnoi game went 12 h3 Bh5!? (Korchnoi is not the type
to back off and give his opponent an easy strategic edge with 12 ... Bxf3 13
Nxf3 a6 14 Bf4) 13 Qxb7 Rf6 14 Qb3 Rg6 15 Be2! Bh4 16 Rf1 (Black has
an impressive force arrayed against White’s king, who is more well-protected
than he looks) 16 ... Bxf3 17 Nxf3 Bxf2+? (this attempt to storm his way in
fails) 18 Rxf2 Nxf2 19 Kxf2 Qd6 (threatening to sneak into the g3-square) 20
Ng5! (cutting off the ... Qg3+ threat, while menacing a nasty f7-fork) 20 ...
Rf8 21 Qa3! (fantastic defensive play by Karpov, who continues to throw the
attacking pieces out of synch) 21 ... Qd8 (21 ... Qf6?? is met with 22 Bh5!
Rh6 23 Nf7+ Rxf7 24 Bxf7 and White wins) 22 Bf4 h6 23 Nf3 Re8 24 Bd3
Re4!? (The Eagles hit song Desperado’s lyrics come to mind: “Desperado,
why don’t you come to your senses?”) 25 g3 (Karpov correctly ignores the
gift and solidifies his kingside) 25 ... Rf6 26 Qc5 g5 27 Nxg5! hxg5 28 Bxg5
and Black’s disastrous non-attack reminds us of the vaudevillian act which
gets booed by the audience, until a giant hook appears stage left to drag him
off stage, A.Karpov-V.Korchnoi, 6th matchgame, Moscow 1974.
b) The immediate 12 Qxb7 pawn grab has also been tried at the elite level. “Justice is only for those
who are caught,” declares the queen, as she pockets the silverware. Now White’s and Black’s motives
retain their starkly separate identities. Following 12 ... Rf6, V.Ivanchuk-A.Shirov, Dortmund 1998,
White’s chances look superior after 13 Qb3. White’s extra pawn and solid position mean more than
Black’s kingside attacking chances.
12 Qc2
Advancing the queen with 12 Qa4 instead is met with 12 ... Nc6 which
Mikhalevski indicates that the move is “underlining the misplaced position of
the queen on a4”. After 13 Bb5 (13 Qb3 Na5 14 Qa4 Nc6 will lead to a
repetition draw) 13 ... Bh4 (inducing a weakening of White’s kingside
pawns) 14 g3 Bf6 15 Bxc6 bxc6 16 Qxc6 and here, Black has two good
options:
a) 16…Re8 17 Ne5 Bxe5 18 dxe5 Ng5 (18 ... Rxe5 19 Nb3 Rb8 20 Be3
looks playable for Black) 19 f4 Nh3+! 20 Kg2 Rb8 21 c4 (after 21 Qc5 Rb6
22 Rf1 g5 23 Nf3 gxf4 24 gxf4 Rg6, Houdini claims White is okay, but to my
human eyes, Black’s attack looks pretty scary!) 21 ... dxc4 22 Nxc4 Qd3 23
Ne3 Rxb2+! (forcing the draw) 24 Bxb2 Qd2+ 25 Kh1 Nf2+ 26 Kg1 (26
Kg2?? Nd1+ forces mate) 26 ... Nh3+ 27 Kh1 Nf2+ with perpetual check,
V.Anand-V.Kramnik, Wijk aan Zee 1999.
b) Black can also try Gelfand’s 16 ... Rb8 idea which is best met with 17
Qe6+! (I don’t trust 17 c4 which weakens the d4-point and after 17 ... f4! 18
Nxe4 dxe4 19 Qxe4 Bxf3 20 Qxf3 fxg3 21 hxg3 Bxd4 22 Bf4 Rxb2 (White’s
pawns fall left and right) 23 Rf1 Rxf2! 24 Rxf2 Bxa1, Black is up a pawn
with all the winning chances, I.Cheparinov-B.Gelfand, Sochi 2008) 17 ...
Kh8 18 Nxe4 dxe4 19 Ne5 Bxe5 20 Qxe5 Bf3 21 Bf4 Rxb2 22 Qc5 Rg8,
J.Garcia Garay-J.Pina, correspondence 2003. I slightly prefer White’s
chances after 23 Rab1 Rxa2 24 Ra1 Rb2 25 c4. Now if 25 ... Qe8 26 d5 Qh5
27 h4 Qg4 28 Kh2, Black’s queenside pawns are about to fall and I don’t see
a way for him to proceed effectively with his own attack.
12 ... Nc6
 Covering the e5-square. Black can also try 12 ... Bd6 and after 13 Ne5 Bh5
14 b4 Nc6 15 Ndf3 Re8 16 Bb2!? Qf6 17 Qb3, P.Leko-V.Anand, Linares
2005, chances look balanced following 17 ... Bf7.
13 b4

Dual purpose:
1. White seizes useful queenside space.
2. White hopes to prevent Black from contesting for the e5-square, by
forcing b4-b5, which chases the c6-knight away.
13 ... a6
Stalling b4-b5 and preparing to gain the a-file when White finally gets in
the push.
14 Rb1
Preparing a2-a4, followed by b4-b5. Morozevich also has experience with
14 a4 Bd6 15 Ba3 with the idea of b4-b5. Note that 15 Rb1 transposes to the
game’s continuation. After 15 ... Kh8 16 b5 (16 Reb1 Nxd2 17 Nxd2 Qh4 18
Nf1 Ne7 19 f3 Bh5 20 b5 Bxa3 21 Rxa3 axb5 22 Rxb5 b6 and Black looks
fine, A.Morozevich-B.Gelfand, Yerevan 2008) 16 ... axb5 17 Bxd6 cxd6!
(seizing control over the e5-square, while preparing to pressure the c3-pawn
with a future ... Rc8) 18 axb5 Na5 19 h3 Bh5 20 Nh2 Rc8 21 Ra3 b6 22 Ndf1
f4! (cutting off the intended Ne3) 23 f3 (23 Bxe4 dxe4 24 Qxe4 Re8 25 Qb1
Rxe1 26 Qxe1 Qe8 offers Black full compensation for the pawn) 23 ... Qh4
24 Re2 Ng3 25 Nxg3 fxg3 26 Nf1 Bxf3! (a dangerous piece sacrifice) 27
gxf3 Rxf3 28 Nd2?! (28 Rg2 Rcf8 29 Be2 Rf2 30 Nxg3 Nc4! 31 Qc1 Qf6 32
Bxc4 Rxg2+ 33 Kxg2 Qf3+ 34 Kh2 Qf2+ is drawn) 28 ... Rf2 29 Rxf2 gxf2+
30 Kg2 Qg5+ 31 Kxf2 Rf8+ 32 Nf3 Qf4 33 Qe2 Qh2+ 34 Ke1 Qg3+ 35 Qf2
Rxf3 and Black regained the sacrificed piece with all the winning chances,
R.Ponomariov-M.Adams, Wijk aan Zee 2005.
14 ... Kh8

A new move and a possible improvement over:

a) 14 ... Bd6 15 h3 Bh5 16 b5 Na5, Hydra (computer)-M.Adams, London
2005. Black’s game looks slightly uncomfortable after 17 bxa6 bxa6 18 c4!.
b) The main continuation is 14 ... b5, halting White’s queenside advance.
After 15 a4 Rb8 16 axb5 axb5 17 Ra1, D.Stellwagen-D.Fridman, Netherlands
2007, Black achieves an acceptable position with 17 ... Qd6.
15 a4 Bd6
Covering the e5-square, while clearing the e7-one for his knight.
16 b5 axb5 17 axb5
After 17 Rxb5 b6!, White can’t play 18 Rxd5? Nxd2! (not 18 ... Bxh2+??
19 Nxh2 Qxd5 20 Nxe4 Rae8 21 Nxg4 fxg4 22 Re2 with a winning position
for White) 19 Bxd2 Bxf3 20 Rxd6 Qxd6 21 gxf3 Ne7, Black is up an
exchange for a pawn, and has potential against White’s king.
17 ... Ne7
 The knight heads for the kingside. Also possible is 17 ... Na5 18 Ne5,
V.Rasulov-K.Sakaev, St Petersburg 2011. Now Black should just play the
calm 18 ... Bh5 with a balanced game.
18 Ne5
At last, White is able to occupy the e5-square, while preparing f2-f3.
18 ... Bxe5
Wang Yue plugs up the e5-hole with a white pawn. Mikhalevski suggests
the possibility 18 ... Ng6!? when White should probably ignore the g4-
temptation and play 19 Ndf3 Bxf3 20 Nxf3 c5 21 bxc6 bxc6 with a decent
position.
19 dxe5 Nc5?!

In such highly charged positions of equilibrium, the slightest movement

from either party has a way of jostling the status quo, which in turn creates a
chain reaction, further inflaming the position. Black voluntarily abandons his
powerful e4-knight post, when he should ignore the coming f2-f3 and play 19
... Ng6! 20 f3 Qh4 21 Rf1 Bh5 when he stands no worse. If White tries 22
fxe4 fxe4, his bishop has no place to go, so he has to try 23 Rb4 exd3 24
Rxh4 dxc2 25 Rxh5 Ra1 26 Nb3 Rb1 27 e6 Kg8 28 Rxd5 Rxb3 29 Re1 Re8
when the game remains dynamically balanced.
20 f3 Bh5
Alternatively, 20 ... Nxd3 21 Qxd3 Bh5 22 Nb3 f4 23 Rb2 has White’s
passed e-pawn and d4-knight outpost giving him a clear strategic edge.
21 Nb3 Ne6
Black attempts an e6-blockade of White’s passed e-pawn, as we have seen
earlier in the chapter. But in this case, Black’s blockade won’t hold for long.
22 Nd4
Challenging the blockade immediately.
22 ... Qd7 23 Rb3!
A powerful strategic idea, preparing Ba3, which in turn endangers Black’s
f5-pawn. I remember an Amazon customer review (I begin to despise my
species if I read the negative and troll reviews, so I don’t read them much
anymore!) where one reader took offense at the fact that I included a
blindfold game in my book. What he failed to understand is that world class
players are probably still playing at minimum IM strength without sight of
the board!

23 ... Bg6
Reinforcing the f5-pawn.
24 Ba3
Also simple and strong is 24 Nxe6 Qxe6 25 f4 with a clear advantage for
White, due to his superior opposite coloured bishop and protected, passed e-
pawn.
24 ... Rf7
 Unpinning the e7-knight at the cost of a very passively placed rook.
25 Nxe6 Qxe6
Principle: The queen is the worst possible blockader, since an attack by
any minor piece forces her to back away.
26 f4
Also tempting is 26 b6! Nc6 27 bxc7 Rxc7 28 Bd6 Nd4 29 Qb2 Nxb3 30
Bxc7 Nc5 31 Bb5 with a continued strategic grip for White.
26 ... Nc8!

Activating the knight by preparing to transfer it to the b6-square, where it

clamps down on potential c3-c4 breaks from White.
27 Bc5 Nb6 28 Rbb1!
The rook has done its job on the b3-square and now prepares to challenge
the a-file.
28 ... Rd7 29 Ra1 Rb8!
Black avoids the trap 29 ... Rdd8?? which fails miserably to 30 Rxa8 Rxa8
31 g4!, winning on the spot since Black’s control over f5-square collapses.
30 Bxb6!?
 Madness arises only when the subconscious mind wrests dominance from
our day-to-day conscious mind. But then the problem for the crazy mind is to
blend into society, without causing suspicion. So the now dominant insane
mind must feign normalcy! Moro’s last original – an incredibly strange
strategic decision – is something which would never have occurred to me.
Indifferent to the dictates of convention, he decides to hand over the bishop
pair – including his ultra-powerful dark-squared monster – merely to stick
Black with a bad remaining bishop. To me, the move is as strategically
disorienting as a witch who wakes up one Sunday morning and decides that
Satan isn’t her cup of tea. So she retires her flying broom, smashes her
cauldron, gives away her cat to the Humane Society, puts on a frilly pink
dress, and goes to church. Let’s get back to the game.
30 ... Qxb6+
I would play 30 ... cxb6 to use the open c-file to pressure White’s c-pawn.
31 Kh1
Also, 31 Qf2 d4 32 c4 Re8 favours White.
31 ... Qe6
Back to the drudge work of the minimum wage job of blockading from the
e6-square. White’s queen takes cruel pleasure in observing her e6-sister’s
embarrassed state.
32 c4?!
 This move is really weird as well, since it activates Black’s d7-rook. The
idea is to break the e6-blockade, but I think White pays too high a price.
32 ... dxc4 33 Bxc4
Finally, Black’s queen is asked to step aside.
33 ... Qe7
The c4-bishop’s ‘inducements’ manage to overcome the queen’s shyness.
34 Be2
Thinking about transferring to the f3-square, where it bears down on the
b7-pawn.
34 ... Qb4
The position displays a leavening of only partially formed, hidden truths,
made a million times more obscure from the fact that this is a blindfold game.
Black can try 34 ... Rd4!? 35 g3 Bf7! when 36 Qxf5? should be avoided, due
to 36 ... Bd5+ 37 Kg1 Qc5 38 Kf1 Rf8 39 Qg4 Qb4! with a nasty rook
sacrifice threat on the f4-pawn.
35 Red1!?
Stronger is 35 e6! Re7 36 Bc4 and Black is kept tied up.
35 ... Rxd1+ 36 Rxd1 Qxf4 37 Qxc7
Now the b7-pawn comes under fire.
37 ... Re8

38 b6
 Exercise (critical decision): Can Black get away with 38 ... Qxe5?
38 Bf3! h5!, offering Black’s king access to both h7- and h6- squares?

38 ... h6?
Answer: Sometimes we lose games from seeing ghosts. The answer is ‘Yes’. Black can just take the
pawn and get away with the crime, with 38 ... Qxe5! 39 Qxe5 Rxe5 40 Bf3 (40 Rd8+? is met with 40 ...
Be8) 40 ... h5! and the trouble is, 41 Bxb7? (correct is 41 g4! fxg4 42 Bxb7 Re6 43 Bc6! Rxc6 44 b7
Be4+ 45 Kg1 Rb6 46 Rd8+ Kh7 47 b8Q Rxb8 48 Rxb8 with a draw) is met with 41 ... Rb5 and Black
picks off the b6-pawn with some winning chances.
39 e6?!
Moro finds a tactical solution. 39 Qxb7! offers White good winning
chances.
39 ... Qe3 40 Bc4 Bh5?!
40 ... f4! intending ... f5-f4-f3 allows Black to escape. For example, after
41 Bd5 Qe2 42 Rg1 Be4 43 Qd7 Bxd5! 44 Qxe8+ Kh7 (Black threatens ...
f4-f3!) 45 Qd8 Bxg2+ 46 Rxg2 Qf1+, the game ends with perpetual check.
41 Rf1 Be2 42 Qf7!
The once missing cohesion reappears, magically reanimating White’s
position. This zwischenzug seizes the advantage.
42 ... Bxc4!
Stronger than 42 ... Rg8?! 43 Bxe2 Qxe2 44 Rg1 Qe4 45 e7 Kh7 46 Qf8
when Black still struggles.
43 Qxe8+ Kh7 44 Rg1 Bxe6?!
Stronger is 44 ... Bd5! 45 Qb5 (45 e7 Bxg2+! 46 Kxg2 Qe4+ is drawn by
perpetual check) 45 ... Qxe6 when Black looks fine.
45 Qb5!
Cutting off ... Bd5.
45 ... Qd4 46 Qb1!
Once again cutting off ... Bd5.
46 ... Qc5 47 Rc1 Qf2 48 Rf1 Qc5 49 h3 Bd7 50 Rc1
More accurate was 50 Qd3 Be6 51 Rb1.
50 ... Qf2?
White’s win is not so easy after 50 ... Qd5!.
 Exercise (planning): When we put off needful measures for later,
we must realize that there may come a time where there is no more
‘later’. Black’s last move was a blunder and his apprehensions are
far from allayed. How did Morozevich take advantage of the last move?

Answer: Seize the seventh rank, targeting the g7- and b7-pawns.
51 Rc7! Qd4
Other futile options:
a) 51 ... Ba4 is met with 52 Qb4! Qe3 53 Qb2! Qg3 54 Rxb7 Bc6 55 Re7
Qxh3+ 56 Kg1 Qg3 57 b7 and White wins.
b) 51 ... Bc6?? is easily refuted by 52 Rxc6 bxc6 53 b7 and White
promotes.
52 Rxb7 Bc6 53 Qxf5+ 1-0
Summary: Black should be fine theoretically in the 8 Re1 9 c3 variation. The only problem for us
is to familiarize ourselves with the mountain of theory!

Game 20
W.So-A.Giri
Corus B, Wijk aan Zee 2010

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 Re1 Bg4 9 c3 f5 10 Qb3 0-0

Question: Can Black get away with a piece sacrifice with 10 ... Bxf3?
 Answer: It’s unsound at worst and semi-sound at best. I see only one game in the database (Black
won it so now the sac scores 100% for Black!). White should consolidate after 11 gxf3 Qd7? (11 ...
Na5 looks like a better try when after 12 Qa4+ c6 13 fxe4 fxe4 14 Qd1! 0-0 15 Bf1 Bh4 16 Be3 Qf6 17
Qc2 Rf7 18 Na3 Raf8 19 b4 Qg6+ 20 Bg2 Rxf2 21 Bxf2 Rxf2 22 Qxf2 Bxf2+ 23 Kxf2 Qd6 24 Kg1 b6
25 Re2 Nb7 26 Nc2, White’s two rooks and minor piece are clearly worth more than Black’s queen and
two pawns) 12 fxe4 fxe4 13 Bf1 0-0-0 14 Be3 g5 15 Nd2 Qf5 16 c4 h5 17 cxd5 Rxd5, J.Elburg-
J.Morgado, correspondence 1976. White is winning after 18 Bh3! g4 19 Bg2 Nxd4 20 Bxd4 Rxd4 21
Bxe4 and Black can resign.
11 Nbd2 Na5 12 Qa4 Nc6 13 Qb3
This repetition really changes nothing from the previous game, except a
pair of extra moves for each side.
13 ... Na5 14 Qc2 Nc6 15 b4 a6 16 a4

Question: What is the difference between this move order

and 16 Rb1, which we saw in the previous game?

Answer: I think the main difference is that with the 16 a4 move order, White cuts out early ... b7-b5
options, so it may be a shade more accurate.
16 ... Bd6 17 Ba3
 A divergence from the 17 Rb1 plan.

Question: What is the idea behind 17 Ba3, since b4-b5

doesn’t look possible, due to the hanging a3-bishop?

Answer: The a3-bishop won’t hang when White plays Qb2. Then b4-b5 is enabled.
17 ... Kh8
This is considered more accurate than 17 ... Rf6 18 b5 Nxd2 19 Qxd2! (19
Nxd2 allows 19 ... Bxh2+! and now if 20 Kxh2 (20 Kf1!? axb5 21 axb5 Rh6!
and now 22 bxc6 is met with 22 ... Bd6! (threatening mate) 23 g3 Rxa3 24
cxb7 Rxa1 25 Rxa1 Rh1+ 26 Kg2 Rxa1 27 Qb2 Qb8 28 Qxa1 Qxb7 29 f3
Bh5 30 Bxf5 Bf7 with an equal position) 20 ... Rh6+ 21 Kg1 Qh4 22 f3 Qh1+
23 Kf2 Qh4+ 24 Kf1 Qh1+ with perpetual check) 19 ... Bxf3 20 bxc6 Rg6 21
g3 bxc6 22 Bxd6 cxd6 23 Bxf5 Rf6 24 Qd3 g6 25 Qxf3 Rxf5 26 Qe2 Rf7 27
a5 with an edge for White, who owns the e-file and ties Black down to
defence of the a6-pawn, A.Morozevich-A.Karpov, Prague 2002.
18 Qb2
White protects his a3-bishop, which in turn keeps Black concerned about
the b4-b5 push. The immediate 18 b5 forces Black to compromise his
structure with 18…axb5 19 Bxd6 cxd6! 20 axb5 Na5 when Black’s potential
pressure down the c-file gives ample compensation, R.Ponomariov-
M.Adams, Wijk aan Zee 2005.
18 ... Ne7
Let’s examine other choices:
a) 18 ... Re8 19 b5 axb5 20 Bxb5 Qf6 21 Bxd6 cxd6 22 a5 (the
undermining a5-a6 is in the air) 22 ... Reb8 23 Ra4!?, P.Svidler-P.H.Nielsen,
Dortmund 2005. Here Black can take the pawn with 23 ... Nxa5!? 24 Rea1
Qd8 25 Qa3 Bxf3! 26 Nxf3 Nc4! 27 Bxc4 Rxa4 28 Qxa4 dxc4 29 Qxc4 g6
and Houdini claims White has compensation for the pawn. In my opinion,
Black stands no worse here.
b) 18 ... Qe7 adds pressure to the a3-pawn. This discourages b4-b5, while
bolstering support of the e5-square. The downside of the move is that it eats
up the e7-square, normally reserved for the c6-knight. After 19 b5 Bxa3 20
Rxa3 axb5 (the point of 18 ... Qe7 is that White is unable to recapture on b5
with his a-pawn) 21 Bxb5 Bxf3 22 Nxf3 Nd8! 23 Bd3, D.Andreikin-
A.Rakhmanov, Dagomys 2009, chances are even after 23 ... Qd6 24 Ne5
Nc6!.
19 Ne5 Bxe5 20 dxe5 Ng6!
In life-threatening situations, we tend to overlook the niceties of polite
conduct. Giri correctly commits to a piece sacrifice to generate a kingside
attack.
21 f3 Nxe5!?
 This is in essence a speculative piece sacrifice though Black has decent
options:
a) 21 ... Qg5! looks like a huge improvement. Following 22 fxg4 Nxd2 23
Qc2 Nh4 24 Rad1 Nhf3+! 25 gxf3 Nxf3+ 26 Kg2 fxg4! 27 Re2 Qf4 28 Kh1
Nh4 (threat: ... Qf1+ with a back rank mate!) 29 Rg2, K.Stalmach-
F.Kunzelmann, correspondence 2010, Black looks better after 29 ... Nxg2 30
Qxg2 Rae8.

Question: Why is Black obliged to sacrifice when he has 21 ... Nxd2?

Answer: You avoid the piece sacrifice in this way, but Black completely loses the initiative.
b) 21…Nd2 22 Qxd2 Bh5 23 b5 Re8 24 bxa6 bxa6 25 f4 ensures strategic
domination for White:
1. Who owns a protected, passed e-pawn.
2. Who owns the bishop pair.
3. Bc5 is coming, when White effectively blockades Black’s queenside
pawn majority.
4. White dominates the dark squares.
5. Black’s remaining bishop is a bad one.
22 Bf1 Qf6!
Houdini doesn’t like this, but I do. Giri refuses to gather pawns in exchange for slight strategic
inferiority after 22 ... Nxd2 23 Qxd2 Nxf3+ 24 gxf3 Bxf3 25 b5! Rf7 (25 ... Rf6?? 26 Be7 Rg6+ 27 Kf2
and White wins) 26 Qf4 Be4 27 bxa6 bxa6 28 Bc5 when White’s grip on the dark squares means that
his extra piece is worth more than Black’s three pawns.
23 fxe4 fxe4 24 Kh1?!
This king tuck is a misstep, after which Black takes over the initiative.
White should seize control over both the f2- and d4- squares with 24 Nb3 Rf7
25 b5 Bd7 26 bxa6 bxa6 27 Nd4 Qg5 28 c4! c5 29 Bxc5 Rc8 30 Bd6 Nxc4
31 Bxc4 Rxc4 32 Rf1 Rxf1+ 33 Rxf1 h6 34 Ne2 (intending Be5, with a dark
square blockade) 34 ... d4 35 Qb3! Qe3+ 36 Qxe3 dxe3 37 a5 and Black will
be hard pressed to hold the position.
24 ... b5!
Giri is determined to defy steep odds with élan. This brilliant shot,
imposing some order to the chaos, effectively slams the door on White’s a3-
bishop, since b4-b5 is denied.
25 Nb3
Accepting the pawn sacrifice would lead to a highly aesthetic drawing line
after25 axb5 axb5 26 Bxb5 Qb6! 27 Bf1

27…Bd1! (a startling clearance shot). Now if 28 Raxd1 Ng4 (threatening

smothered mate, starting with ... Nf2+) 29 Re2 Nxh2! 30 Kxh2 Qh6+ 31 Kg1
Qb6+ when perpetual check cannot be avoided.
25 ... Qh4!
The queen regards the white king’s vehement protests as the tantrums of a
petulant child, which are best left ignored. Now White must be on the lookout
for both ... Rf2 and ... Rf5-h5.
26 Qd2 Rf5! 27 Nd4
Most definitely not 27 Qxd5?? Raf8 28 Qxe4 Nf3! 29 h3 Qg3! and it is
curtains for White.
27 ... Rh5 28 h3
 28 ... Rf8?!
Black decides to abbreviate his rage and place it aside for some other time.
This is too slow and now White can defend the threats, while remaining up
material. There are certainly better choices:
a) Black should indeed sacrifice with 28 ... Bxh3! 29 gxh3 Ng4! 30 Re2
Rf8 (invasion of the f2-square is imminent) 31 Qe1 Nf2+ 32 Rxf2 Rxf2 33
Bg2 h6 34 Qe3 Rg5 35 Rg1 Rg3! 36 Qxf2 Rxh3+ 37 Bxh3 Qxf2 38 Rf1 Qd2
and amazingly, Black’s queen and pawns are enough against three minor
pieces and rook. For example, 39 Ne6 Qxc3 40 Rf8+ Kh7 41 Bf5+ g6 42
Rf7+ Kg8 43 Rf8+ draws by perpetual check.
b) Also fine for Black is 28 ... Nd3! 29 Re3 (29 Bxd3? Bxh3 30 Kg1 Bf5!
31 Rf1 Qh2+ 32 Kf2 Rf8 33 Ke1 exd3 34 Qf2 Rf6 35 Bc1 Bh3 36 Bf4 Qxg2
37 Qxg2 Bxg2 and White will be lucky to draw) 29 ... Nf2+ 30 Kg1 Rf8 31
Ra2 Nxh3+! 32 gxh3 Rxf1+! 33 Kxf1 Bxh3+ 34 Rxh3 Qxh3+ 35 Qg2 Qd3+
36 Re2 Qb1+ 37 Kf2 Rh6 38 Nc2 Qd1 39 Ne1 Rf6+ 40 Ke3 bxa4 when
Black’s five pawns and initiative offer him full compensation for the two
pieces.
29 Re3!
So understands his own weak points and shores them up immediately. This
move, reinforcing the f3- and h3-weaknesses, makes it difficult for Black to
strengthen his attack.
 29 ... Rf2 30 Qe1 Qf6?
Black has to try 30 ... Rh6 31 Kg1 Rhf6 32 hxg4 Qxg4 33 Rh3 Rxf1+ 34
Qxf1 Rxf1+ 35 Rxf1 h6 when Black still struggles, but less so than in the
game’s continuation.

If a falsehood is repeated enough, eventually you will believe it.

Sometimes when I’m attacking and sense that it’s going downhill, I overhear
my subconscious mind coaxing my conscious mind with trite phrases like
“Easy does it! You’ll see, it’s going to work!” – sort of a ‘fake-it-until-you-
succeed’ strategy.

Exercise (combination alert): Black’s last move was a

serious error. White to play and win more material:

Answer: Double attack/unpinning.

31 Kg1!
A move which reverberates with not-so-hidden menace. The king marches
up and down, exposing incompetence and issuing manifestos. Suddenly, the
f2-rook and g4-bishop hang simultaneously.
31 ... Rf4 32 hxg4
Membership of Black’s elite fighting forces dwindle, mainly by way of the
grave.
 32 ... Nxg4 33 Rh3 Rxh3 34 gxh3 Nf2 35 Qe3!
The queen’s reptilian personality is spiteful and cold, in equal parts. This
cold-blooded move should lead to a win.
35 ... Qg5+ 36 Bg2 Nd1!?
A move based on the fact that after sunset all cats look black. This is as
good as any, since everything loses. Giri is busted but isn’t quite ready to beg
for mercy. So he sets a last desperate trap. A few remaining perils lie in
ambush for White in the implementation of his nearly inevitable
consolidation.

37 Qc1?
White missed two consolidation methods:
a) 37 Qe2! Nf2 38 Bc1! Nxh3+ 39 Kh2 (the king’s skill at dissimulation
and evasion has come from a lifetime of steadily increasing (and well
justified!) paranoia) 39 ... Qh4 40 Be3! Ng5+ 41 Kg1 Qg3 42 Bxf4 and it’s
game over for Black.
b) 37 Rxd1! is the simplest way as White deliberately allows Black’s
‘combination’ with 37 ... Rf1+ 38 Kxf1 Qxe3 39 Ne2 and White has way too
much for the queen.
37 ... Qg3! 38 Ne2??
White still has chances to convert after the correct 38 Qxd1 Rf2 39 Qf1!
Rxf1+ 40 Rxf1 h5 41 Ne2 Qd3 42 Rf8+ Kh7 43 Bf1 bxa4 when White still
stands better.

Exercise (combination alert): Find a shot which

discredits White’s last move and sends him reeling.

Answer: Attraction.
38 ... Rf1+! 0-1
A crushing riposte indeed. After 39 Kxf1 (“I rule with such consummate
wisdom, that no improvements over my decision-making are possible,”
theorizes White’s king, who isn’t as wise as he imagines), Black commits
bloody murder with 39 ... Qf2 mate! Strangely enough, the queen likes to
watch her executions accompanied by a string quartet, which she claims
greatly enhances her pleasure.
Summary: Giri’s piece sacrifice looks sound, especially if we factor in the improvement 21 ...
Qg5!.

Game 21
A.Volokitin-L.Dominguez Perez
Bermuda 2005

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 Re1 Bg4 9 c3
In the next game, we will look at 9 c4.
9 ... f5 10 Nbd2
 Question: Does it make a difference if White plays 10 Nbd2, rather than 10 Qb3?

Answer: As mentioned in a previous game, this shouldn’t alter matters, since the two move orders
tend to merge and transpose.
10 ... 0-0 11 Qb3 Na5 12 Qc2 Bd6!?

This is a slightly altered move order from the last two games, which
allows White his next move. Black refuses to retreat his a5-knight and
instead, encourages White’s knight to hop into the e5-square.
13 Ne5!
This is the only way to try and exploit Black’s move order.

Question: Can White transpose to lines we looked at previously with 13 b4?

Answer: No, Black can avoid those lines with a little equalizing trick, 13 ... Bxf3!. This well-timed
move weakens White’s coverage of the c4-square. Following 14 Nxf3 Nc4 (this is one of the points of
the 12 ... Bd6 move order), if White proceeds ‘normally’, believing him or herself to be transposing to
the main lines, he gets tricked into a nice version for Black. After Black’s knight reaches c4, White has
nothing. After 15 Bxc4 dxc4 16 Be3 Re8 17 Rad1 Qf6, if anyone stands better, it’s Black, V.Anand-
B.Gelfand, Moscow 2004.
13 ... Bxe5!?
This move may well be inaccurate. Such concessions tend to be made in
the name of expediency, with the hope that the strategic indiscretions won’t
come back to haunt us in the future.

Question: Does Black have to hand over the bishop pair?

Answer: No. Black can also try 13 ... Bh5! which seems to offer decent chances:
a) 14 f3 is met with 14 ... Qh4 15 Rf1 Rae8! and if 16 fxe4?! fxe4, Black has full attacking
compensation for the piece.
b) 14 b4 Nc6 15 Ndf3 Re8 16 Bb2 Qf6 17 Qb3 Kh8 18 Be2?! (White should try 18 Qxd5 Rad8 19
Bxe4 Bxe5 20 dxe5 Qf8 21 Qc5 Qxc5 22 bxc5 fxe4 23 Rxe4 Bxf3 24 gxf3 Rxe5 25 Rae1 Rxe4 26 fxe4
Ne5 27 Kg2 though Black stands no worse) 18 ... Rxe5 (18 ... Bxf3! 19 Nxf3 g5! 20 Bb5 g4 21 Ne5
Nxe5 22 dxe5 Bxe5 23 Bxe8 Rxe8 offers Black huge compensation for the exchange) 19 dxe5 Nxe5 20
Nxe5 Bxe5 21 Bxh5 Bxh2+! (remember this trick as it crops up a lot in the 9 c3 variation) 22 Kxh2
Qh4+ 23 Kg1 Qxf2+ 24 Kh2 and the game was agreed drawn (can’t argue with perpetual check
anyway) in P.Leko-V.Anand, Linares 2005.
14 dxe5 Qh4 15 Nf1
15 Rf1! secures an edge in P.Smirnov-V.Ivanchuk, Yerevan 2004. Houdini
gives 15 ... Bh5 16 f3 Nxd2 17 Bxd2 Nc4 18 Rae1 Nxd2 19 Qxd2 Qe7 when
White has the better bishop and the passed e-pawn is superior to Black’s
queenside majority.
15 ... Nc6 16 Bf4 Rae8

Black dares White to win a piece with f2-f3.

17 f3!?
Tempting, but White should decline the piece offer. We are better off
enduring hunger, rather than eating food laced with poison. 17 e6! insures a
slight advantage, F.Patocka-C.Tschohl, correspondence 2006. A confident
player may label such a move ‘over-cautious’ while the naturally cautious
among us (including your writer) may deem it ‘good common sense’.
Pointedly ignoring someone (in this case White ignores his ‘threat’ of f2-f3)
is in a strange way, calling attention to yourself. After 17 ... Rxe6 18 f3 Bh5
19 Bxc7 Qe7 20 Bf4 Nc5 21 Rxe6 Nxe6 22 Bg3 Bg6 23 Re1 Qf7 24 Qd2,
White’s bishop pair still offers a small edge.
17 ... Bxf3
The bishop, a believer, now believes even more. Black is unhampered by
inhibition of any kind.
18 Bg3
This move wins a piece, at a high cost.
18 ... Nxg3 19 hxg3 Qh5 20 gxf3 Nxe5 21 Be2 f4!

An unwelcome problem emerges for White. This move underscores

Black’s not-so-covert ambition. Now the black kingside declares itself an
autonomous region, no longer under White’s authority. Volokitin may have
expected 21 ... Nxf3+?! 22 Bxf3 Qxf3 23 Qf2 when White’s safe king insures
him the advantage.
22 g4?!
The defensive obligations tax even Volokitin’s defensive competence.
Now Black stands better. When we try and fix a problem which isn’t a true
problem, it’s a little like dreaming we are bitten by a venomous snake, and
then drinking the antidote to the poison: both snake bite and antidote are
unreal, since it was all a dream. A better try is 22 Qd2 fxg3 23 Qxd5+ Kh8 24
f4! Qh6 25 Nxg3 (25 fxe5?? Rxf1+! forces mate) 25 ... Qxf4 26 Qg2 Qe3+
27 Kh1 Qh6+ 28 Qh2! (28 Kg1? is met with 28 ... Qb6+! 29 Kh1 Rf4! with a
dangerous attack) 28 ... Qc6+ (now a draw is as unalterable as a pendulum’s
return swing) 29 Qg2 Qh6+ and it’s perpetual check.
22 ... Qg5?!
This move loses a precious tempo for his attack and evens out the game.
Stronger is 22 ... Qh4! (the queen declares her sinister intention by hitting the
Re1 and the threat of ... Nxf3+! gains a tempo) 23 Red1 h5! 24 Rxd5 (24
gxh5? Rf6 25 Rxd5 Qxh5 26 Nh2 Rh6 and Black has a winning attack) 24 ...
hxg4 25 Rxe5 Rxe5 26 Bc4+ Kh8 27 Qh2 Qxh2+ 28 Nxh2 g3 29 Ng4 after
which Black’s rook and two powerful pawns are probably superior to White’s
two minor pieces.
23 Qd2
Attacking the d5-pawn.
23 ... c6 24 Nh2
Now he has time to adequately cover the g4-pawn.
24 ... h5
Dominguez decides that the g4-pawn is in dire need of a good soak and
scrub down.
25 Rf1 hxg4 26 Nxg4 Qh4
 27 Nh2?
White was okay after either 27 Bd1! or 27 Rf2!.
27 ... Rf6
Black’s attack gains momentum.
28 Rf2
Desperately attempting to cover his second rank. Also insufficient is 28
Qe1 Qg3+! 29 Qxg3 fxg3 and White loses his extra piece, with a busted
position.
28 ... Rg6+ 29 Rg2 Ree6
Black’s rooks creep closer to White’s king. Every piece except the king
participates in the assault.
30 Rf1
White is just one move away from Rff2, but he doesn’t have time.
30 ... Rxg2+ 31 Kxg2
 Exercise (combination alert): What is Black’s best continuation of the attack?

Answer: Drag the white rook to the h-file, then force the king off the defence of the hapless knight.
31 ... Rh6!
This wins back the piece. White may just barely survive the inaccurate 31
... Rg6+ 32 Kh1 Qg3 33 Bd1 Nc4 34 Qe2! Ne3 35 Re1 when Black has no
forced win.
32 Rh1
Euthanasia is possible with 32 Ng4?? Qg3 mate!
32 ... Qg3+ 33 Kf1 Rxh2 34 Rxh2 Qxh2
Black is up two pawns.
35 Qd4
Attacking the e5-knight and a7-pawn.
35 ... Qh3+ 36 Ke1
After 36 Kf2, White’s king is suffering from the classic battered spouse
syndrome: the more abuse heaped upon him by Black’s queen, the more he
considers her misunderstood, and loves her even more. Following 36 ... Qg3+
37 Kf1 Nxf3, the threat of ... Nh2 mate wins a third pawn.
36 ... Nxf3+ 37 Bxf3 Qxf3 38 Qxa7
 Exercise (planning): Black to play and force the win:

Answer: Simplification/pawn sacrifice. White can resign as the resulting king and pawn ending is a
trivial win for Black.
38 ... Qe3+
The queen clinks a spoon to her wine glass to attract everyone’s attention.
39 Qxe3
Hey, I said: “White can resign!”
39 ... fxe3 40 Ke2 Kf7 41 Kxe3 g5
The g-file carries a precious cargo and the outside passer offers White zero
hope.
42 a4 Kf6 43 a5 Ke5 44 Kf3 Kf5
“Who is this finger-fluttering fop, who dares approach me as an equal?”
asks Black’s king of his f3-brother, who now wishes he hadn’t waved hello.
45 Ke3 g4 46 Kd4 g3 47 Ke3 Kg4 0-1
Summary: 12 ... Bd6!? is an interesting side line which may throw off a booked up opponent. We
can improve on the game continuation with 13 ... Bh5!.

Game 22
A.Naiditsch-V.Kramnik
Dortmund 2008

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 Re1 Bg4 9 c4
 In this version, White chooses dynamism over safety. His last move
attempts to undermine the e4-knight, at the concession of weakening the d4-
pawn.
9 ... Nf6
Black’s wisest option. We simply retreat our knight from the e4-danger
zone, while reinforcing the d5-pawn. Other tries, which we will not cover in
detail, are:
a) 9 ... Bxf3 10 Qxf3 Nxd4?! 11 Qe3 Nf5 12 Qh3 Nfd6 13 cxd5 Nf6 14
Bg5 and White menaces Bxf6 with Black finding himself dangerously behind
in development, E.Lobron-E.Handoko, Indonesia 1983.
b) 9 ... f5?! 10 cxd5 Qxd5 11 Nc3 Nxc3 12 bxc3 Bxf3 (12 ... 0-0-0 looks a
bit better, but I still don’t like Black’s position after 13 Rb1) 13 Qxf3 Qxf3
14 gxf3 Rf8 15 d5 Na5 16 Ba3 Rf7 17 Re5 (threat: d5-d6! with a double
attack on the e7-bishop and the a5-knight) 17 ... Kd7, Y.Liao-O.Sulliman,
Dubai 1986. Black is busted after 18 Bxf5+ Kd8 19 Rae1.
c) 9 ... Nxd4?? simply hangs a piece to 10 Bxe4 dxe4 11 Qxd4.
10 Nc3!?
This is a dangerous pawn sacrifice for our side, which we need to know
cold. Note that 10 cxd5 isn’t as dangerous for Black and he should counter
with 10 ... Bxf3! (this move forces queens off the board and equalizes) 11
Qxf3 Qxd5 (with the d4-pawn under attack, White has no choice but to enter
an even ending) 12 Qxd5 Nxd5 13 Be4 0-0-0 14 Nc3 Bb4 15 Bd2 Nf6 16
Bxc6 bxc6 17 Be3 Nd5 18 Rac1 Nxe3 19 fxe3 c5 20 Rf1 f6 21 Rf5 Rhe8 22
dxc5 Rxe3 23 Na4 c6 and Black already stands better, since his rooks control
both open files and his bishop is superior to White’s knight, G.Kamsky-
A.Karpov, Linares 1994.
10 ... Bxf3!?
Kramnik accepts the challenge. Current theory says we should accept
White’s pawn offer.

Question: What happens if we decline with 10 ... 0-0?

Answer: This line is also playable and has the benefit of avoiding a big chunk of tricky theory.
After 11 cxd5 Nxd5 (also playable is 11 ... Nxd4 12 Bxh7+ Nxh7 13 Qxd4 Bxf3 14 gxf3 Bf6,
S.Kapnisis-S.Malikentzos, Athens 2012. Houdini gives White an edge following 15 Qd3 Qd7 16 Bf4
Rfe8 17 Re4 Ng5 18 Bxg5 Bxg5 19 Qb5. To me, Black’s ending looks perfectly playable after 19 ...
Qxb5 20 Nxb5 Red8 21 Rd1 Rd7 so I don’t think White’s extra pawn means all that much and feel
Black should hold this.) 12 h3 Be6 13 a3 Re8 14 Qc2 h6 15 Be3 Qd7, I would be perfectly fine with
Black’s position in this isolani position, C.Sandipan-Le Quang Liem, Subic Bay 2009.
11 Qxf3 Nxd4

12 Qd1

Question: Why passively retreat when we

have the g3-square available for the queen?
 Answer: The d1-square may actually be the queen’s best one, since she is vulnerable to ... Nf5, ...
Nh5 and ... Bd6 tempo-gaining ideas if she is placed on the g3-square. For example, after 12 Qg3 dxc4
13 Bxc4 0-0 14 Bg5 (14 Bh6?! Nf5 15 Qe5 Nxh6 16 Qxe7 Qd4! 17 Bb3 Nhg4! and Black takes over
the initiative, while retaining the extra pawn) 14 ... Bd6 15 Qh4 h6 16 Bxf6 Qxf6 17 Qxf6 gxf6 18 Re4
c5 19 Rh4 Kg7 20 Ne4 Be7 21 Ng3 f5!, White’s initiative has been exhausted and Karpov went on to
convert his extra pawn in the ending, E.Lobron-A.Karpov, Hannover 1983.
12 ... Ne6
Black blocks the e-file, while safeguarding his insecurely posted d4-
knight. Let’s look at the alternatives:
a) Kasparov mentions that 12 ... c6 13 Be3 Ne6 14 cxd5 Nxd5 15 Nxd5
Qxd5 16 Qc2 offers White “obvious compensation for the pawn” with the
bishop pair and development lead. 16 ... 0-0-0 is met with 17 Bc4 Qa5 18
Qe4 and even here, Black looks perfectly fine after 18 ... Bf6.
b) 12 ... dxc4 13 Bxc4 c5 14 Qa4+ Qd7 15 Qxd7+ Kxd7! (the only move,
but a good one which seems to solve Black’s troubles) 16 Be3 (after 16 Bxf7
Nc2 17 Rd1+ Kc6 18 Rb1 Rad8 19 Bg5 Rxd1+ 20 Rxd1 Rd8 21 Rxd8 Bxd8,
Black’s active king and queenside pawn majority more than make up for
White’s bishop pair) 16 ... Rhd8 17 Bxd4 cxd4 18 Bb5+ Kd6 (Black’s king is
surprisingly safe in the middle of the board) 19 Rad1 Kc5! 20 Re5+ Kd6 21
Ree1 Kc5 22 Re5+ and the players agreed to a draw, I.Smirin-A.Giri,
Khanty-Mansiysk 2010.
13 cxd5
Another try is 13 Bf5 dxc4 14 Qa4+ c6 15 Bxe6 (15 Qxc4 Nd4 16 Rd1
Qa5! and Black escapes his difficulties) 15 ... fxe6 16 Qxc4 0-0! (Black just
returns the pawn for a perfectly playable game) 17 Qxe6+ Rf7 18 Be3 Qd6
19 Qxd6 and a truce was declared, V.Anand-V.Kramnik, Mainz (rapid) 2001.
13 ... Nxd5 14 Bb5+
“I bring moral authority to this sadly uncivilized society,” declares the
bishop. This check is a little disruption trick from White. On the other hand,
14 Qa4+ isn’t dangerous for Black as after 14 ... c6 15 Nxd5 Qxd5 16 Bc4
Qd6 17 Bxe6 fxe6 18 Bf4 Qb4 19 Qxb4, and in this even position, a draw
was agreed, P.Wolff-P.Zarnicki, Buenos Aires 1997.
14 ... c6 15 Nxd5 cxb5
15 ... Qxd5?? would be a suicide-inducing blunder as after 16 Qxd5, 16 ...
cxd5 is an illegal move!
16 Bf4
 This move prevents immediate castling. Other options are:
a) 16 Qh5 0-0 17 Nxe7+ (17 Be3 Bg5! 18 Rad1 Bxe3 19 Rxe3 Rc8 and
Black stood no worse, B.Gelfand-A.Yusupov, Horgen 1994) 17 ... Qxe7 18
Qxb5 Rfd8 19 Be3 a6 20 Qb3 Rac8 and Black harmoniously completed his
development with equal chances, V.Ivanchuk-V.Kramnik, Dortmund 2008.
b) 16 Qb3 0-0 (you know the trick by now. Just return the extra pawn to
complete your development) 17 Be3 (Svidler declines to re-establish material
equality by chopping the b5-pawn) 17 ... Bc5 18 Rad1 Bxe3 19 Rxe3 Qa5 20
Qc2 Rfe8 21 f4 b4 22 f5 Nf8 23 Ne7+ Kh8 24 Qc4 (threat: Qxf7, followed by
mate on g8) 24 ... Ne6! (this interference trick solves all of Black’s
difficulties) 25 Nd5 Rad8 26 Ree1 Qc5+ 27 Qxc5 Nxc5 28 Rxe8+ Rxe8 29
Nxb4 g6 30 fxg6 and a draw was agreed, P.Svidler-B.Gelfand, Moscow
2009.
16 ... Nxf4
Two ways to go downhill:
a) 16 ... Bd6? fails to 17 Nf6+! Ke7 (17 ... Qxf6 18 Bxd6 Rd8 19 Qd5 a6
20 Rad1 is lost for Black) 18 Bxd6+ Qxd6 19 Nd5+ Kf8 20 Qh5 and White’s
enduring initiative is worth much more than Black’s extra pawn.
b) 16 ... 0-0? 17 Nxe7+ Qxe7 18 Bd6 picks off the exchange.
17 Rxe7+ Kf8 18 Re5
This was a suggestion of Karpov’s. Alternatively, 18 Re4 Qxd5 19 Rxf4
Rd8 20 Qc2 Qd2 looks fine for Black.
18 ... Qd6 19 Qd2!?

The attacking party strikes at the stillest part of the night. Naiditsch
unleashes a shocking new move and Black’s margin of safety seems to
constantly reduce.

Question: Is White getting enough compensation?

Answer: My Everyman editor IM Richard Palliser answers: “It may surprise subscribers (of
chesspub.com) that I haven’t given Naiditsch’s novelty an exclam or even two, but I’m simply not
convinced that it’s so good! Kramnik now thought for a long time, possibly due to over berating
himself for failing to spot 19 Qd2 at home, before avoiding the critical line”. The comps all say Black is
doing just fine, with a 0.00 evaluation.

Question: What is White’s point behind 19 Qd2?

Answer: The idea is to meet 19 ... Qxe5 with 20 Qb4+! when Black’s king enters a hurricane.
Getting the rook out of danger with 19 Rf5 offers White nothing as 19 ... Rd8 20 Ne3 (not 20 Qf3?? g6
21 Rxf4 Qxd5 22 Qxd5 Rxd5 23 Re4 Kg7, H.Koronowski-A.Thelen, Germany 199 and the trouble is
24 Re7 is met with 24 ... Rhd8 25 g3 R8d7 with an extra pawn in the ending for Black) 20 ... Qxd1+ 21
Rxd1 Rxd1+ 22 Nxd1 Ne6 23 Rxb5 b6 24 Nc3 Ke7 is just even and a draw was agreed,
R.Kasimdzhanov-A.Jussupow, Essen 2001.

Question: Can White try 19 Qd4?

 Answer: It’s too contorted. Black wins material with the short circuiting 19 ... f6! and just like that,
White is busted, U.Kaminski-J.Howell, German League 1992.
19 ... Ng6?
Before we are able to subjugate a problem, it first must be correctly
identified. Kramnik, still reeling from Naiditsch’s opening novelty, goes
astray. In his confusion, he embraces a mirage, declaring it to be truth. Now
the whole collapses due to a tiny fragmented element. Correct is 19 ... Qxe5!
(the sneaky shopkeeper rarely supplies her customers with the correct return
change) as this move gives him a perfectly playable game after 20 Qb4+ Ke8
(20 ... Kg8?? costs Black his queen and after 21 Ne7+, Black must hand over
his queen, since 21 ... Kf8?? is met with 22 Ng6+):
a) 21 Re1 Ne2+ 22 Kf1 Rc8! and White can win the black queen in
various ways, but 23 f4 Qxd5 24 Rxe2+ Kd7 25 Rd2 Rc5 26 Rxd5+ Rxd5 27
Qb3 (27 Qa3 “leaves Black even slightly better if anyone is, since his two
rooks are at least the equal of White’s queen,” writes Palliser) 27 ... Kd6 28
Qb4+ Kc7 29 Qc3+ Kb8 30 Qxg7 Rc8 31 Kf2 a6 32 h4 Rc2+ 33 Ke3 Rdd2
34 g3 Rxb2 35 Qxh7 Rdc2 36 Ke4 Rc4+ 37 Kf5 Rc5+ 38 Kf6 Rc6+ 39 Kg5
Rg2 40 g4 Rg6+ 41 Kf5 R6xg4 42 h5 is just equal, J.Olano Aizpurua-
F.Velilla Velasco, correspondence 2008. Of course such brilliantly
convoluted comp-generated correspondence analysis is well beyond human
perception over the board. So it was almost impossible for Kramnik to find
such an idea during the game.
b) Also, 21 Qxb5+!? Kd8 22 Rd1 looks crushing, but it’s merely an optical illusion and Black is just
fine after 22 ... Ne2+ 23 Kh1 Nd4 24 Qxb7 Rc8 25 h3 with even chances:
b1) After 25…g5, a draw was agreed in A.Gabrielian-V.Papin, Voronezh
2010. The truce was quite appropriate as 26 Qxf7 Rc5 27 f4 Qe6 28 Qg7
Rxd5 29 Qxh8+ Qe8 30 Qf6+ Qe7 31 Qh8+ Qe8 32 Qf6+ is perpetual check.
b2) If Black tries to win with 25 ... Rc5?, he runs into 26 Qa8+ Kd7 27
Qxa7+ Kd6 28 Nb4 Ke6 29 Nd3 Rc7 30 Qa3 Qg5 31 Re1+ Kf6 32 Qd6+
Ne6 33 Re5 Rd8 34 Rxe6+ fxe6 35 Qxd8+ Re7 and White has recouped his
investment and retains the initiative.
20 Ree1!
 Preparing to unravel with Rad1 next. Kramnik is fatally behind in
development and never recovers.
20 ... f6?!
Another defensive trip up. Black doesn’t have time for this leisurely ... Kf7
and ... Rhe8 unravelling plan. His last chance to unravel lay in 20 ... h5! 21
Rad1 Rh6! 22 Qd3 Ne5 23 Qxb5 Re6 24 Qxb7 Rae8 25 Ne3 Qb6 26 Qxb6
Rxb6. Black is down a pawn, but this is still his best drawing chance.
21 Rad1 Kf7
Black might also have plugged the e-file with 21 ... Ne5, but after 22 Qc3
he would have remained in huge trouble: for example, 22 ... Rd8 23 Qb3 Rd7
24 Nf4 and White’s pieces will soon invade.
22 Qe3!
White’s attack receives a much-needed infusion of troops and he sets up
the threat of Qb3!.
22 ... Rhe8?!
A better way to lose the queen is 22 ... Rhd8!. The secret to defence is
dogged persistence, coupled with a personality which refuses to get
discouraged under prolonged adversity. This position is so tough, that
Kramnik, one of the world’s premier defenders, is unable to find the best
path. After 23 Ne7 Qxe7 24 Qb3+ Kf8 25 Rxe7 Nxe7, at least in this version
Black gets to hang on to his b5-pawn, although this factor probably won’t
save him in the end.

To deliberately hold back help to a dying person, is in a way, murder (or in

this case, suicide) by default.

Exercise (combination alert): To criticize a player who blunders in

an already lost position is akin to a police officer reprimanding and
ticketing the corpse in an accident for speeding. White to play and win
material:

Answer: Clearance/discovered attack. Qb3+ is now unavoidable.

23 Ne7! Qxe7
The queen displays a shocking lapse of even rudimentary courtesy to her
visitor. However, 23 ... Rxe7 24 Rxd6 Rxe3 25 Rxe3 is equally hopeless for
Black.
24 Qb3+
White’s queen isn’t quite as democratic minded as she would like
everyone to believe, since her single vote counts for 51% of the sum total. “In
a society of equals, is it my fault that some of us are more equal than others?”
she asks, quite reasonably.
24 ... Kf8 25 Rxe7 Rxe7 26 Qxb5
In the 22 ... Rhd8! version, Black still kept this pawn.
 26 ... Rae8
Threatening a back rank cheapo. Kramnik has to pray for a fortress.
27 g3 Ne5 28 Kg2 Nc6 29 b4
Naiditsch soon begins to advance on both flanks to induce further
weakness in Black’s camp.
29 ... a6 30 Qb6 h6 31 a4 Ne5 32 Qc5 Kg8 33 b5 axb5 34 axb5
Now Black’s knight no longer has access to the c6-square.
34 ... Nf7 35 h4 Kh8 36 Rd2 Kg8 37 Kh3! Kh8 38 f4 Kg8 39 h5 Kh8 40
Qf5 Nd8

All Black can do is shuffle around and pray.

Exercise (planning): By now, Black’s coming loss is as predictable

as some celebrity’s misconduct being the lead story on CNN, over
actual newsworthy world events. How did White force the win?

Answer: A decisive invasion of the seventh rank.

41 Rd7! Ne6 42 Qd5! 1-0
Now if 42 ... b6 (42 ... Nc7 43 Rxe7 Nxd5 44 Rxe8+ Kh7 45 Rb8 b6 46 f5
is zugzwang for Black as the b6-pawn falls), 43 Rxe7 Rxe7 44 Qd6 Re8 45
Qd7 wins for White.
Summary: Black’s resources are perfectly playable after 10 Nc3!?. We can either accept the pawn
(we must know our theory deeply to enter this line and survive), or play the safer 10 ... 0-0, which has
the benefit of dodging a big chunk of theory. To survive this line, we must possess the danger-tendrils
of a nervous rabbit in a wolf-infested forest, so study it well.

Game 23
V.Topalov-B.Gelfand
Corus A, Wijk aan Zee 2008

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 Nc3

White agrees to a touch of structural damage to rid himself of Black’s

powerfully posted e4-knight.
8 ... Nxc3
“That which cannot be understood, must be destroyed,” thinks Black’s
knight, who hates all things unorthodox. Let’s examine other tries:
a) 8 ... Bf5 is a perfectly playable alternative, but eventually, we do need to
take the c3-knight, simply because White outnumbers us on coverage of the
e4-square. After 9 Re1 Nxc3 10 bxc3 Bxd3 11 Qxd3 0-0 12 Rb1 Na5
(clamping down on White’s c3-c4-break) 13 Bf4 Bf6 14 h3 c6 15 Nh2 Bg5
16 Be5 b6 17 Qg3 Nc4 18 Nf3 (18 Bxg7?? hoping for pin tricks on the g5-
bishop, fails miserably to 18 ... Bh4! (zwischenzug) 19 Qg4 f5 and White
hangs his bishop) 18 ... Bh6 19 Re2 f6 20 Bf4 Bxf4 21 Qxf4 Qd7 22 Rbe1
Rae8, Black stands structurally better and it is White who must think about
drawing the game, A.Naiditsch-A.Giri, Linares 2013.
b) I would avoid the passive 8 ... Nf6?! 9 h3! 0-0 10 a3 Re8 11 Re1,
S.Ganguly-R.Ramesh, Visakhapatnam 2006. This looks like a sour Exchange
French position for Black, since:
1. White has the useful extra moves a2-a3 and h2-h3.
2. White’s light-squared bishop is more actively posted than its c8-
counterpart.
c) The uncompromising 8 ... f5 weakens the e5-square. 9 Ne2 0-0 10 c4
Be6 11 Nf4 Bf7 12 cxd5 Bxd5, A.Grischuk-B.Gelfand, Bastia 2003. White
stands clearly better after 13 Ne5! (threat: Nxd5, followed by Bc4) 13 ...
Nxe5 14 dxe5 Nc5 15 Be2 Bf7 16 Be3 Ne6 17 Qc1! Nxf4 18 Bxf4. Black’s
position looks slightly uncomfortable since Rd1 is coming and White’s
kingside pawn majority looks superior to Black’s queenside model, since he
owns a passed e-pawn.
9 bxc3 Bg4
If he doesn’t develop this bishop now, White will curtail its activity with
h2-h3! next. Following 9 ... 0-0 10 h3! (cutting down on the c8-bishop’s
options) 10 ... h6 11 Re1 Bd6 12 Ne5 Qh4 13 Rb1 Nxe5 14 dxe5 Bc5 15 Qf3
b6 16 Be3 Bxe3 17 Rxe3 Be6 18 Rb4! (the rook is brought into the attack
horizontally) 18 ... Qe7 19 Rf4 c5?, now Gashimov uncorked 20 Rf6!! with a
winning attack. Black has no good answer and goes down after 20 ... Rfd8 21
Qh5 Kf8 (21 ... gxf6?? 22 Rg3+ Kf8 23 Qxh6+ Ke8 24 Bb5+ Bd7 25 Rg8+
and White mates) 22 Rg3! c4 23 Rxg7! cxd3 24 Rg8+! Kxg8 25 Rxh6 1-0,
V.Gashimov-B.Lalic, Cappelle la Grande 2007. Be warned…this line can be
a dangerous weapon for White in the hands of a skilled attacker.
10 Re1
White can also put the half-open b-file to use with 10 Rb1 Rb8 11 Re1 0-0
12 Bf4 Bd6 13 Bxd6 Qxd6 14 Re3 (14 Bxh7+?? fails to 14 ... Kxh7 15 Ng5+
Kg6 16 Qxg4 f5 17 Qh4 Rh8 18 Re6+ Qxe6 19 Qxh8 Qe7! 20 Qh7+ Kxg5
and White’s attack ends even before it begins) 14 ... f6 15 h3 Bh5 16 Qe2
Qd7 17 g4 Bf7 18 Nh4 Rfe8 19 Nf5 Rxe3 20 Qxe3 Re8 21 Qg3 Nd8! 22 h4
Bg6 23 h5 Bxf5 24 Bxf5, V.Anand-B.Gelfand, Monte Carlo blindfold (rapid)
2006. Black stands no worse after 24 ... Qc6 25 Qd3 h6 26 Bh7+ Kf8 27 Bg6
Re7 28 Qf5 Qd6 29 Qc8 c6 30 Kg2 b6.
10 ... 0-0 11 Bf4 Bd6
 Black challenges White’s grip over the e5-square. This is Black’s main
line, but it may be simpler to play 11 ... Bh5 which we look at next game.
12 Bxd6 cxd6?!
Gelfand incorrectly believes that necessity thrusts this uncomfortable
concession in his face.

Question: Why can’t Black recapture on d6 with the

queen, as in that Anand-Gelfand note from above?

Answer: This time, the idea fails for Black.

a) After 12 ... Qxd6?,
 White can pull off the Greek Gift sacrifice 13 Bxh7+! Kxh7 14 Ng5+ Kg6
15 Qxg4 f5 16 Qh4 Rh8 17 Re6+ Qxe6 18 Qxh8. GM Alex Baburin writes:
“The black rook is on a8 rather than b8 – and that makes a huge difference!
It’s amazing what including such mundane moves as Rb1 and …Rb8 can do
to a position!” Now Black loses material after 18 ... Rxh8 19 Nxe6 Rc8 20
Rb1 b6 21 Nf4+ Kf7 22 Nxd5 with two extra pawns. Black can avoid the
doubled, isolated d-pawn with the zwischenzug.
b) Black does have an improvement over Gelfand’s pawn recapture with
12 ... Bxf3! as Black decides that his bishop’s time and energy may be more
profitably applied on another track. Following 13 Qxf3 (13 Bxc7 Bxd1 14
Bxd8 Bxc2 15 Bxc2 Rfxd8 and Black isn’t worse, since the c3-pawn is every
bit as weak as the d5-one) 13 ... Qxd6 14 Re3 Rae8 15 Rae1 Rxe3 16 Rxe3,
S.Fedorchuk-Y.Solodovnichenko, LLucmajor 2014, I think White’s edge is
mainly symbolic. Black should be fine after 16 ... g6 17 h4 Nb8! 18 h5 Nd7
19 g3 c6 20 Kg2 b5. Now White must be on the lookout for future ... Qa3
raids. After 21 Qe2 Kg7 22 Re7 Rd8 23 f4 Kf8 24 hxg6 hxg6! 25 Re3 Kg7,
26 f5 is met with the bypassing 26 ... g5 27 Re7 Kf8 28 Re3 Kg7 and a
repetition draw is reached.
13 Re3!
 Dual purpose:
1. White prepares to strengthen his hold on the e-file.
2. The rook covers the f3-knight, allowing White’s queen to unpin.
13 ... Qd7
Other possible tries:
a) Black is unable to immediately challenge the e-file, due to the tactic 13
... Re8? 14 Rxe8+ Qxe8 15 Bxh7+!.
b) A decent continuation is 13 ... Na5 14 h3, L.Fressinet-B.Macieja,
Plovdiv 2008. I don’t think Black stands worse if he reinforces d5-pawn and
the c4-square with 14 ... Be6! 15 Rb1 Rc8 16 Qe2 Qf6 17 Nh2 Rfe8.
14 h3 Bh5
Weaker is 14 ... Bxf3? 15 Qxf3 and the d5-pawn comes under great
pressure. If 15 ... Ne7?, 16 Rae1 wins the d5-pawn, since Black walks into a
fatal pin after 16 ... Rfe8?? when 17 Bxh7+! wins, since 17 ... Kxh7 18 Qxf7
regains the piece and winning.
15 Qd2
Unpinning and keeping tabs on the e3-rook if Black challenges the e-file.
15 ... Rae8 16 Nh4!
White pieces begin to gather for a potential kingside assault.
16 ... Rxe3 17 Qxe3
Threat: Qg5, simultaneously attacking the h5-bishop and d5-pawn.
 17 ... Qe6?!
This is inaccurate. The position is fraught with blind-alley plans for Black.
He seems to be holding things together after the correct 17 ... Bg6! 18 Nxg6
hxg6 19 Qg5 Ne7 20 Re1 Re8 21 a4 a6 22 a5 Kf8 23 Rb1 Rc8 24 Rb3 Rc7
25 Qf4 Qd8 26 h4 Rd7 27 Ra3 (covering the a5-pawn. 27 ... Rc7 28 g3 Qd7
29 Rb3 Qe6 30 Rb6 Nc8 31 Rb3. After chasing his quarry for so long, the
trail grows cold for White, who still stands a shade better but progress won’t
come easily.
18 Nf5!
Threat: Qxe6, followed by Nxd6.
18 ... Rd8!?
Worth considering is 18 ... Qxe3 19 fxe3 Rd8 20 Rb1 b6 21 Rb5 Bg6 22
Rxd5 Bxf5 23 Rxf5 g6 24 Rf3 Na5 25 Kf2 Rc8 26 Rf6 Nc4 27 g4 Kg7 28 g5
when White is unable to hold on to his extra pawn, but still has winning
chances after 28 ... Nb2 29 Ke2 Rxc3 30 Be4.
19 Qg5

Threatening both mate and the h5-bishop.

19 ... Bg6 20 Ne3
Topalov decides to indulge in a bit of nuisance-mongering. Now the d5-
pawn comes under assault.
 20 ... Be4 21 Re1
This may allow Black to equalize. Instead, 21 Bxe4! dxe4 22 Rb1 b6 23
d5 Qe7 24 Qg4 Ne5 (24 ... Na5 25 Rb4 Re8 26 Nf5 Qf8 27 Rxe4 is even
worse for Black) 25 Qxe4 has White winning a pawn but his backward c-
pawns offer Black a target for counterplay.
21 ... h6 22 Qh4 Qg6
Stronger is 22 ... Bxd3! 23 cxd3 Re8 24 Qf4 Ne7 25 c4 dxc4 26 dxc4 b6
and I think Black should be able to hold the game.
23 Be2!
Rerouting the bishop to a more dangerous diagonal.
23 ... Bxc2
Natural, but not the best move. 23 ... Kf8 intending ... Qg5! gets Black is
pretty close to equality, if not already there.
24 Bh5
Stronger is 24 Bf3! when the weak d5-pawn’s defenders are seen sprawled
about in various attitudes of drunken repose. Now if 24 ... Be4 25 Nxd5,
Black should chop the knight since 25 ... Bxf3?? walks into 26 Qxd8+!
(removal of the guard/knight fork) 26 ... Kh7 27 Nf4 Qxg2+ 28 Nxg2 Nxd8
29 Ne3 and Black’s one pawn is insufficient compensation for the exchange.
24 ... Qe4
The queen dances about with the nervous vitality of a person on her third
triple espresso.
25 Qg3 Bd3?!
A better defence is 25 ... Bb1! 26 Kh2 Kf8 27 Bf3 Qg6 28 Bxd5 Qxg3+ 29
Kxg3 Bd3 and I think Black will save the game.
26 f3?!
The right pawn but the wrong square. Houdini gives White a clear edge
after 26 f4! Re8! (Black is forced to offer charity, possibly past his own
financial ability as 26 ... Kf8 27 f5 is strategically tough for Black) 27 Nc2
(the ingrate of a knight reminds us of the sulky terrier who bites the hand of
the veterinarian trying to heal him) 27 ... Bxc2 (if a starving man is offered a
bowl of soup with a dead fly floating in it, he will just eat around the
obstruction) 28 Rxe4 Bxe4 29 f5 Bxf5 30 Qxd6 Be6 31 Bf3 Rd8 32 Qc5 Rd7
33 c4! Ne7 34 cxd5 Nxd5 35 Qxa7 and I think White’s chances for a win are
higher than Black’s chances for a fortress draw.
26 ... Qe7??
 Gelfand may actually be okay after 26 ... Qh7! 27 Qf4 (27 Nxd5? walks
into the double attack 27 ... Qf5! 28 Nf4 Bc4 when Black stands better) 27 ...
g5! 28 Qg3 Ne7 29 f4 Be4! 30 fxg5 Qg7 31 Rf1 f5! 32 g6 b5 and I don’t
believe Black stands any worse here.

Exercise (combination alert): So far Black negotiated his defensive trials

reasonably well, but now he stumbles, getting distracted by a project
peripheral
to his overall goal of survival. We sense that a combination whispers to us,
just below the threshold of perception. Now comes the time for us to bring
it to the surface. Black’s last move walks into a forced loss. How?

Answer: Double attack/queen trap.

27 Ng4! Qg5
After 27 ... Qf8, 28 Nxh6+ Kh7 29 Nxf7 is crushing.
28 f4! 1-0
There is unspoken complicity in the murder plot between f-pawn, bishop,
knight and white queen as 28…Qxh5 29 Nf6+ forks king and queen.
Summary: Let’s not underestimate the 8 Nc3 line. White gets a sustained initiative for just a small
degree of weakening, in an Exchange French-like position.

Game 24
J.Smeets-B.Gelfand
Amber rapid, Nice 2010

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d4 d5 6 Bd3 Nc6 7 0-0 Be7 8 Nc3 Nxc3 9 bxc3 Bg4 10
Re1 0-0 11 Bf4 Bh5

In my opinion, this is slightly superior to the previous game’s 11 ... Bd6,

since it cuts down on White’s kingside attacking prospects.

Question: Why back up the bishop when White hasn’t even played h2-h3?

Answer: The idea is to transfer the bishop to the g6-square where it neutralizes White’s powerful
light-squared bishop. It also reinforces the king’s cover on the g-file, just in case White manages to lift
a queen to g4, or a rook to g3. This, in turn, keeps Black’s king safer than in the lines we looked at in
the previous game.
12 Rb1
After the unpinning 12 Re3:
a) 12…Bd6 13 Bxd6 Qxd6:
 Question: With White’s rook on the e3-square,
the h7-sacrifice is a great temptation. Is it sound?

Answer: The sacrifice doesn’t hold up under the comp’s glare.

a1) After 14 Bxh7+? Kxh7 15 Ng5+ Kg6 16 Qd2 Kxg5! (I admit that you
must be a very confident and capable calculator to work out such a line over
the board) 17 Rg3+ Kf6 18 Qg5+ Ke6 19 Qxh5 g6! 20 Re1+ Kd7 21 Qh3+
Kd8 22 Rf3 Qd7 23 Qh6 Re8 24 Rfe3 Rxe3 25 Rxe3 Kc8! 26 Qf8+ Nd8 27
Re7 Qc6 28 Re8 Qf6 29 Re7 b6 Black should be able to unravel and win.
a2) 14 Qe2 Nd8 15 Bf5, S.Karjakin-B.Macieja, Internet (blitz) 2004. Here Black can get away with
15 ... Ne6! since 16 Bxe6 fxe6 17 Rxe6 is met with 17 ... Bxf3 18 gxf3 Rae8! 19 Re3 (19 Rxe8 Qg6+
20 Kh1 Rxe8 and Black has the initiative) 19 ... Rxe3 20 fxe3 Qc6 21 Rb1 b6 22 c4 dxc4 23 e4 a6 and I
slightly prefer Black, since his king is less exposed than White’s.
b) If Black is ambitious, he can try the crazy-looking but effective
unplayed idea 12 ... g5!!:
b1) After 13 Rxe7! Qxe7 14 Bxg5 Qxg5 15 Nxg5 Bxd1 16 Rxd1 h6 17
Nh3 Na5 I think White will be the one fighting for the draw in the ending,
since his rook and pawn don’t quite measure up to full compensation for the
exchange.
b2) 13 Bg3? is met with 13 ... f5 14 Be5 g4 15 Ne1 Nxe5 16 Rxe5 Bd6!
when White seems to be in serious trouble.
12 ... Rb8
 Question: Which is a better way to deal with the attack
on the b7-pawn? Defending it with the text move or 12 ... b6?

Answer: Both are okay. It’s a matter of style. The text move creates no queenside pawn weakness,
but ties down Black’s rook, while 12 ... b6 is the opposite, creating weakness, while leaving the a8-rook
free to participate. After 13 Qe2 Bd6 14 Qe3 Re8! 15 Ne5 (15 Qxe8+ Qxe8 16 Rxe8+ Rxe8 17 Bxd6
cxd6 is just fine for Black, who gets play down the newly opened c-file) 15 ... Bxe5 16 dxe5 Qd7 17
Qg3 Qg4 18 c4 dxc4 19 Bxc4, Li Chao-Bu Xiangzhi, Tianjin 2011, White’s bishop pair is easily offset
by Black’s light square control and superior structure after 19 ... Qxg3 20 hxg3 Nd4.
13 Qe2
This is the main line but it seems odd to me, since it violates the principle:
When doubling major pieces on an open file, the queen should be behind the
rook. Other options are:
a) 13 h3 Bd6 14 Bxd6 Qxd6 15 Re3 f6 16 Qd2 Rfe8 17 Nh4 Rxe3 18
Qxe3 Re8 19 Qd2 Nd8 20 Nf5 Qd7 21 Qf4 Bg6 22 h4 Kh8 23 Qg4 Qf7 24
Qf3 b6 25 g3 Nb7 26 Kg2 Bxf5 27 Qxf5 g6 28 Qf3 Nd6 29 h5 g5 30 g4 Qe6
31 h6 Re7 32 Rh1 Kg8 33 Qd1 c6 34 Rh3 Qe1 35 Qf3 Qe6 36 Qd1 Qe1 37
Qf3 and the players agreed to a draw in a couple of moves, V.Topalov-
V.Kramnik, Nice 2008.
b) 13 Re3 Bg6 14 Qe2 Bd6 15 Bg3 a6 16 Nh4 Qd7 17 Nxg6 hxg6 18 Re1
b5 19 Qf3 Bxg3 20 Qxg3, T.Gharamian-N.Guliyev, Nancy 2013. The game
looks even after 20 ... Rfe8 21 Qh4 Rxe3 22 Rxe3 and 22…Qd8! doesn’t give
White time to load up on the h-file.
13 ... Bd6 14 Qe3 Qd7
Preparing to place a rook on the e-file.
15 Ne5!?
 This move, which violates the principle: the side who attacks should avoid
unnecessary swaps, is probably not such a great idea. White can also try:
a) 15 Bxd6 Qxd6 16 Nh4 Bg6 17 Nxg6 hxg6 18 h4 a6 19 a4 Na5 20 g3
Qc6 21 Qg5, L.Fressinet-Li Chao, Wijk aan Zee 2011. Here I would continue
with 21 ... Nc4 to bring the knight into the game. After 22 h5 gxh5 23 Qxh5
g6 24 Qh6 Rfe8 25 Kg2 Rxe1 26 Rxe1 Re8 27 Rh1 Qf6, Black is just fine.

Question: Is 15 Ng5 a waste of time for White?

Answer: White doesn’t have anything better to do and the move does reroute his knight to the more
active f4-square.
b) 15 Ng5 h6 16 Nh3 Bg4 17 Qg3 Bxh3 18 Bxd6 cxd6 19 Qxh3 Qxh3 20
gxh3 Rfd8 21 Rb5 Kf8 22 Reb1 b6 23 Rxd5 Rbc8 and Black’s play down the
c-file, combined with White’s weakened pawn structure, gives him fully
adequate compensation for the pawn. Black threatens to play ... Rbc8 and ...
Na5, which ties White up. Following 24 c4 Ke7 25 Re1+ Kf8 26 Rb1 Ke7 27
Re1+ Kf8, the players agreed to a draw, E.Bacrot-Wang Yue, Nanjing 2010.
15 ... Bxe5 16 dxe5!?
Very risky. Determination to get the job done – but without the necessary
resources to accomplish it – reaches only the 50% threshold of the equation.
White continues to further weaken his structure without the slightest display
of remorse. This move looks like it is made with the philosophy: there is no
such thing as an “honorary” member of an attack. Either we are in, or we
are out, and damn the structural cost. I’m not sure if this is an expression of
rage or exultation. Smeets isn’t interested in the boring line 16 Bxe5 which is
an admission that White got absolutely nothing from the opening. Here we
sense the fist-unclenching release of the position’s tension. After 16 ... Rfe8
17 Qg3 Nxe5 18 Rxe5 Rxe5 19 Qxe5 Bg6 20 Bb5 Qd8 21 Bd3 Qd7 22 Bb5 a
repetition draw is reached.
16 ... Bg6
Principle: When your opponent has the bishop pair, swap one of the
prelates off.
17 Bxg6 hxg6 18 Rbd1 Qe6

Question: How would you evaluate the position?

Answer: The comp says this position is dead even, while I prefer Black:
1. His structure is clearly superior.
2. Black will have a monster knight if given a chance to play ... Na5 and ...
Nc4.
3. Black has a grip on the light squares.
4. I don’t believe White’s kingside attacking chances fully compensate for
his strategic deficiencies.
19 Bg5
The bishop continues to connive plots, mostly illegal, immoral or a
combination of both. With this move, White disallows …Rb8, and in doing
so, weakens the d5-pawn.
19 ... Rbe8
Recent unauthorized activity nourishes the black king’s suspicions, so he
doubles his guard, ordering both rooks close to him.
20 Qd2?!
White’s queen abnegates her responsibility, abandoning the e5-pawn, the
way my wife assigns to me the brunt of the candy-giving (to both deserving
young and undeserving college-aged trick-or-treaters) every Halloween night.
The idea is to swap the e5-pawn for Black’s d5-pawn. I think White has
better drawing chances in the line 20 f3! and if 20 ... Qxe5 (20 ... b6 21 Qd2
Nxe5 22 f4 f6 23 Bh4 Qg4 24 fxe5 Qxh4 25 Qxd5+ Rf7 26 e6 Rfe7 27 Re4
Qh5 28 c4 should be okay for White) 21 Qxe5 Nxe5 22 Bf4 Nc4 23 Rxe8
Rxe8 24 Rxd5 c6 25 Rd7 Re1+ 26 Kf2 Rb1 27 Ke2 a5 28 Rc7 Na3 29 Kd3, I
think White should hold the draw without too much trouble.
20 ... Nxe5

21 Bf4
Already, White must be very careful:
 a) Not 21 Qxd5??, which is a horrible blunder. Black wins after 21 ...
Qxd5 22 Rxd5 Nf3+.
b) The howler 21 f4? is met with 21 ... Qb6+! when White must swap
queens and avoid 22 Kh1? Nc4 when the d5-pawn is taboo due to 23 Qxd5??
Ne3.
21 ... Qf5 22 Bxe5 Rxe5 23 Rxe5 Qxe5 24 Qxd5 Qxc3
The frowning queen, deeply dissatisfied with how the universe is run,
ponders deeply on the changes she plans to make. Black’s queen is destined
to pick off more pawns than White’s.
25 Qxb7 Qxc2 26 Qb3 Qxb3
Swapping queens is Black’s best shot at a win.
27 axb3

So in the aftermath, Black is up a pawn in a rook and pawn ending, with

reasonable winning chances.
27 ... Rb8 28 Rd3
Not 28 Rc1?? Rxb3 and White is unable to capture the c7-pawn due to the
back rank mate problem.
28 ... Kf8 29 Rc3 Rb7 30 Re3
Cutting off Black’s king. When under pressure and on the defensive, our
single joy in life is to frustrate the opponent’s desires – even if in a trivial
way. Instead, after 30 Kf1 Ke7 31 Ke2 Kd6 32 Kd3 c5 33 Kc2 Re7 34 Kd2,
Black continues to make progress.
30 ... Rb6 31 Kf1 Re6 32 Rc3

Question: Does White have any chance to hold a draw in the king and pawn ending?

Answer: Not a chance, if Black exercises reasonable care. Sometimes we must choose between two
lines which lose anyway. The trick then is to pick the one containing the least pain for ourselves and the
greatest potential for error from our opponents. The king and pawn ending would be easier for Black to
win than the rook and pawn ending which occurred in the game after 32 Rxe6?! fxe6 33 Ke2 Ke7 34
Kd3 Kd6 35 Kc4 Kc6 36 f4 a5 37 g3 Kb6 38 Kd4 Kb5 39 Kc3 (39 Ke5 Kb4 40 Kxe6 Kxb3 41 Kf7 a4
and it’s not even close) 39 ... Kc5 (Black’s king now precedes his c3-competitor by a full rank) 40 h3
Kb5 41 h4 Kc5 42 g4 Kd5 43 Kd3 e5 44 fxe5 Kxe5 45 Ke3 c5 and White is in zugzwang. With 46
Kd3, the white king bemoans, “I live in a world which rewards fools, while the wise – namely, myself
– are casually brushed aside,” with obvious sour grapes at his fast-rising brother. After 46 ... Kf4 47 g5
Kg4 48 Kc4 Kxh4 49 Kxc5 Kxg5 50 Kb5 Kf5 51 Kxa5 g5 52 b4 g4 53 b5 g3 54 b6 g2 55 b7 g1Q 56
b8Q, White’s planned draw is doomed to be wrecked on the shoal of the position’s inherent
incompatibility with the idea and following 56 ... Qa1+ 57 Kb6 Qb2+, queens come off the board,
except that Black gets to make a new one with his g-pawn.
32 ... c6 33 Rc4 Ke8 34 Rd4 Re5
Both sides attempt to restrict the other’s king.
35 h4 Ke7 36 g3 Ke6
If Black’s king can’t come across, then he will go up the board instead.
37 Kg2 Rb5
Principle: In rook endings, tie the opponent’s rook to a weak pawn.
38 Rd3
 Note that both the black rook and king now clearly outshine their d3- and
g2-counterparts.
38 ... a5 39 g4 Rd5 40 Rf3 Rd4! 41 Kg3 c5 42 Re3+ Kd7 43 f3 Rb4 44
f4 Kc6
Black needs his king to support his future passed queenside pawn.
45 h5 gxh5 46 gxh5 Rb7 47 Kg4 Kd5 48 Kf5 Kd4 49 Rg3
Point B may not be a great distance from point A, as the crow flies. But if
we are left without a direct path, the trip may still require a long time. White
makes progress on the kingside but it’s too slow in comparison to his
opponent’s queening attempts on the other side of the board.
49 ... f6 50 Kg6
Also insufficient is 50 h6 gxh6 51 Kxf6 Rb6+ 52 Kf7 a4! 53 bxa4 c4 54 f5
c3 55 Rg4+ Kd5 56 Rg1 Ke5 57 a5 Rc6 58 Rf1 c2 59 Rc1 Kxf5 60 a6 h5 61
a7 Rc7+ 62 Ke8 Rxa7 63 Rxc2 Ra4 64 Kf7 h4 when Black wins, since
White’s king is unable to help halt the passed h-pawn.

The years have not treated White’s king kindly, as we observe hollow eye
sockets and a pasty complexion, ravaged by wrinkles caused by stress and the
passage of time.

Exercise (planning): We sense the inevitability of an ancient Greek tragedy

encircling White’s position. Come up with a clear winning plan for Black.
 Answer: Pawn sacrifice/creation of a passed pawn.
50 ... a4!
Black king, rook and queenside pawns cement their arrangement with
conspiratorial nods at the sight of White’s rook, who is inadequate in his role
as defender. They agree that the lowly a-pawn must sacrifice himself to get
into the history books and add significance to an otherwise inconsequential
life. It’s a mistake to avoid sacrificing a pawn here with 50 ... c4? 51 bxc4
Kxc4 52 f5 a4 53 h6 gxh6 54 Kxf6 Ra7 55 Ra3 Kb4 56 Ra1 a3 57 Kg6 a2 58
Kxh6 Kb3 59 f6 Kb2 60 Rxa2+ Kxa2 61 Kg6 and White draws.
51 bxa4 c4
The c-pawn is too fast.
52 f5 c3 53 Rg1
“To presume to a social status above your natural station is laughable
indeed,” White’s rook chides Black’s ambitious c-pawn, who isn’t
discouraged in the least.
53 ... c2 54 Rc1 Kc3 55 h6 gxh6 56 Kxf6 Kd2 57 Rh1 0-1
Summary: 11 ... Bh5 looks like an easier path to full equality than the
previous game’s 11 ... Bd6.
Chapter Four
Fifth Move Alternatives
In this chapter, we cover White’s numerous fifth-move dodges of the Old
Main Line. Some of them are bizarre, and clearly intended to throw us off
theoretically, like the following diagrams:

The idea here is to cut off Black’s natural ... d6-d5 (which it usually
doesn’t). White can even follow with Nc3 and when we swap, play dxc3,
with a position similar to the games we see next chapter, but to my mind, an
inferior version for White.
 This borrows an idea from the Kopec Sicilian. White inverts the normal
order and plays Bd3, blocking the d-pawn. The idea is to reach a normal
looking position with c2-c3, Bc2 and d2-d4 next.

Then we must familiarize ourselves with a couple of very drawish lines,

neither of which pops up at us with the force of a newly opened bottle of
champagne.
 In this one, White kicks our knight back and will play d3-d4 next, reaching
a dull, symmetrical line of the Exchange French, with White on the move.
Neither this line nor the next one can be accurately described as a cash cow
for White. Equality is not an issue. Winning with Black is, since White
wilfully drains the position of dynamism. If we are happy with a draw (as is
really the goal of our opening), then there are no worries.
 A law in one nation can be contradicted as a crime in another. This one is
even more obnoxious. White wants the symmetrical Exchange French line,
but with queens off the board, enhancing the position’s drawing potential.

In all cases, we are on theoretically solid ground and not a single line gives us any real problem.

Game 25
S.Dvoirys-S.Mamedyarov
Aeroflot Open, Moscow 2004

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 c4

IM John Watson told me this move is quite dangerous for Black, while
GM Konstantin Sakaev is dismissive of the move and writes: “This is an
original move, but that’s about the most positive thing that can be said about
it.” I think the move is dangerous only if Black gets caught by surprise.
Otherwise, I’m with Sakaev.

Question: Why did White play c2-c4 rather than d2-d4?

Answer: For a couple of reasons:

1. By playing c2-c4 first, White inhibits ... d6-d5, a move which normally
helps support the e4-knight.
2. White may just skip d2-d4 altogether and play Nc3, allowing ... Nxc3
and dxc3, which is a structure similar to the Nimzowitsch Variation, our next
chapter in the book.
5 ... Nc6
This is given an exclam by Sakaev. This move is considered slightly more
accurate than 5 ... Be7, since it dissuades d2-d4 from White.
6 d4!?
He plays d2-d4 anyway! Sakaev frowns on this move, giving it a ‘?!’
mark. Instead, 6 Be2 is too meek to hope for an advantage. After 6 ... Be7 7
0-0 0-0 8 d4 Bf6 (this looks like a rather nice version of an Alekhine’s
Symmetrical Exchange line for Black who threatens to add more heat to the
d4-pawn with ... Bg4 next) 9 d5 (9 h3 Bf5 isn’t very inspiring for White
either) 9 ... Ne7!? (Giri spurns the natural e5-square and his knight heads for
the g6-square instead) 10 Nd4 Re8 11 Na3 (awkward, but 11 Nc3 allows
Black to double White’s c-pawns, while 11 Nd2?? loses on the spot to 11 ...
Nxf2) 11 ... Nf5 (Black already stands slightly better) 12 Nac2 Nc5 13 Nb3,
it’s a bad sign if you own the White pieces, yet are looking for exchanges to
equalize, L.Nisipeanu-A.Giri, Wijk aan Zee 2010. Black looks slightly more
comfortable after 13 ... Na4 14 Rb1 Bd7.
In the next game, we will look at 6 Nc3.
6 ... d5
For a long time this was thought to be at least equal for Black, but
Dvoirys’ next move offers our side a new challenge.
7 c5!?
 The Petroff is not an opening normally conducive to roaming our private,
experimental caprices. Most lines must be followed with military obedience
to theoretical norms. Yet this move may be an exception to the rule. Now the
player’s interests run counter-parallel. This move, which may alter the
previous negative evaluations, isn’t mentioned by Sakaev or Karpov in their
Petroff books (in fact, the move was still an unknown TN when Karpov
wrote his book). White seizes queenside space at the cost of time. Karpov
gives 7 Nc3 Bb4 when it is White is fighting for equality. After 8 Qc2 Qe7!
(this move disrupts White’s developmental flow) 9 Be3 Bf5 (threat: ... Ng3,
with a double attack on White’s queen and also the h1-rook) 10 Qc1?! (White
had to take his chances with 10 Bd3 dxc4 11 Bxc4 Ng3 12 Qa4 Nxh1 13 d5
Be4 14 dxc6 Bxc6 15 Bb5 Bxb5 16 Qxb5+ c6 17 Qc4 0-0 18 0-0-0 Nxf2 19
Bxf2 Rfe8 with advantage to Black, who has the safer king while White’s
knight pair looks less effective than Black’s rook and two pawns) 10 ... Nxc3
11 bxc3 Ba3 12 Qd2 (12 Qd1 is met with 12 ... Bb2) 12 ... Nb4!, White’s
position was close to complete collapse, V.Kupreichik-A.Mikhalchishin,
Kujbyshev 1986.
7 ... Bg4
Black logically applies pressure to a defender of the d4-pawn. Also
playable is 7 ... Be7 8 Bb5 Bd7 (Black can dare White to swap on c6 with 8
... 0-0 9 Bxc6 bxc6 10 Qa4 Bf6 and if White gets greedy with 11 Qxc6!?,
Black seizes the initiative after 11 ... Rb8 12 0-0 Rb4!? 13 Bf4 Rxb2 14 Bxc7
Qe8 15 Qxd5 Bb7 16 Qc4 Nxf2! 17 Rxf2 Qe3 18 Nbd2 Bxf3 when White is
under heavy pressure) 9 0-0 0-0 10 Re1 Bf6 11 Bf4 Nxd4!? (11 ... a6 also
looks fine for Black) 12 Nxd4 Bxd4 13 Qxd4 Bxb5 14 f3, S.Dvoirys-J.Bick,
Biel 2003. The position looks balanced after 14 ... Ng5 15 Bxg5 Qxg5 16
Nc3 Bc6 17 Re5 Qg6 18 Nxd5 Bxd5 19 Rxd5 Rae8 20 Re5.
8 Bb5 Qf6!
Covering the c6-knight, while pressuring the one on f3 and the d4-pawn.
9 Nc3
Instead, 9 h3?! looks like a waste of a tempo. After 9 ... Bxf3 10 gxf3 Ng5
11 Bxg5 Qxg5 12 Nc3 (White may try 12 Qa4 which is met with 12 ... 0-0-0!
13 Bxc6 bxc6! 14 Nc3 Re8+ 15 Kf1 Re6 when I slightly prefer Black’s
chances) 12 ... Be7 13 Qa4 0-0 14 Bxc6 bxc6 15 Qxc6, I.Jakic-Z.Licina,
Opatija 2012. Black should continue 15 ... Rae8 16 Qxd5 Bxc5+ 17 Kf1
Qxd5 18 Nxd5 Bxd4 19 Nxc7 Rc8 20 Rd1 Bxb2 21 Nd5 Rc2. Only Black
has chances to win in this ending.
9 ... Bxf3 10 Qxf3!
“Please sit. Perhaps you would care for a glass of wine?” White’s queen
asks her sister, as she surreptitiously adds a few drops of hemlock into the
glass. White offers the d4-pawn for a dangerous development lead. However,
the evaluation remains even. White’s last move is far more dynamic than the
lacklustre 10 gxf3 Nxc3 11 bxc3 Be7 when if anyone is better, it is Black.
10 ... Qxd4
When we grab material like this at the cost to our development, our sense
of avarice overpowers our fear.
11 0-0 0-0-0
“Perhaps Black should play the cold-blooded 11 ... Nxc3 12 bxc3 Qe4
when White still has to demonstrate what compensation he has,” writes GM
Nigel Davies.
 Question: Do you agree?

Answer: This pawn grab looks pretty dangerous to me. I prefer White after 13 Qd1 Qf5 14 Re1+
Be7 15 Ba3 Qd7 16 Qe2. White’s queen, who is gifted in equal parts with good looks, an unlimited
cash flow and the power of life and death over her subjects, doesn’t care a counterfeit dollar about her
e8-brother’s health and wellbeing. After 16 ... Kf8 17 Qf3, White has loads of compensation and I
wouldn’t be tempted to take on Black’s defensive task.
12 Be3 Qf6
An untried idea is 12 ... Qe5!? when after 13 Bxc6 (13 Rfe1 Nd4 14 Bxd4
Qxd4 15 Rad1 Qxf2+!? 16 Qxf2 Nxf2 17 Kxf2 Bxc5+ 18 Kf3 c6 and Black
gets three healthy pawns for the piece in the ending, which looks
approximately balanced) 13 ... bxc6 14 Rfe1 Re8 15 Nxe4 Qxe4 16 Qh3+
(Black stands better after 16 Qxf7? Bxc5) 16 ... Re6 17 Bd2 Qg6 18 Qa3
Kb7, Black looks okay.
13 Qg4+ Qe6
The queen tacitly acknowledges her sister’s ominous presence and takes
preventative measures. Otherwise, after 13 ... Kb8 14 Nxe4 dxe4 15 Rad1 (15
Qxe4 Nd4 16 Bc4 Bxc5 and I don’t see full compensation for White) 15 ...
Nd4 16 Rxd4 Rxd4 17 Bxd4 Qxd4 18 Rd1 Qf6 19 b4 e3 20 fxe3 g6 21 Qf4
Bg7 22 Qxf6 Bxf6, the position appears to be heading for a draw.
14 Qxe6+ fxe6 15 Nxe4 dxe4 16 Bxc6 bxc6
 Question: Isn’t Black strategically lost with the burden of five isolanis?

Answer: Menace is that entity which is invisible and inaudible, yet tangibly real. Yet here I don’t
see it for Black. I sense an arising of both hope and foreboding. Davies thinks Black is in trouble, while
I think Black is perfectly fine in the coming ending, for the following reasons:
1. First, Black is up a pawn, so it costs White time to win one back.
2. Black can set up a d5-outpost for his rook with ... Rd5.
3. White’s c5-pawn is fixed on the wrong colour of White’s remaining
bishop, so Black is the one with the better bishop. If White backs up the c5-
pawn with b2-b4, then Black counters with …a7-a5, which reduces his
isolani population.
4. Black’s king can be easily centralized by ... Rd5, ... Kd7, ... e6-e5 and ...
Ke6.
Now saying all this, I am reminded by your words: “the burden of five
isolanis,” which means matters can easily sour for Black if he isn’t careful.
17 Rac1
Intending Rc4 and Rxe4, equalizing the material.
17 ... Rd5 18 Rc4 Be7 19 b4
White can try an attempt at improvement with 19 b3, which allows him
future Ra4 ideas. Following 19 ... Rhd8 20 h3 a5 21 Rfc1 Bg5 22 Bxg5 Rxg5
23 Rxe4 Rd2 24 Rxe6 Rxa2 25 Rc3 g6 26 Re7 h5 27 Re6 Kb7 28 h4 Rf5 29
Rxg6 Rfxf2 30 Rg5 Rfb2 31 Kh2 Rd2 32 Rf3 Rd5 33 Rff5 Rd4 34 Kh3, the
players agreed to a draw, E.Raby-K.Owen, correspondence 2008. The game
is even after 34 ... Rd3+ 35 Rf3 Rdd2 36 Rxh5 Rxg2 37 Rh7 Rg1 38 Rff7
Rag2 39 Rxc7+ Ka6 40 Rxc6+ Kb5 41 Rcc7 Rg3+ with perpetual check.
19 ... Kd7 20 Rxe4 a5!?
This appears as a pro forma courtesy, rather than any real contrition for his
past adoption of so many isolanis. Mamedyarov rids himself of his isolated a-
pawn and gives White a few isolanis of his own.
21 bxa5 Ra8 22 Ra4 Bxc5 23 Bxc5 Rxc5 24 a6 Rb5
The rook goes after White’s a-pawn.
25 Rd1+ Ke7 26 g3 c5
After 26 ... Rb6 27 a7 Rb7 28 Rg4 g6 29 Rc4 Raxa7 30 Rxc6 Rxa2 31 Re1
Kd7 32 Rcxe6 Rb1, only Black can win, since his outside passed c-pawn is
still a bit of an issue for White.
27 Rd3 Ra7 28 Kg2 Rb6 29 Rda3
White hangs on to his a6-pawn, but in doing so, ties up his rooks. I think
Black actually stands better now.
29 ... Kd6 30 Kf3 Kc6 31 Ra5 Rb5
The a6-pawn falls if one pair of rooks comes off the board.
32 Ke3
After 32 R5a4 Rb4 33 Ra5 Kb6, White’s rooks feel tied up.
32 ... Rxa5 33 Rxa5 Kb6 34 Ra4 Rxa6 35 Rxa6+!
It takes confidence in one’s calculation ability to play such a move.
35 ... Kxa6
 King and pawn endings are tricky things, where we must carefully sort
through our promotion itinerary, working out various routes, layovers and
connecting flights to our 8th rank destinations. Black is up a pawn in the king
and pawn ending, which is normally fatal for the pawn down side. However,
White’s outside passed a-pawn offers him equality.
36 Kd3 Kb5 37 Kc3 e5 38 f3 c4 39 h3 g6 40 h4 h6
 Exercise (planning): How did White force a draw?

Answer: Pawn sacrifice/deflection. White sacrifices the outside passed a-pawn to deflect Black’s
king.
41 a4+! Kc5
Black’s king is unable to hide the peevishness in his voice, since he no
longer can win the war. Taking the pawn with 41 ... Kxa4 is no better as 42
Kxc4 Ka3 43 Kd5 Kb4 44 Kxe5 c5 45 h5! (creation of a passed f-pawn
insures the draw) 45 ... gxh5 46 f4 c4 47 f5 c3 48 f6 c2 49 f7 c1Q 50 f8Q+
Qc5+ 51 Qxc5+ Kxc5 52 Kf5 Kd5 53 Kg6 Ke5 54 Kxh6 h4 55 gxh4 Kf6 is
drawn.
42 h5!
White continues to gambit pawns left and right, as if the game was a
King’s Gambit, rather than a king and pawn ending!
42 ... gxh5
Also leading to parity is 42 ... g5 43 a5 Kb5 44 a6 Kxa6 45 Kxc4 c6 46
Kc5 Kb7 47 Kd6 g4 48 fxg4 e4 49 g5 e3 50 gxh6 e2 51 h7 e1Q 52 h8Q
Qb4+! (52 ... Qxg3+?! 53 Qe5 offers White some chances, since his h-pawn
is faster than Black’s c-pawn) 53 Ke6 Qg4+ 54 Kf7 Qf3+ and the game ends
in perpetual check.
43 a5 Kb5 44 a6
I insist. The a-pawn considers herself a second missing link, genetically
superior to all other pawns on the board. So naturally, common strictures of
law and morality are clearly inapplicable to her. “If I am the law itself, then
how is it possible for me to commit a crime?” she asks.
44 ... Kxa6 45 Kxc4 Ka5 46 Kd5 Kb4
The trouble with this move is that when White promotes his pawn, it will
be with check.

Question: Does that mean that Black may win with 46 ... Kb5?
 Answer: No. White can force Black’s king to the b4-pawn like this: 47 Kxe5 c5 48 f4 c4 49 Kd4!
(White’s king patiently explains the facts to his brother, as if to an obtuse child) 49 ... Kb4 (step by
unwilling step, Black’s king is induced to the unfavourable b4-square) 50 f5 c3 51 f6 c2 52 f7 c1Q 53
f8Q+ with a very likely draw.
47 Kxe5 c5 48 f4 c4 49 f5 c3 50 f6 c2 51 f7 c1Q 52 f8Q+

Exercise (calculation): Can Black force a draw by meeting White’s last

 check, with 52 ... Qc5+? Work out the king and pawn ending, and
evaluate.

Answer: The move works and is the safest path to the draw.
52 ... Qc5+! 53 Qxc5+ Kxc5 54 Kf5
Black’s kingside pawns are as defenceless as new-born sparrow chicks,
left unprotected in the nest from a predator eagle. The hoped-for crescendo
turns out anticlimactically, since the game is now obviously drawn.
54 ... Kd5 55 Kg6 Ke4 ½-½
After 56 Kxh6 Kf3 57 Kxh5 Kxg3, I’m nearly certain this is a theoretical
draw!
Summary: 7 c5 turns a formally inferior line into a playable one for White, but there is no serious
danger for our side if we are familiar with the analysis.

Game 26
A.G.Giaccio-P.Zarnicki
Buenos Aires Zonal 2000

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 c4 Nc6 6 Nc3

White hopes to get a position similar to the one reached in the next chapter
of the book.

Question: What is the difference?

 Answer: In this version White’s c-pawns are pushed up one square, to the c4- and c3-squares, rather
than on the c2- and c3-squares.

Question: Isn’t that better for White, since now he influences the d5-square?

Answer: I think it’s actually slightly inferior to the next chapter’s version, for the following reasons:
1. The d3-square is weakened.
2. In the next chapter, we often see White castling queenside, while here it
would be much riskier, since White’s queenside pawn front has been
loosened.
6 ... Nxc3 7 dxc3
7 bxc3 is much less common but still playable. White follows the
principle: Capture towards the centre. In doing so White gets an open b-file
and exerts more central influence than in the 7 dxc3 line. On the downside,
White creates three pawn islands and his central pawn complex is more
vulnerable in the ending. After 7 ... Be7 8 d4 Bf5:
a) 9 Be2 0-0 10 Bf4 Bf6 11 0-0 Na5 12 Re1 Re8 13 Qa4 b6, K.Piorun-
D.Fridman, Wroclaw 2009, the position is about equal.
b) 9 Bd3 Qd7 10 0-0 0-0 11 Rb1, K.Kiik-E.Turunen, Helsinki 1990. The position looks balanced
once again after 11 ... b6.
7 ... Be7
Black’s main move. Let’s look at the alternatives:

Question: Why couldn’t Black preempt White’s possible Bd3 with 7 ... Bf5 instead?

Answer: I think the bishop trade offer is very playable.

a) After 7 ... Bf5 8 Bd3 Qd7 9 0-0 Be7 10 Bxf5 Qxf5, K.Miton-
V.Gashimov, Havana 2007, I don’t believe White can extract much with 11
Re1 0-0 12 Qd5 Qc2! 13 Bf4 Rab8! 14 Rab1 Rfe8 15 c5 dxc5 16 Bxc7 Ra8
and White’s edge is not enough here to really worry Black.
b) Sakaev favours 7 ... g6 which is tricky, since it creates dark square
weakness in Black’s camp. Also, Black’s bishop will fianchetto into a c3-
wall. Following 8 Bg5 (one point of Sakaev’s idea is that 8 Bd3?! is met with
the disruptive 8 ... Qe7+! when White has nothing better than the meek
retreat 9 Be2, since 9 Be3?! is met with 9 ... Bh6!) 8 ... Qd7 9 Qd2 Bg7 10 0-
0-0 0-0 11 h4 h5 12 Bh6 a6 13 b3 Qf5 14 Bxg7 Kxg7 15 Kb2 Re8 16 Bd3
Qf6 17 Ng5 Bf5 and if given a choice, I would take Black’s position, since I
see no real possibility of an attack from White, A.Vovk-A.Montanes Arribas,
Madrid 2012.
8 Bd3
Non-critical is 8 Be2 which poses no problems for Black after 8 ... 0-0 9 0-
0 Re8 10 h3 Bf6 11 Be3 b6 12 Nd4 Nxd4 (12 ... Ne5 is also possible) 13
cxd4 Bb7 14 Qd2 h6 15 Rfe1, Peng Xiaomin-V.Ivanchuk, Elista 1998. Black
looks completely equal after 15 ... Be4 where White’s extra central space is
more a symbolic advantage than a factual one.
8 ... Bg4
Now that White committed his bishop to d3, this pin is logical.
9 Be4
White’s main move. He covers the f3-knight and aims his bishop at the d5-
square and c6-knight.
9 ... 0-0
Black can also prepare the annoying neutralizing move ….Bf5 with 9 ...
Qd7 when 10 Be3 Bf5 11 Bd5 Bf6 12 0-0 0-0 13 Re1, V.Durarbeyli-
M.Melnichuk, Kemer 2009 is about equal. Here Black can play 13 ... Ne7! 14
Bd4 (14 Bxb7 Rab8 15 Ba6 Rxb2 16 Bd4 Ng6 17 Bxf6 gxf6 18 Nd4 Ne5 and
Black looks okay in this messy position) 14 ... Nxd5 15 cxd5 Be7 16 c4 c5 17
Bc3 (17 dxc6 bxc6 is riskier, since it opens the game for Black’s bishop pair)
17 ... Rfe8 when Black’s king is absolutely safe and he can prepare a ... b7-b5
break.
10 Qc2 f5
The text move is a more aggressive response than 10 ... h6 11 Be3 Be6 12
b3 Bf6 13 0-0-0 a5 14 a4 Ne7 15 h4! Qc8? (15 ... Re8 may still be okay for
Black) 16 Bg5! and Black was in deep trouble, H.Stefansson-I.Sokolov,
Reykjavik 2000.
11 Bd5+ Kh8 12 Bd2
 A new move, which prepares long castling. Other tries:
a) White can also play the safer 12 0-0 Bxf3 13 Bxf3:
a1) 13 Bg5! (principle: When your opponent has the bishop pair, eliminate
one of them) 14 Re1 Ne5!? (14 ... Bxc1 15 Raxc1 Ne5 16 Bxb7 Rb8 17 Rcd1
Qg5 18 Bd5 c6 19 Rxe5 dxe5 20 Bxc6 e4 when this one is hard to evaluate.
White has two pawns for the exchange and a queenside super-majority, while
Black has chances on the kingside. I think I will take the gutless route and
declare the position unclear) 15 Bxb7 Rb8 16 f4 Bxf4 17 Bxf4 Ng6 18 Qf2
Rxb7 19 b3 Qf6. I don’t think Black stands worse.
a2) 13 ... Ne5!? 14 Bxb7 Rb8 15 Bd5 c6, Repina,V-Oganessian,R Moscow
2009. Here White can seize the advantage with 16 Be6! f4 17 Rd1 f3 18 g3
Qe8 19 Qe4. Black’s position is loaded with pawn weaknesses and I don’t
believe in his compensating attacking chances.
b) 12 Bf4 Rb8 13 0-0-0 Bxf3 14 Bxf3 Bg5! and the point is, White’s
assassin bishops always work in tandem, so Black’s bishop decides to break
up the team. Black has equal chances once he rids White of the bishop pair,
B.Miljanic-P.Zarnicki, Internet (blitz) 2004.
12 ... Rb8
Covering the b7-pawn, which enables Black to move his knight.
13 0-0-0 Ne5!
 This move inflicts damage to White’s structure, although it does open the
g-file for White’s rooks.
14 h3 Bxf3 15 gxf3 c6 16 f4!
The less accurate 16 Be6?! is met with 16 ... Nxf3 when White is unable to
recapture on f5, since 17 Bxf5?? Nxd2 18 Rxd2 Bg5 hangs material.
16 ... cxd5 17 fxe5 dxe5 18 Be3! d4
Also, 18 ... f4 19 Rxd5 Qc7 20 Bd2 Bf6 21 b3 Rbd8 22 Rd1 is about even.
19 f4?!
White loses the initiative with this move. Correct was 19 cxd4 f4 20 dxe5
Qa5 21 Bd4 Qxa2 22 Rhg1 Rbc8 23 e6 Rg8 24 b3 Ba3+ 25 Kd2 Bb4+ 26
Kc1 (26 Kd3!? is playable but perhaps unwise after 26 ... Qa6) 26 ... Ba3+
with a draw.
19 ... Qa5
Also promising was 19 ... Qc7 20 cxd4 exf4 21 Bf2 Rbc8 22 b3 Bf6.
20 cxd4 exd4
Black had a better way with 20 ... exf4! 21 Bxf4 Qxa2! and if 22 Bxb8?
Bg5+ 23 Rd2 Rxb8 24 Kd1 Re8 25 Rf1 g6 26 d5 Re4 27 b3 Qa1+ 28 Qc1
Qxc1+ 29 Kxc1 Rd4 30 Rfd1 Kg7 31 Kc2 Bxd2 32 Rxd2 Rxd2+ 33 Kxd2
Kf6, he has a won king and pawn ending.
21 Bxd4 Qxa2 22 Rhg1 Bf6 23 Bxf6
Black’s initiative is minimized by 23 b3! and 23 ... Qa3+ is met with 24
Qb2 when Black has nothing better than 24 ... Qxb2+ 25 Kxb2. White’s
threat to invade the seventh rank should hold the game, despite Black’s extra
pawn.
23 ... Rxf6
Black’s extra pawn doesn’t matter as much as the king safety issue, which
is in Black’s favour.
24 Rd7 Rg8 25 Rg3
The blunder 25 Rxb7?? runs into 25…Rc6! which wins on the spot, since
26 b3 is met with 26 ... Qa1+ 27 Qb1 Qc3+ 28 Qc2 Qe3+, picking off a rook.

Exercise (combination alert): Black has access to tactics here. What would you play?

Answer: Double attack.

25 ... Qa1+! 26 Qb1 Qa4!
Here is the first double attack. Black simultaneously hits the d7-rook and
the c4-pawn.
27 Qd3
This is forced. Instead, 27 Rxb7?? walks into an even more deadly double
attack trick after 27 ... Qxc4+ 28 Qc2 Qxf4+, picking off the g3-rook.
27 ... Re6
Threat: ... Re1+, followed by ... Rd1+, winning White’s queen.
 28 Kd2?
White had to take his chances in the ending 28 Rg1 Re4 29 Kb1 Qxc4 30
Qxc4 Rxc4 31 Rxb7 Rxf4 32 Rxa7 g5.

Exercise (combination alert/planning): We must sort through the

chaos to cast even a glimmer of illumination to reveal Black’s
correct plan. White’s last move allows Black a forced win. How?

28 ... Qa1?
In positions of great complexity, and with our clock running low, it almost
feels as if the formulation of a plan is a pointless exercise in the absence of
easily available, hard data. After this momentary lapse, Black risks negating
all his previous strivings.
Answer: Black should double rooks, and White is helpless against the dual invasion threats on the
e1- and e2-squares. After 28 ... Rge8! (threat: ... Re1, followed by ... Rd1+), whisperings in the vicinity
of White’s king convey sinister portent:
a) 29 Rg1 Re2+! 30 Kc3 (30 Kc1 Re1+ 31 Rxe1 Rxe1+ 32 Kd2 Rd1+ 33
Ke2 Rxd3 and unfortunately for White, his Rd8+ back rank trick is covered
by x-ray, courtesy of Black’s d3-rook) 30 ... Qa5+! (a sweet-voiced wheedle
here and a pitiful whine there, is all White’s king hears from his importunate
sister, whose attention-seeking personality drives him crazy) 31 Kb3 Qb6+
32 Ka3 Qxb2+ 33 Ka4 Qa2+ 34 Kb4 Rb2+ 35 Kc5 Qa5+ 36 Kd4 Rd2 and
White’s queen is toast.
b) 29 Rgxg7 Qb4+! (after hearing of the white king’s overthrow, Black’s queen closes her eyes and
sways gracefully, as if to the cadence of music only she hears) 30 Qc3 Re2+ 31 Kd1 Re1+ 32 Kd2
R8e2+ 33 Kd3 Re3+ 34 Kd2 R1e2+ 35 Kd1 Rxc3 and there is no perpetual check after 36 Rxh7+ Kg8
37 Rhg7+ Kf8 38 Rgf7+ Ke8 and the e7-check square is covered.)
29 Kc2?
Better resistance is offered after 29 Rxb7 Re4 30 Re3 Rxf4 31 Rf3 Rxf3
32 Qxf3 Rd8+ 33 Kc2 Qa4+ 34 Qb3 and Black still has a tough fight ahead
of him.
29 ... Rb6
This still should win but much stronger was 29 ... Qa4+! 30 Kd2 Rge8
transposing to the answer to the last combination alert exercise.
30 Qd4 Rxb2+!

The king is prodded forth, to the accompaniment of recriminations,

warnings and outright threats. This shot fatally exposes White’s king.
31 Qxb2 Qa4+ 32 Rb3 Qxd7 33 Rxb7 Qe6
Taking aim at the c4-pawn.
34 Kb3 Qc6 35 Rb4 a5!
Principle: Passed pawns must be pushed. In this case, the pawn advance comes with tempi as it hits
the rook and also helps to push the king into less secure territory.
36 Rb5 a4+! 37 Kc3 Rc8
Black’s long out-of-play rook, who suffered a kind of spiritual anaesthesia
on his dismal past, now experiences a full recovery.
 38 c5
Also decisive is 38 Qb4 Qf3+ 39 Kb2 Qxf4.

Exercise (combination alert): White’s king, subject to lengthy and

painful interrogation, is about to talk. How did Black force the win?

Answer: Overloaded defender/deflection.

38 ... a3!
“There is one of you and one of me, so to my mind we begin this battle on
equal terms,” reasons the a-pawn to White’s queen, who towers above him.
White’s overloaded queen is unable to capture the a-pawn, which marches
forward.
39 Qe2 a2!
The same theme. White’s queen is staked out and tied down by a single
a2-Lilliputian.
40 Ra5 0-1
White made his move and resigned, seeing 40…a1Q+! (deflection) 41
Rxa1 Qxc5+ and White’s king, who at one point looked like he was
everywhere at once, is now nowhere at all. Black forces mate.
Summary: After 7 dxc3, I think White simply gets a slightly inferior version of the positions
reached next chapter.
Game 27
A.Shirov-A.Karpov
Corsica Masters, Bastia (rapid) 2002

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Qe2

A very strange opening choice for a player of Shirov’s fiery temperament.

The reason you don’t see any of my games in this book – unlike most of my
other opening books – is that some of my more grovely, cringing, snively
opponents at the San Diego Chess Club play ultra-drawish lines, like this one.
In an email, I whined to editor GM John Emms of the sad fact that Black
really can’t easily play Petroff for a win if White is determined to draw, and
he responded to those who play such lines with a resounding “Cowards!!”
proclamation. Be careful though. This line actually contains a bite to it if we
mishandle the defence, since we often land in a queenless symmetry, two full
moves down.
5 ... Qe7
The only move which doesn’t lose a piece.
6 d3

Question: Can White play 6 Nc3 like in the next chapter?

Answer: It doesn’t make sense to play this way to self-inflict slight distortion upon White’s
structure if queens immediately come off the board after 6 ... Nxc3 7 dxc3 Qxe2+ 8 Bxe2. Essentially,
White just got a milquetoast version of next chapter’s variation. After 8 ... Nc6 9 Bf4 Be7 10 0-0-0 Bg4
11 Rhe1 0-0-0 12 Nd4 Bxe2 13 Rxe2 Bf6 14 Nxc6, the players agreed to a draw in this not-so-
fascinatingly even ending, R.Hübner-Y.Seirawan, Manila 1990.
6 ... Nf6 7 Nc3
White’s main line is 7 Bg5, although the two can easily later transpose.
The threat is to inflict damage to Black’s structure with Bxf6:
a) 7 ... Qxe2+ 8 Bxe2 Be7 9 Nbd2 (9 Nc3 c6 10 0-0-0 Na6 11 Nd4 Nc7 12
Rde1 Ne6 13 Nxe6 Bxe6 14 f4 d5 15 Bf3 Kd8 16 Ne2 Re8 17 Ng3 g6 18
Ne2 Bd7 19 c4 dxc4 20 dxc4 Ng8! when Karpov equalized and actually went
on to win, J.Timman-A.Karpov, Tilburg 1980) 9 ... Nd5 (exchanges benefit
Black, who eliminates his bad bishop) 10 Bxe7 Nxe7 11 0-0-0 Nbc6 12 Rhe1
Bd7 13 d4 0-0-0 14 Bc4 d5 15 Bd3 Rde8 16 c3 Bf5, and the game is dead
even, Li Chao-Wang Yue, Shenzhen (rapid) 2013.
b) 7 ... Nbd7 8 Nc3 Qxe2+ (8 ... c6?! is weak, since it allows 9 Ne4 d5 10
Nd6+ Kd8 11 Qxe7+ Kxe7 12 Bf4 Nh5 13 Nxc8+ Rxc8 14 Be3 with an edge
for White, due to the bishop pair) 9 Bxe2 h6 10 Bh4 g5!. Here, Black gets
aggressive, freeing his d7-knight from having to cover the f6-counterpart,
while gaining kingside space. I think this means more than Black’s slight
weakening of f5. Following 11 Bg3 Nb6! 12 Nd2 Nfd5 13 Nxd5 Nxd5 14 h4
g4 15 0-0-0 Be6 16 d4 Bg7 17 Bb5+ Kd8 18 c3 a6 19 Bd3 Kd7, the game
remains in balance, Kr. Georgiev-E.Rozentalis, Ermioni Argolidas 2005.
7 ... Qxe2+
This move is labelled “very solid” by Sakaev.
 Question: Doesn’t this help White’s development?

Answer: It does, but the position is so lacking in dynamism that Black remains fine. Also, at some
point or another, Black nearly always is the one forced later on to do the swapping on e2, since White’s
development lead tends to force it. 7 ... Nbd7 is perhaps just a shade more dynamic, since it may lead to
the Georgiev-Rozentalis note from above.
8 Bxe2 Be7
The position is in dead symmetry, except White has two extra moves in
hand. However, the drab nature of the position renders those two moves
pretty much meaningless if Black defends properly.
9 0-0
a) 9 Nd4 0-0 (After 9 ... Bd7, Shirov got overtly creative with 10 g4?! (this
isn’t a position conducive for flights of imaginative fancy) 10 ... Nxg4! 11
Rg1 h5 12 h3 Ne5 13 Nd5 Bd8 14 f4 Ng6 15 f5 Ne7 16 Nxe7 Bxe7 and the
g7-pawn can’t be touched. White lacked compensation for the pawn and was
unable to hold the game, A.Shirov-J.Lautier, Plovdiv 2003) 10 Bf4 c6 11 0-0
d5 12 Rfe1 Na6 13 Bf3 Bd8 14 a3 Bb6 15 Nb3 Re8 16 Na4 Bc7 17 Bg5 Bf5
18 Bxf6 gxf6 (Black can handle this weakening of his kingside pawns) 19 g3
Bd6 20 Na5 Rxe1+ 21 Rxe1 Rb8 22 Nc3 Bd7 23 Nb3 Nc5 24 Nxc5 Bxc5 25
Kg2 Kf8 and Black’s bishop pair easily made up for his structural damage,
V.Ivanchuk-Wang Yue, Beijing blindfold (rapid) 2011.
 b) 9 Bg5 can transpose to the Timman-Karpov note from above.
9 ... c6
This flexible move cuts out future Nb5 worries and prepares the
manoeuvre ... Na6-c7, which allows Black to block the e-file if necessary
with ... Ne6 later on.
10 Nd4
Seizing control over the f5-square, while clearing the f3-square for his
bishop.
10 ... Na6 11 Re1
After 11 Bf3 (threat: Nxc6) S.Vega Gutierrez-J.Garcia Padron, Las Palmas
2012, Black should play 11 ... 0-0 12 Re1 Re8 13 Bd2 Bd7 with full equality.
11 ... Nc7 12 Bf3 Kf8!?

Black’s king, whose new home lacks much in the way of creature
comforts, escapes the e-file pin, while also dodging Nxc6 tactics, but at a
steep cost. Perhaps Karpov, in secretly playing for the win, goads Shirov into
rash action by taunting him at a subliminal level. White doesn’t have much,
but if I had a choice, I would still take his side since Black’s lazy h8-rook’s
work ethic is in little evidence, and will remain out of the game for a while.
Let’s examine the alternatives:
a) Much safer is 12 ... Bd7 13 b4, S.Vega Gutierrez-S.Milliet, Khanty-
Mansiysk 2010. Here Black can try 13 ... Ng4!? (intending ... Ne5) 14 Rb1
Ne5 15 Bh5 0-0 16 f4 Ng6 17 Bxg6 Bf6! 18 Bxh7+ Kxh7 19 Be3 d5 20 Bf2
Rfe8 and Black’s bishop pair and light square control offer full compensation
for the pawn, but probably few actual winning chances.
b) Nobody has tried 12 ... Be6!? 13 b4 a6 14 Nxc6 bxc6 15 Bxc6+ Kd8 16
Bxa8 Nxa8 17 Ne2 Kd7. I suspect that chances are close to even in this
unbalanced ending with a rook and two pawns against Black’s two minor
pieces.
13 b4!
This move implies menace lying in wait, just beneath the position’s
surface.

Question: What is the point of 13 b4?

Answer: It follows the principle: Create confrontation when leading in development. Black must be
on the constant lookout for b4-b5 ideas.
13 ... Bd7?!
This move allows White to realise his intent unchallenged. Instead 13 ...
Ng4! (intending ... Ne5) which may be an improvement. After 14 Bd2 Ne5
15 Bd1 f5 16 f4 Nf7 17 Bf3 Bf6 18 Be3 g6, Black will be fine if he has time
for ... Kg7 and ... Bd7. If 19 b5 c5 20 Nde2 Rb8 21 d4 b6 22 Rad1 Kg7, I
don’t believe White stands any better.
14 b5
Now White has a shade of an edge, since Black’s development lag
becomes a source of concern.
14 ... d5
This allows White to pry open the b-file. Also, after 14 ... c5 15 Nde2 Rb8
16 a4 h6 17 Ng3, Black is unpleasantly tied down.
15 bxc6 bxc6 16 Rb1 Bd6 17 Rb7 Nfe8 18 Be3
Going after the a7-pawn, but Black’s next move doesn’t give him time.
18 ... Bc8 19 Rb3 Bd7
A draw offer.
20 Reb1
... which is declined.
20 ... Ke7!
 Karpov enlists the aid of his king to defend his queenside, while preparing
to activate his h-rook.
21 Rb7 g6!
Preparing ... Ng7-e6. So subtle and wisp-like is Karpov’s defensive play
that its details tend to disappear at the moment of perception.
22 Nb3
Uncovering an attack on the a7-pawn.
22 ... Ng7!
The blank-faced knight’s thoughts are as unreadable as a porcelain doll.
23 Na5
Shirov isn’t satisfied with 23 Rxa7 Rxa7 24 Bxa7 Ra8 25 Be3 Nf5 26 Bd2
Ne6 27 a4 Nfd4 28 Bd1 Be5. Black’s activity, coupled with White’s passive
stance, offers Black full compensation for the pawn.
23 ... Nge6 24 Na4!?
After 24 Ne2 Rhc8 25 g3 Nc5, Black’s defensive difficulties are at an end.
24 ... c5!
We note that White’s initiative tide noticeably slackens and recedes.
25 Nxc5!?
I’m not so sure this ad-hoc addition is beneficial to White’s cause. Shirov
continues to press, rejecting a safer line like 25 Nb2 Nd4 26 Bg5+ f6 27
Re1+ Be6 28 Be3 Nxf3+ 29 gxf3 Kd7 30 Na4 d4 31 Bf4! Rhc8 32 Bxd6
Kxd6 33 c4 Bf5 34 Rb3 Re8 35 Nb7+ Kc6 36 Na5+ with perpetual check.
25 ... Nxc5 26 Rxc7 Bxc7 27 Bxc5+
The bishop differs with the scriptures on the notion of all life being
sacrosanct, since he feels his own is a tad more sacrosanct than others.
27 ... Bd6 28 Bd4 Rhc8 29 Bxd5 Rab8 30 Re1+ Kf8 31 Nc4 Bc5
In the aftermath White gets two healthy pawns for the exchange and the comp assesses the position at
close to even, while I favour Black (maybe the bias comes from the fact that Black won the game).
32 Be5 Re8
Pinning the e5-pest.
33 Kf1 Rbd8 34 Bf6 Rxe1+ 35 Kxe1 Re8+ 36 Be4?
Correct is 36 Kf1 Be6 37 Bxe6 Rxe6 38 Bb2 Ke7 when Black only stands
slightly better.

Exercise (combination alert): White’s last move allowed a combination

from Black, which leads to a winning position. What should he play?

36 ... Bb5?
Answer: Tempo gain and then exploit the pin.
Karpov must have been in severe time pressure to miss 36 ... Re6!. The
bored rook taps his fingers rhythmically on his desk, as White’s flustered
bishops plead their case. Now a new source of revenue awaits Black, since
this move clears the way for ... f5, which kills the pinned e4-bishop. 37 Bc3
f5 38 Ne5 Rxe5 39 Bxe5 fxe4 40 dxe4 and White may have three pawns for
the piece, but in this instance he is completely lost. His queenside pawns are
split, rendering them targets for Black’s powerful bishops.
37 Ne5?
Shirov must have been just as low on the clock, since he allowed the same
combination, but this time in an even more fatal version. Instead, 37 Ne3!
avoids the combination.
37 ... Re6!

“When a person of power – namely myself – hints or insinuates, it would

be in your best interest to listen carefully,” the rook advises White’s dark-
squared bishop. Now Karpov sees it. White’s f6-bishop and knight watch
stupidly, as their e4-counterpart is forced into a deadly pin, the way
miscommunicating tennis doubles partners just stand there, both thinking the
other will hit the ball. Now heavy material loss is inevitable.
38 c4 Ba4 39 Bh8
The bishop changes accommodations from the palace to the rustic
simplicity of his new home, the dungeon.
39 ... f6 40 Ng4
White’s minor pieces walk slowly and carefully around Black’s rook, as if
their feet hurt. Also hopeless is 40 f4 fxe5 41 Bxe5 Bc6.
40 ... f5 0-1
 The f-pawn doesn’t believe in negotiation, since he prefers to acquire
100% of his desires, rather than just 50%.
Summary: Most of your opponents who play 5 Qe2 will be after a draw. Don’t underestimate this
line, since in most variations we get queenless symmetry, two full moves down, which can easily turn
into something tangible for White if we mishandle the defence.

Game 28
A.Shirov-M.Adams
European Team Championship, Leon 2001

1 e4 e5
Compare our line with the Kopec Sicilian: 1 ... c5 2 Nf3 Nc6 3 Bd3!?, and
after 3 ... e6 4 c3 Nf6 5 Bc2, d2-d4 is coming and White’s once odd bishop
gains a degree of respectability after 5 ... d5 6 e5 Nd7 7 d4, D.Kopec-
C.Lakdawala, Internet (blitz) 2014.
2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Bd3

This line borrows an idea from the Kopec Sicilian.

Question: Did White forget to play d2-d4 first?

Answer: I admit the move looks amateurish, but isn’t as weak as it appears.

Question: What would be the reason White blocks his own development like this?
 Answer: The block of development is merely temporary. White, inverting the normal order of
things, plays it like a Kopec Sicilian, planning a quick c2-c3, Bc2 and d2-d4, when harmony returns to
his position.
5 ... Nf6
This simple retreat is considered Black’s most efficient path to equality.

Question: Why retreat when Black can back up the knight with 5 ... d5?

Answer: This is also possible. However, I think White can extract an edge.
a) After 5…d5 6 Qe2 Qe7 7 0-0 Nd6:
a1) 8 Qd1 g6 9 Re1 Be6 10 Bf1 Bg7 11 d4 0-0 12 Bf4 Nc6 13 c3 Qd7 14 Nbd2 Rfe8 15 h3 f6 and
Black equalized in A.Morozevich-A.Shirov, Astana (rapid) 2001.
a2) 8 Nc3! c6, J.Magem Badals-A.Shirov, Liepaja 2001. White is better after 9 Re1! Qxe2 10 Nxe2
and Black’s problem is 10 ... Be7 is met with 11 b3! when Black can’t castle, since 11 ... 0-0? is met
with 12 Ned4 Bd8 13 Ba3 Bc7 14 Re7 Nd7 15 Ng5! and Black is busted, since 15 ... g6? is hit with 16
Nxh7! Kxh7 17 Ne6 which is decisive.
b) 5 ... Nc5 doesn’t really gain a tempo, since White later regains time on
the c5-knight after 6 Be2 Be7 7 0-0 0-0 8 d4 (White regains the lost tempo).
After 8…Ne4 9 c4 c6 10 Bd3 d5 11 Re1 f5!? (risky, since it creates an e5-
hole and 11 ... Nf6 is safer, although I like White’s game after 12 Nc3) 12
Nc3, N.Short-V.Otsmann, Puhajarve 2012, I prefer White after 12 ... Be6 13
Qb3 Qb6 14 Qxb6 axb6 15 cxd5 cxd5 16 Nb5 Bd7 17 Ne5 when White
exploits the b5- and e5-holes.
6 0-0
The immediate 6 h3 cuts off Black’s ... Bg4 option, but that shouldn’t
matter much since Black usually fianchettos later in this line with ... b6. After
6 ... Be7 7 c3 0-0 8 Bc2 b6 9 0-0 Bb7 10 d4 Nbd7 11 Nbd2 Re8 12 Nc4 a5 13
Ne3 g6 14 Ba4 Bf8 15 Ng4 Nxg4 16 hxg4 Bg7, N.Short-Wang Yue,
Poikovsky 2012, chances are even.
6 ... Be7
6 ... Bg4 is also possible. Following 7 Re1+ Be7 8 h3 Bh5 9 c3 (9 Qe2
Nc6 10 g4 Bg6 11 Bb5 Qd7 12 Bxc6 bxc6 13 Nc3 Kf8 Loss of castling rights
isn’t the end of the world here for Black, since he gains kingside light square
counterplay with a future ... h7-h5) 9 ... 0-0 10 Bc2 Nbd7 11 d4 Re8 12 Nbd2
Bf8 13 Rxe8 Qxe8 14 Nf1 Bg6 15 Bf4 Bxc2 16 Qxc2 Qe4, Black has
equalized, R.Rabiega-P.Haba, Internet (blitz) 2004.
7 h3 0-0 8 c3 Re8
A good alternative is 8 ... b6 9 Bc2 Bb7 10 d4 (10 Re1 Re8 11 d4 Nbd7 12
Bg5 Nf8 13 Nbd2 Ng6 14 Ba4 Nd7 15 Be3 a6 (preparing ... b7-b5) 16 Nf1 b5
17 Bb3 Nb6 18 Ng3 Nd5 and the future world champion got nothing from the
opening, M.Carlsen-K.Lahno, Lausanne 2004) 10 ... Nbd7 11 Nbd2 Re8 12
Nc4 Bf8 13 Bg5 h6 14 Bh4 g6 15 Ne3 Bg7 16 Nh2 Qc8 17 Nhg4 Nxg4 18
Nxg4 f5 19 Ne3 Nf6 when Black unravelled with an excellent position,
D.Navara-B.Gelfand, Wijk aan Zee 2012.
9 Bc2 b6!

This is a reliable equalizer. Black scores over 50% from this position.

Question: Isn’t this fianchetto unnatural, since

it weakens Black’s queenside light squares?

Answer: I think it is Black’s best move for the following reasons:

1. The fight is for control over the e4-square.
2. The fianchetto is logical since White denied Black’s light-squared
bishop access to both the g4- and f5-squares.
3. White isn’t in a position to exploit the superficial weakening of Black’s
queenside light squares.
I think Adams’ last move is slightly more accurate than 9 ... Nbd7 10 d4
c6 11 Re1 Nf8 with a passive but very solid position for Black, J.Haakert-
P.Staller, Germany 2001.
10 d4 Bb7 11 Bg5
 11 Re1 transposes to the M.Carlsen-K.Lahno game, mentioned in the notes
to Black’s 8th move.
11…h6 12 Bh4 Ne4!

Principle: Swaps benefit the cramped side. This idea is borrowed from
Lasker’s line of the Queen’s Gambit Declined.
13 Bxe7 Rxe7 14 Nbd2
White can gain space with 14 d5 (cutting off support for Black’s e4-
knight) 14 ... Nd7 15 c4, E.Real de Azua-F.Fiorito, Villa Ballester 2004. Here
Black can try undermining the white centre with 15 ... b5!? 16 cxb5 Nef6 17
Nc3 Nb6 18 Bb3 a6 19 bxa6 Rxa6 20 Re1 Rxe1+ 21 Nxe1 Qa8 22 Rc1 Ra5
when Black regains the sacrificed pawn with a good position.
14 ... Nxd2 15 Qxd2!?
 Such a radical decision indicates a natural attacker’s correspondingly
belligerent temperament. Shirov, whose creativity proves to be an eternally
renewable resource, wilfully encourages damage to his structure to generate
play down the newly opened g-file, and enhance his control over the light
squares. Yet I feel that his decision catches glimpses of the position’s partial
truth, rather than the all-encompassing one he desires. Your deeply
unimaginative writer would have undoubtedly played the limp 15 Nxd2 and
after 15…Nd7, Black has already equalized completely.
15 ... Bxf3
Adams is up for the challenge, refusing to back down with 15 ... Nd7.
16 gxf3
White’s carbuncular kingside pawns don’t make a pretty sight.
16 ... Nd7 17 Kh2
Preparing to load up the rooks along the g-file.
17 ... Re6
A precaution to cover against future Qxh6 tricks.
18 Rg1 Qf6 19 Rg3 Kh8
A move impelled by the stark practicality of survival. Adams wisely
removes his king from the g-file, preparing ... Rg8.
20 Rag1 Rg8
Black has sufficient defensive resources to counter White’s g-file build-up.
 21 f4 Re7 22 Qd3
White’s forces are never happy unless they agree upon a foreign threat to
unify their fear with hate and xenophobia. In this case, Black’s king supplies
their psychological need admirably. “A senile, bedridden king is not fit to
rule,” the queen tells her h8-brother, who is neither senile nor bedridden.
White threatens a Qh7-cheapo. When we desire someone to comply with our
wishes, the first step is gentle, logical persuasion, and failing that, follows
rebuke and outright threats.
22 ... Nf8
Black’s king gains the protective services of a beneficent friend on f8.
23 Qf3 c5
Principle: Counter a wing attack with a central distraction.
24 d5
This move places a pawn on the same colour as White’s remaining bishop,
yet I think it’s the correct decision. White can also trade pawns with 24 dxc5
dxc5 25 Bb3 Ne6 26 f5 Nc7 27 Qh5, J.Dudas-M.Galyas, Budapest 2002. The
players agreed to a draw here though I think Black’s position still offers some
promise after 27 ... Ne8, intending ... Nd6.
24 ... b5

Houdini evaluates this at even, while I feel like Black is the only one who can win. White’s
kingside attack is at a standstill, with no way to increase its energy. Meanwhile, Black can poke and
prod on the queenside, hoping to distract White’s kingside forces.
25 R1g2
White’s silence on the matter of the kingside grows strained. There is
simply no way to increase the pressure.
25 ... Re1 26 Rg1 Re7
Perhaps a draw offer.
27 a3
Shirov still harbours ambitions. I think he should have offered to repeat
moves.
27 ... c4!
Now I clearly prefer Black. This move cuts out Qd3 and opens the way for
... Nd7-c5.
28 Qg4 Nd7 29 a4?!

Opening the queenside can only favour Black. This move violates the
principle: Don’t create play on your opponent’s strong wing. I would either
just shuffle, or challenge the e-file with 29 Re3.
29 ... a6
Black can also consider opening the queenside with 29 ... bxa4 30 Bxa4
Nc5 31 Bc2 Rb7 which forces White on the defensive with 32 Rb1 Rgb8 33
Kg2 a5 (certainly not 33 ... Rxb2?? 34 Rxb2 Rxb2 35 Qc8+ and Black is
mated) 34 h4 Nd3! 35 Bxd3 cxd3 36 Rxd3 Rxb2 37 Rxb2 Rxb2 38 f5 Qe5.
Black’s passed a-pawn and superior structure offers him all the chances.
30 a5!?
Now the a5-pawn may become a target. He may have been better off with
the line 30 axb5 axb5 31 Ra1 g6.
30 ... Nc5 31 Re3 Nb3?!
This move allows White to retain the balance. White is in deep trouble
after 31 ... Rxe3 32 fxe3 Re8 33 Qf3 Na4! 34 Bxa4 bxa4 35 Ra1 Qf5.
32 Rge1?!
This move weakens White’s argument to some extent. White’s faltering
attack grows in dysfunction in inverse proportion to its actual efficiency.
Instead, he should hold the game after 32 Bxb3! cxb3 33 Rge1 Rxe3 34
Rxe3.
32 ... Rxe3
A concession which allows White to reconnect his pawns, but this is the
only way for Black to make progress.
33 fxe3 Nc5
White is in deep trouble after the counterintuitive 33 ... Nxa5! 34 Ra1 Nb7
35 Rxa6 Nc5 36 Rb6 Re8 37 Qg3 Nd3! 38 Rxb5 Qe7 39 Bxd3 cxd3 40 Rb4
Qa7! 41 Rd4 Rb8 42 b4 Qa2+ 43 Qg2 Qe2 44 Kg3 Qxe3+ 45 Qf3 Qg1+ 46
Qg2 Qe1+ 47 Qf2 Qxc3 when White is unlikely to save himself.
34 h4
Houdini suggests the inhuman 34 Kg3!? which follows Steinitzian principle: The king is a fighting
piece. Use it.
34 ... Re8
Also promising is 34 ... b4!.
35 Rg1
White’s best chance to save the game lay in 35 Qh3.
 Exercise (planning): Come up with a winning plan for Black.

Answer:
Step 1: Force the creation of a passed pawn.
35 ... Nd3!
A dual interference/double attack threat has been issued. Now the b2- and
e3-pawns are simultaneously attacked, forcing White’s next move. Black’s
last move contains two deadly points:
1. It forces White to hand Black a dangerously passed d-pawn.
2. By blocking White’s bishop on the b1-h7 diagonal, Black enables the
previously unplayable ... Rxe3, since White’s back rank threats are negated.
36 Bxd3 cxd3 37 Rg3 b4!
Step 2: Queenside breakthrough. This pawn push allows Black’s queen to
infiltrate with deadly effect.
38 cxb4 Qxb2+ 39 Rg2 Qc3 40 Qd7 Rg8
The rook fears White’s queen more than Howard Hughes feared airborne
germs. This safety-first move is unnecessary but still winning. Black gets
away and wins with 40 ... Rxe3! 41 Qxf7 d2 42 Qf8+ Kh7 43 Qf5+ Kg8
(there is no perpetual check) 44 Qg4 h5! (overloading White’s queen) 45
Qxh5 Rh3+ 46 Kg1 Qe3+ 47 Rf2 Qe1+ 48 Kg2 Qh1 mate!
 41 Qg4
The parties wrestle for possession of the d1-square.
41 ... Qe1!
Black’s queen cuts off Qd1 and there is no defence to the coming ... d3-d2.
42 f5
Intending f5-f6. White attempts one last lunge with the totality of his fast-
waning strength.
42 ... d2 0-1
White is unable to issue sufficient threats before the pawn promotion after
43 f6 (43 Rg1 g6! 44 f6 Re8 leaves White paralyzed) 43 ... g6 44 Qf4 g5 45
hxg5 d1Q.
Summary: The odd looking 5 Bd3 – which is a hybrid idea borrowed from the Kopec Sicilian,
looks like a person sawed off the top of a totem pole and then cemented the head of a Grecian marble
statue on it. It is best neutralized by retreating our e4-knight, and then later fianchettoing on the
queenside.

Game 29
C.Camano-M.Ginzburg
Argentine Championship, Buenos Aires 1994

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Be2

Obviously White is happy to draw.

Question: Why develop the bishop to such a passive

 square when the more ‘active’ d3-square is available?

Answer: In this version, White essentially meets Petroff with a Petroff setup! He plans to play it
super passively with d2-d3 (without d3-d4 later on) and dares us to come and beat him.
5 ... d5 6 0-0 Bd6
Just because our opponent plays passively doesn’t mean we have to follow
suit. Black’s bishop grabs the best diagonal.
7 d3 Nf6 8 Bg5 0-0 9 Nc3
Nxd5 is an obvious and yet strategic threat.
9 ... c6
Not the careless 9 ... Re8? 10 Nxd5 Bxh2+ 11 Nxh2 Qxd5 12 Bxf6 gxf6
13 Bf3 Qd6 14 Ng4 when Black has no compensation for the structural
damage to his kingside pawns.
10 Re1
White won’t even play d3-d4 and plays the position as if he were Black,
intent on a draw.
10 ... Bf5 11 Qd2 Re8 12 Bf4
Welcome to my life at the San Diego Chess Club. White is intent on
draining all the dynamism out of the position.
12 ... Qc7 13 Bxd6 Qxd6 14 Nd4!?

Question: Is this the beginning of ambition from White’s side?

 Answer: Maybe just a tad. Perhaps White is trying to provoke Black into a future ... c6-c5. His
intent is to play for a queenside bind of the dark squares with a2-a4-a5, Nb3 and Na4. The problem
with that plan is that the b7-pawn – White’s lone target – is easily defended, and that White removes
key defenders from his king. If White wants a draw, then he should go into vacuum cleaner mode and
continue to exchange pieces with 14 Bf1 Nbd7 15 Rxe8+ Rxe8 16 Re1 Rxe1 17 Qxe1.
14 ... Bg6 15 a4!?
Following through with his queenside plan.
15 ... Nbd7
I would think about 15 ... a5 securing the b4-square and preventing
White’s next move.
16 a5 a6
Now ... c6-c5 is enabled, since Nb5 is no longer possible.
17 Nb3 Re6 18 Bf1 Rae8 19 Rxe6?!
Nothing worthwhile comes from White’s proposed transaction. I would
have avoided this swap, which can only benefit Black.
19 ... fxe6!

At long last, Black achieves a slight alteration of the nauseatingly endless

symmetry in the structure. With his last move, Black opened the f-file for his
rook and grabs central influence.
20 Re1 e5
White has yet to really make a major mistake, yet he stands worse since he
operated in ultra-passive mode from the get-go.
21 g3!?
White attempts to prod along his immovable mass. He feels the need to
substantially modify his original intent – to do absolutely nothing – into a
pocket-sized economy version. Now control or loss of the kingside light
squares will be an equal partner in White’s coming joys or sorrows. The
alternative is to sit there and do absolutely nothing.
21 ... Rf8!
An attack brews on the kingside with ... Ng4 in the air.
22 Bh3
The bishop embraces his new-found freedom.
22 ... h6!
Now White must worry about ... Nh7-g5 ideas.
23 Na4
This plan of seizing the queenside dark squares feels distant and
unattainable. White’s play on that wing is like a man who shows up at the
bank loan manager’s office unshaven, unshowered, and dressed in a tattered
tee shirt and jogging pants. He probably won’t get the loan, since outer
appearances matter.
23 ... e4
I also prefer Black after 23 ... Nh7.
24 Bxd7?
 In chess there are two kinds of mistakes:
1. Pardonable, where the position’s complexity level is too high and we
either misjudge or miscalculate a sequence.
2. Unpardonable, where the damage to our position is entirely self-
inflicted.
Duty calls in vain, since the bishop has no intention of sticking around to
defend his king. In a single move, it is as if White’s luck decided to pack its
toothbrush and bags, and withdraw all support to his king. This move
eradicates the lone protector of White’s kingside light squares. 24 d4 was
necessary.
24 ... Nxd7! 25 d4
The only move.

Question: Why can’t White win a pawn with 25 dxe4 since

Black is unable to recapture with either pawn or bishop?
 Answer: Black has no intention of recapturing and plays 25 ... Ne5! threatening a f3-megafork. So
enamoured with himself is Black’s knight, that he prefers his own company to that of everyone else.
After 26 Re3, Black initiates a decisive attack with 26…Bxe4!. If 27 Qd1 Qg6! (attacking the c2-pawn)
28 Nd4 Ng4 29 Re2 Nxh2!, Black has a crushing position, since taking the knight leads to mate after 30
Kxh2?? Qh5+ 31 Kg1 Qh1 mate!
25 ... Nf6 26 Nc3 Nh7!
With the idea of ... Ng5.
27 h4
Forced, but White’s position grows worse by the move.
27 ... Bh5 28 Nc5
Attacking the b7-pawn but Black can’t be bothered with such trifle issues
when he is aiming to gun down the white king.
28 ... Bf3
The bayonet thrust 28 ... g5! gives Black a winning attack. After 29 Nxb7?
Qd7 30 Nc5 Qh3, White is doomed.
29 Ne2
Also, 29 Nd1 g5 gives Black a winning attack.
29 ... g5!
Black flails away at the white king’s cover with the force of an egg whisk.
30 hxg5
Now White’s king is stripped naked and left shivering in the wind. This
loses, but White had no way out in any case.
 30 ... hxg5 31 g4

Exercise (combination alert): Black to play and force mate.

Answer: Transfer the queen to the h-file, when White is unable to defend his fatally weakened light
squares.
31 ... Qh6 0-1
The move is so obvious that it doesn’t deserve an exclam. The queen
increases her malevolence level of her already hateful gaze by at least 25%, if
not more as 32 Ng3 Qh3 is decisive.
Summary: Obviously, 5 Be2 isn’t intended to strike fear into our hearts. White’s idea is to go into
crouch mode with d2-d3 and ask our side how we intend to make progress.

Game 30
J.Maslowski-M.Sarwinski
Bydgoszcz Open 2001

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 c3

 White usually intends d2-d4 and Bd3 next.

Question: What is the point of inverting the move order and playing 5 c3 first?

Answer: White intends a vulgar cheapo of truly epic proportions with Qa4+, which wins the e4-
knight if he misses the threat.
5 ... Nc6
Let’s look at other options:
a) Eleven (!) unfortunate souls fell for the trap 5 ... Be7?? 6 Qa4+ when
Black hangs his e4-knight.
b) The logical continuation is 5 ... d5 6 d3 Nf6 7 d4 c5 (the most
enterprising move) 8 Be2 Be7 9 0-0 0-0 10 dxc5 Bxc5 11 Bg5 Be7 12 Nbd2
Nc6 13 Nb3 Re8 14 Re1 Bg4 15 Qc2 Qd7 and the players agreed to a
premature draw in this isolani position, V.Hort-A.Karpov, Montreal 1979.
6 d4 d5
It isn’t too late to fall for White’s trap with 6 ... Be7?? 7 d5 and when
Black’s knight moves, Qa4+ once again picks up the dangling e4-knight.
7 Bd3 Bd6
Aggressively seizing control over the e5-square. If you want to transpose
to a position we covered in the previous chapter, then play 7 ... Be7 8 0-0
Bg4 9 Re1 f5 10 Qb3 0-0 11 Nbd2 Na5.
 8 Nbd2 0-0 9 0-0
9 Nxe4? is obviously weak in light of 9 ... dxe4 10 Bxe4 Re8 11 Qe2 Bf5!
(less accurate is 11 ... f5 12 Ng5! Rxe4 13 Nxe4 fxe4 14 Qxe4 when White
stands only a tad worse) 12 Bxf5 Rxe2+ 13 Kxe2 when Black obviously
holds all the winning chances. White actually has enough material for the
queen, but Black’s leads in development and White’s king isn’t completely
secure.

9 ... Bf5
Harmoniously completing his development. The resulting position is
drawish.

Question: How does Black play if we want to win?

Answer: If Black wants to play for a win, then we should go for 9 ... f5 10 Qb3 Kh8 when Black
gets a superior version of the previous chapter, since his bishop is better placed on the d6-square, rather
than the e7-one, A.Simonik-M.Pajersky, Czechia 2012.
10 Qc2 Re8 11 Re1 Nxd2
Played in all three games in my database. This move achieves drawish
equality.
12 Bxd2 Bxd3 13 Qxd3 h6
 Covering against Ng5 ideas. However, 13 ... Rxe1+ 14 Rxe1, J.Araguas
Bestue-J.Torres Martinez, Barcelona 2002, is a drawish line instead.
14 Qf5
The queen hopes to make trouble for Black’s king on the kingside. If
White is less ambitious, he can play 14 Rxe8+ Qxe8 15 Re1 Ne7 16 g3 Qd7
17 Kg2 Re8 with a draw as the most likely outcome.
14 ... Ne7
Black redeploys a defender for his king.
15 Qg4
Threat: Bxh6.
15 ... Qc8
Oh, no you don’t.
16 Qxc8
After 16 Qh5, Black forces the swap in any case with 16 ... Qf5 and White
must cooperate, since 17 Qh4? is met with 17 ... Ng6 18 Qh3 Qxh3 19 gxh3
when White had to swap anyway, while taking on three unnecessary isolanis.
16 ... Raxc8
I get this kind of position all the time from the Black side of the Exchange
French, against players rated between 2100 and 2400 who are intent on a
draw. When I fail to win, I invariably berate my opponent – with obvious
sour grapes – for being a shameless (yet successful) groveler!
 17 Ne5 c5?!

Black is determined to carry out his win-at-all-costs project to its

inevitable conclusion.

Question: Isn’t wilfully taking on an isolani a

risky undertaking with queens off the board?

Answer: I would certainly docket this decision in the ‘rash’ category. Normally it is, and I think
Black’s decision is the classic overpressing of the higher rated player trying too hard, and risking loss.
However, I have learned that irrationality and being wrong about something are not necessarily
synonymous. Sometimes we try an irrational idea at the board, only to be later awarded with a win, just
as in this game. However, I could never see myself playing such a move, no matter how badly I wanted
to create an imbalance. In fact, getting a pure positional player like your writer to engage in a move like
... c5?! is a little like asking Joan of Arc to renounce her visions of God as unreal and getting her to fill
out a prescription for anti-psychotic medications to prevent such episodes from recurring. If I had this
position against a lower rated opponent, I would jump on the imbalance 17 ... Bxe5! 18 Rxe5 (Black’s
knight is no worse than White’s remaining bishop) 18 ... f6 19 Re2 g5! (cutting off Bf4) 20 Rae1 Kf7
21 f3 Nf5 22 Rxe8 Rxe8 23 Rxe8 Kxe8 24 Kf2 with an equal ending.
18 dxc5 Bxc5 19 Nd3 Bb6 20 Be3
Principle: Exchanges favour the side without the isolani.
20 ... Bxe3 21 Rxe3 Nf5 22 Rxe8+ Rxe8 23 Re1
He can also try 23 Kf1 d4 24 c4 when I prefer White’s queenside majority
to Black’s passed d-pawn.
23 ... Rd8!?
 Black has better drawing chances by trying his luck in a knight ending
with 23 ... Rxe1+ 24 Nxe1 Kf8 25 Nc2 Ke7 but the problem is Black never
intended to play for the draw in the first place!

Black’s rook half-heartedly objects, but decides not to make a scene and
steps aside for his presumptuous colleague. Conceding the e-file looks
wrong.

Exercise (planning): Come up with a clear plan for White

to seize the initiative and secure a draw at a minimum.

24 Nc5?!
When we sense the presence of the correct plan, yet are unable to find it,
it’s as if we landed the lead role playing Hamlet on stage, receive our cue and
face the audience to deliver our soliloquy, only to forget the most important
line: “To be or not to be?: that is the question.” This move allows Black to
escape his difficulties.
Answer: White can press with 24 Re5! g6 25 g4! (seizing the seventh rank) 25 ... Nh4 26 Re7 b5 27
Ne5 (threatening the f7-pawn while menacing Nc6 in some cases) 27 ... d4! 28 Kf1! (28 Nxf7?? dxc3!
– threatening back rank mate as well as pawn promotion tricks – 29 Kf1 cxb2 30 Re1 Kxf7 and White
loses a piece) 28 ... dxc3 29 bxc3 a6 30 Nxf7 Kf8 31 Ra7 Rd2 32 a3 Ra2 33 Nd6 Rxa3 34 Nxb5 Rb3
35 Nd4 Rxc3 36 Rxa6 and White is the only one with winning chances in this ending, which is a
probable draw.
24 ... d4!
 Now Black is okay again.
25 cxd4
Only White can win in the line 25 g4! Nh4 26 Rd1 Re8 27 cxd4 b6 28 Nb3
f5 29 gxf5 Nxf5.
25 ... Rxd4 26 Nb3 Rb4
Now I slightly prefer Black’s position, although the correct result should
be a draw.
27 Rd1 g6 28 h3?!
White later comes to regret this decision, which subtly weakens White’s
kingside dark squares. 28 g3 is more accurate.
28 ... b6
Perhaps intending ... a7-a5-a4, undermining White’s knight.
29 Rd2 Rc4 30 Kf1 h5 31 Ke2 h4
Fixing the g2-pawn as a potential future target.
32 Kd3 Rb4 33 Kc3 Rf4 34 f3
There was no urgency to weaken his kingside dark squares and pawns like
this.
34 ... Ne3 35 Kd3 Nd5 36 Ke2 Rf5 37 Kf2 Nf4

38 Nd4?!
More accurate is 38 Ke3! since after 38 ... g5, Black’s rook is denied use
of the square now occupied by his g-pawn.
 38 ... Rg5 39 Kf1 Rd5 40 Nb3 Rb5
I would try my luck in the knight ending after 40 ... Rxd2 41 Nxd2 Kf8
since White’s king is tied to g2.
41 Nd4 Ra5 42 a3 Rd5 43 Kf2 Rg5 44 Kf1 a5 45 Nc2
Instead, 45 Ne2! is better as Black’s knight is the sole irritating factor in
the white king’s otherwise untroubled life, so it makes sense to eject it from
the f4-square.
45 ... Rc5 46 Ne3 Kg7 47 Rd4 g5 48 b4?!
An inaccuracy. White hopes to exchange off the queenside pawns. White
should hold the game after the safer 48 Rd6 Re5 49 Kf2 b5 50 b4 axb4 51
axb4.
48 ... Rc3

Exercise (critical decision): Should White play 49 Nd5 or

49 Nf5+? One line should hold the draw, while the other loses.

49 Nf5+?
Fatigue is that strange place where the real and illusory merge in our
minds, producing strange outcomes. Now the position verifies White’s most
pessimistic suppositions. Soon it feels as if his drawing chances suddenly got
sucked into a crevice of a cosmic wormhole, to be displaced on another chess
board in some other part of the universe, 100 million light years away.
Answer: He should avoid this check and play 49 Nd5!. With this prodding mechanism, White hopes
to expedite the powerful f4-knight’s departure and should still just barely draw. After 49 ... Nxd5 50
Rxd5 Rxa3 51 Rxg5+ Kf6 52 Rb5 a4 53 Ke2! Rb3 54 Rxb6+ Ke5 55 Ra6 a3 56 Kd2 Kf4 57 Kc2 Rxb4
58 Rxa3 Kg3 59 Kc3 Rb1 60 Kd4 Kxg2 61 f4 Rh1 62 Ra7 Kxh3 63 Rxf7 Kg4 64 Ke5 Re1+ 65 Kd4 h3
66 Rh7 Rf1 67 Ke5 Rf2 (67 ... Rxf4 is met with the familiar drawing trick 68 Rxh3) 68 Rg7+ Kf3 69
Rh7 Kg3 70 Rg7+, Black can’t make progress.
49 ... Kf6 50 Nd6 Rxa3 51 bxa5 Ra1+ 52 Kf2 Ra2+
Targeting the g2-pawn.
53 Kf1 bxa5 54 Ne4+ Ke5!
Double attack. White loses material as the d4-rook and g2-knight hang
simultaneously.
55 Rc4 Nxg2
White’s kingside pawns haven’t exactly been a flowering civilization.
56 Rc5+
This allowed Black’s king entry into White’s camp but there wasn’t much
choice, since 56 Nxg5?? walked into the 56 ... Ne3+ knight fork.
56 ... Kf4 57 Nxg5 f5 58 Ne6+ Kg3
It almost feels as if Black is up a piece in this ending – an extra king!
59 Nd4
Also, 59 Rxf5?? Ne3+ is the same old story, since White’s rook gets
forked the other way.
59 ... Ne3+ 60 Ke1 Kxh3
White’s treasury is plagued with insufficient funds to bankroll the defence
of the country. Black’s newly passed h-pawn will be decisive.
61 Rc3
Also futile is 61 Nxf5 Nxf5 62 Rxf5 Kg2 63 f4 h3 64 Rg5+ Kf3 65 Rh5 h2
66 f5 h1R+ 67 Rxh1 Ra1+ and Black wins.
61 ... f4 62 Ne2 Ra1+ 0-1
After 63 Nc1 (63 Kf2 Rf1 and 63 Kd2 Rd1 both end the game
immediately), it’s disconcerting to expect an ambush from an enemy who
you expect from the North, come trotting up behind and tapping you on the
shoulder from the South with 63 ... Kg2 and Black’s h-pawn forces its way
down to the queening square.
Summary: When White plays 5 c3, watch out for the Qa4+ cheapo. Otherwise the move gives us
no trouble at all.

Game 31
V.Samolins-D.Jakovenko
Istanbul Olympiad 2012

1 e4 e6

Question: What?

Answer: Please remain calm. Our Petroff line actually transposes to a line of the Exchange French.
Our move order runs 1 ... e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 d3.

White refuses to enter one of the main lines and in this manner essentially
forces Black into a drab line of the Exchange French, into which we soon
transpose. 5 ... Nf6 6 d4 d5 7 Bd3.
2 d4 d5 3 exd5 exd5 4 Nf3 Nf6 5 Bd3
 And here we transpose to the above note’s line. Not only is this an
Exchange French, but it is a rather boring version since Black is unable to
alter the symmetry even a bit, by developing his g8-knight to the e7-square.
5 ... Be7
This move takes the sting out of any future Bg5. 5 ... Bd6 maintaining the
symmetry, is Black’s most common response. After 6 0-0 0-0 7 h3 h6 8 c4
(an attempt to alter the symmetry, similar to what Jakovenko did in our
game) 8 ... dxc4 9 Bxc4 Nc6 10 Nc3 a6 11 Re1 Re8 12 Be3 b5!? (Giri
enterprisingly grabs some queenside space at the risk of giving White
opportunities for ideas like a future a2-a4) 13 Bf1 Bb7 14 Rc1 Ne7 (Black
has an absolute clamp on the d5-blockade square) 15 Ne5 Ned5 16 Nxd5
Nxd5 17 Bd2 Qf6 18 Qh5 c5 (undermining the e5-knight) 19 a4 c4 (19 ...
cxd4 20 Ng4 Rxe1!? 21 Nxf6+ Nxf6 22 Qh4 Re4 23 g4 Re6 and Black has
huge compensation for the queen) 20 Re4 Bxe5 21 dxe5 Qg6 22 Qh4 Nf6 23
Ree1 Ne4 24 Bc3, Black already has the advantage with a grip on the light
squares and the superior pawn majority, M.Carlsen-A.Giri, Biel 2012.
6 h3
Otherwise ... Bg4 can be annoying. Instead, 6 0-0 Bg4 7 Bf4 0-0 8 h3 Bh5
9 Nbd2 Nbd7 10 Re1 c5!? 11 c3 Re8 12 dxc5 Nxc5 13 Bb5 Rf8 14 Qe2 Bd6
15 Bxd6 Qxd6 16 Qe3 a6 17 Bf1 Rfe8 18 Qd4 Ne6 19 Qa4 b5 has Black
achieving an excellent isolani position and with at least even chances,
T.Luther-R.Vaganian, Istanbul 2003.
6 ... 0-0 7 Nc3!?
The idea may be to shift the knight over to the kingside via the e2-square
later on. Normally White develops Nd2 instead. After 7 0-0 Nbd7 8 Re1 Re8
9 Bf4 c6 10 c3 Bf8 (I would unravel with 10 ... Nf8) 11 Nbd2 g6!? 12 Qc2 a5
13 a4 Bg7 14 Nf1 Nf8 15 Rxe8 Nxe8 16 Qd2 Nd6 17 Re1 Be6 18 Ng5 Qd7
19 Nxe6, the bishop pair doesn’t mean much in such symmetrical structures.
In fact, I often encourage such unbalancing swaps when I’m Black in the
Exchange French against a lower rated opponent who is intent upon a draw.
Following 19 ... Nxe6 20 Be5 f6 21 Bg3 Re8, Black equalized, A.Ivanov-
G.Kamsky, Internet 2014.
7 ... c5!?
Jakovenko, the higher rated player, creates an imbalance in the structure
and willingly takes on an isolani.

Question: If we are happy with a drawish position, how should Black play it?

Answer: The following game is a good example of Black playing it ultra-solid: 7 ... c6 8 0-0 Re8 9
Re1 Nbd7 10 Bf4 Nf8 11 Bh2 Ne6 12 Rb1 Bd6 with a dead even position, I.Kalogeris-P.Michelakos,
Paleochora 2013.
8 dxc5 Bxc5
The bishop recaptures the pawn and fights for the key d4- and e3-squares.
Also logical is to play 8 ... Nbd7 to recapture with the knight. After 9 0-0 (of
course 9 b4? is a futile attempt to hang on to the c5-pawn as it merely creates
weakness in White’s camp after 9 ... a5 10 b5 Nxc5) 9 ... Nxc5 10 Be3 Bd7
11 Nd4 Re8, N.Murshed-Z.Rahman, Dhaka 2012, Black looks very active
and has dynamic equality after 12 Re1 a6.
9 0-0
Not 9 Bg5?! as it loses the initiative after 9 ... Re8+ 10 Ne2 Nc6 11 0-0 h6
12 Bh4 g5! 13 Bg3 Ne4 when I already prefer Black’s position. His activity
clearly outweighs the potential for long term structural weakness.
9 ... h6 10 Bf4 Nc6 11 a3 Be6 12 Qd2 Re8 13 Rfe1
A theoretical novelty, and not an earth-shattering one. It looks slightly
more logical than the previously played 13 Rae1 which looks like the wrong
rook as the f1-rook now looks awkward. Of course it won’t matter which
rook White plays to the e1-square if White doubles rooks on the e-file,
M.Thesing-M.Schaefer, German League 1992.
 13 ... Rc8 14 Rad1 a6 15 Ne5
Swaps tend to favour White since he owns fewer pawn islands.
15 ... Nxe5 16 Bxe5 Nd7 17 Bg3 Nb6!
Targeting the c4-square for his knight.
18 Ne2
Also logical is 18 Be5 Qg5 19 Qxg5 hxg5 20 Ne2 Nc4 21 Bxc4 dxc4 22
Bd4 Bf8 23 Bc3 f6 and now I would try 24 Bb4 to rid Black of the bishop
pair.
18 ... Qg5 19 c3
Swapping queens would help Black, since he then covers the f4-square
with his g5-pawn.
19 ... Nc4 20 Qxg5!?
I think this is a concession, and prefer 20 Bxc4 dxc4 21 Nd4 Qxd2 22
Rxd2 Bd7 23 Kf1 when White has a grip on the dark squares, neutralizing
Black’s bishop pair. I don’t believe he is in any serious danger of losing.
20 ... hxg5 21 Bxc4 dxc4 22 Nd4 Bd7 23 Rxe8+ Rxe8 24 Nf3?!
When one player refuses to recede from an untenable intent, it’s as if he is
from an alternate universe with correspondingly alternate laws of physics.
This move accords another slight concession. The double attack on d7-bishop
and g5-pawn is ineffective, since White can’t actually grab the g5-pawn
without allowing Black’s rook entry to the seventh rank. Why retreat a well-
posted piece when he can play 24 Kf1, intending Re1?
24 ... Bc6
With the threat: ... Re2.
25 Kf1
Not 25 Nxg5? f6 26 Nf3 Re2 27 Rd8+ Kf7 28 Nd4 Rxb2 29 Nxc6 bxc6 30
Rc8 Ke6 31 Rxc6+ Kd5 32 Rxa6 Rb3 when the c3-pawn falls and Black’s
passed c-pawn is obviously more dangerous than White’s inferior a3-
specimen, since Black’s king will help support it down the board.
25 ... f6 26 Bc7 Kf7 27 Rd8?
Almost any other move is better. Black’s winning chances now
dramatically increase with rooks off the board, and we get the nagging sense
that a drawing opportunity slowly slips from White’s grasp since this allows
Black’s king access to a powerfully centralized position, and the b2-pawn is
fatally weakened.
27 ... Rxd8 28 Bxd8
 Exercise (critical decision): Should Black return the bishop pair with 28 ... Bxf3, or should he hang
on to the bishop pair. Evaluate the bishop versus bishop ending.

Answer: We gauge the bishop ending as won for Black, for the following reasons:
1. All of White’s queenside pawns are fixed on the wrong colour of his
remaining bishop.
2. White also has to worry about his now weak kingside pawns.
3. Black’s king is more active.
28 ... Bxf3! 29 gxf3 Ke6 30 Bc7
Also losing is 30 Ke2 Kd6! and now if 31 Ba5 b6! 32 Bb4 Bxb4! (the king
and pawn ending is a win for Black) 33 axb4 Ke5 34 Ke3 g6 (zugzwang).
Now 35 f4+ (35 Ke2 allows a second deadly zugzwang with 35 ... Kf4 and
it’s game over) 35 ... gxf4+ 36 Kf3 Kf5 37 Ke2 b5 38 Kf3 g5 39 Ke2 g4 40
hxg4+ Kxg4 41 Ke1 Kf3 42 Kf1 f5 (zugzwang yet again) 43 Ke1 Kg2 44
Ke2 f3+ 45 Ke3 f4+ and Black wins.
30 ... Kd7! 31 Bg3
Futile is 31 Bb8 Kc6 32 Ke2 Bd6 and Black wins the king and pawn
ending, since 33 Ba7? is met with 33 ... b6, trapping the bishop.
31 ... Bd6!
The bishop is a staunch egalitarian, just as long as the word is applied to
others and not himself.
32 h4
The pawn ending after 32 Bxd6 Kxd6, as previously discussed, is lost.

Exercise (planning): An enemy can be defeated either through a sudden,

unexpected strike, or through the process of patient attrition. Black in
this case undertakes the latter. Work out a clearly winning line for Black:

Step 1: Deflect White’s bishop temporarily off the h2-b8 diagonal.

32 ... gxh4! 33 Bxh4
Step 2: Infiltrate the bishop to the c1-square, after which White loses
decisive material.
33 ... Bf4! 34 Ke2 Bc1 35 a4 Bxb2 36 Kd2 Ba3 37 Ke3 Bc5+ 38 Ke4
Ke6 39 f4 f5+
Principle: Fix your opponent’s pawn on the same colour of his or her
remaining bishop.
40 Kf3
 Exercise (planning): How to maximize production,
while minimizing labour costs? Force Black’s win.

Answer: Simplification. The king and pawn ending is trivially easy.

40 ... Be7! 0-1
White is lost after 41 Bxe7 (41 Bg3 Bf6 picks off the c3-pawn) 41 ... Kxe7
(do you get the feeling that we are covering the same ground over and over
again?) 42 Ke3 Kd6 43 Kd4 b5 44 axb5 axb5 45 f3 Kc6 (zugzwang) 46 Ke3
(46 Ke5 b4 and a black pawn promotes) 46 ... Kc5 47 Kd2 (White’s king is as
outmatched as that Star Trek episode when Captain Kirk had to fight Mr.
Spock with that ridiculous net and trident against an enraged Spock’s sword
and shield, mixed with his Vulcan super-strength) 47 ... b4 48 cxb4+ Kxb4 49
Kc2 c3 50 Kc1 Kb3 51 Kb1 c2+ 52 Kc1 52 ... g6! (certainly not the
boneheaded 52 ... Kc3?? which is stalemate) and White is in zugzwang He is
forced to abandon the c1-blockade and ... Kb2 comes next, forcing
promotion.
Summary: The annoying 5 d3 line forces our side into a really boring,
painfully symmetrical line of the Exchange French. If you are happy with a
draw, then play solidly, with a future ... c7-c6. If you have dreams of
winning, then willingly take on an isolani with an eventual ... c7-c5, as
Jakovenko displayed in this game.
Chapter Five
The New Main Line: Shirov’s
Variation

5 Nc3, a once obscure variation, was given new life in 2002 by Shirov at the Russia vs. Rest of the
World match, where he electrified the chess world with wins in the ‘harmless’ line over GMs
Morozevich and Motylev. So the 5 Nc3 line, which was originally designed to dodge the theoretically
nightmarish lines covered in Chapters Two and Three in this book, soon soared in popularity. The line,
also called the Nimzowitsch Attack line of the Petroff (although it should obviously be called the
Shirov Variation), has had main line status thrust upon it, mainly through the advocacy of most of the
top players in the world. In this chapter we have a split, where White develops his or her bishop on
either the e3- or f4-square. We need to understand the differences thoroughly. The position, which
looks rather vacant right now, may soon teem with hidden life unseen by the eye, like a desert
landscape – especially if Black decides to castle short and turn the game into an opposite wings attack
situation. Some of our overconfident opponents may believe that psychological problems are merely
symptoms of repressed desire, and therefore remove all constraints of ‘right’ and ‘wrong’ by playing
fast and loose with their structure. They may do this by launching a kingside pawn storm with h2-h4,
f2-f3 and g2-g4. This can be intimidating at first, but let’s not equate lawlessness with freedom.
White’s future kingside pawn pushes can easily morph into overextension later in the game if our
opponents fail to deliver mate.

The ... b6 Factor

Probably the most important new development for our side is a setup for Black, which features a
queenside fianchetto:

This is Caruana-Wang-Hao, Bucharest 2013, where the Chinese GM

pulled off an upset of the world’s number two ranked player with his new
idea.

The next diagram is Baramidze-Akobian, Antalya 2013, which was played just a few days after
Wang Hao’s game. This position is a near-superimposition, except that White’s bishop sits on the f4-
square rather than on the e3-one. Having looked closer at the early ... b6 ideas, I couldn’t come up with
even a tiny edge for White, no matter how hard I whipped the comps. So pay close attention to those
two games from the chapter.

Game 32
S.Karjakin-S.Mamedyarov
FIDE Grand Prix, Beijing 2013

1 e4
To digress a little, after 1 b3 e5 2 Bb2 Nc6 3 c4 Nf6 4 Nf3 e4 5 Nd4 Bc5 6
Nxc6 dxc6, Black’s ... dxc6 theme is similar to what we get in this chapter
(but I think Spassky got a better reversed version!), B.Larsen-B.Spassky,
Belgrade 1970. Spassky went on to slaughter Larsen in this game, which
probably hurt sales of my Nimzo Larsen Attack: Move by Move (1 b3) book!.
1 ... e5
Compare our position in the Petroff with the following Nimzowitsch
Sicilian line with 1 ... c5 2 Nf3 Nf6 3 e5 Nd5 4 Nc3 Nxc3 5 dxc3. I think our
Petroff’s version is much more solid for Black.
2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3

Which actor or actress isn’t nervous on the eve of opening night? We enter
another scary place, the Shirov Attack, currently the darling anti-Petroff line
of many of the top players in the world.
5 ... Nxc3
 We inflict slight harm on White’s structure.
6 dxc3
White’s overwhelming choice.

Question: What exactly does White get for allowing us structural superiority?

Answer: White gets virtually unlimited freedom of movement for his pieces (one thing to note is
that even though White gets open lines, he or she lacks centre pawns, which means we can place a
piece on the e6-square, for instance, without fear of d4-d5) and two open central files. These two can
easily translate into potential attacking chances if we fail to conduct the defence with accuracy. We
cover 6 bxc3 in the Odds and Ends, in the final chapter of the book.
6 ... Be7
Black’s most flexible move.

Question: How is 6 ... Be7 any different from 6 ... Nc6? Won’t the lines transpose?

Answer: Not necessarily. In the latter half of the chapter, when we cover the 7 Bf4 lines, where
Black’s main idea is to play ... Nd7! intending ... Nc5-e6, which gains a tempo on White’s f4-bishop. If
we play a too-early ... Nc6, then we deny ourselves this possibility.
7 Be3

This is considered by GM Sakaev as White’s most dangerous line. We

arrive at a great divide. For the second half of the chapter we cover 7 Bf4.

Question: What are the differences between 7 Be3 versus 7 Bf4?

 Answer: Some of the differences are:
1. By posting 7 Be3, White takes direct aim at the a7-pawn. This can be
important if Black decides to castle long, when the a7-pawn becomes a target.
2. By posting 7 Be3, if Black castles kingside, it clears the way for a
kingside pawn storm plan with Nd4, h2-h4, f2-f3 and g2-g4. If White’s
bishop were on the f4-square, it would tend to be in the path of White’s pawn
storm.
3. By posting 7 Bf4, White discourages a future ... Ne5, which can either
lead to swaps, which White doesn’t want, or ... Nxf3, inflicting further
damage to White’s structure.
4. By posting 7 Bf4, White is vulnerable to the above-mentioned tempo
gaining plan ... Nd7-c5-e6.
5. By posting 7 Bf4, White leaves the e-file open. This can be important if
Black plays …Be6, since then we may be vulnerable to Rhe1 and Nd4, or
Ng5 ideas.
7 ... Nc6
As mentioned above, Black often posts …Nd7 if White posts Bf4. Black’s
intent can be to take some of the fire out of White’s position by castling long,
which deprives White of the opposite wing attack he or she may have banked
on.
8 Qd2
White prepares to castle long.
8 ... Be6
We continue our policy of denying White opposite wing attacks by
keeping the option to our castle long. Nevertheless, 8 ... 0-0 is very playable
but gives White just what he or she wants after 9 0-0-0. With 9…Ne5, Black
takes advantage of the fact that White played Be3 by looking for a helpful
swap. After 10 h4 c6 (10 ... Nxf3 inflicts structural damage but is obviously
full of danger for Black after 11 gxf3 since it opens the g-file for White’s
attack) 11 Bg5 f6 12 Bf4 Qa5 13 a3, a sharp, opposite wing attack
middlegame will follow, F.Caruana-B.Gelfand, Wijk aan Zee 2012.
9 0-0-0 Qd7 10 Kb1
A useful move, covering the a2-pawn, while keeping options open. Taking
aim at the kingside with 10 Bd3 (10 h4 h6 11 Kb1 is covered in Game 34)
doesn’t make much sense if Black castles long with 10 ... 0-0-0 11 h4 h6 12
Kb1 a6 13 c4 (13 Nd4 Nxd4 14 Bxd4 Rhg8 15 Bh7 Rh8 16 Be4 Rhg8 17 g3
d5 is about even). Here, I play the typical 13 ... Ne5, menacing both the c4-
pawn and f3-knight, happy to chop on f3 and open the g-file, since my king is
on the other side of the board. Following 14 Nxe5 dxe5, I already prefer
Black’s position since I have the healthy kingside pawn majority,
“RiverDolphin”-C.Lakdawala, Internet (blitz) 2014.
10 ... h6
A useful move which cuts out both Ng5 and Bg5 ideas. Black can also try
10 ... Bf6 (we will examine 10 ... a6 in the next game):
a) 11 h3 (dual purpose: White cuts out ... Bg4 ideas, while preparing g2-g4
and Bg2 when Black castles long) 11 ... h6 (eliminating both Ng5 and Bg5)
12 b3 a6 13 g4 0-0-0 14 Bg2, F.Caruana-R.Ponomariov, Dortmund 2014.
Black achieves equality after 14 ... Kb8 15 Rhe1 Rhe8.
b) 11 Ng5 Bxg5 12 Bxg5 f6 13 Be3 0-0-0 (I’m not impressed with White’s bishop pair and feel that
Black equalized) 14 Bb5 a6 15 Ba4 b5! (Black’s queenside pawn weakness isn’t so easy for White to
exploit) 16 Bb3, A.Matikozian-C.Arvind, Los Angeles 2012 and Black’s position looks harmonious
after 16 ... Rhe8.
11 h3
I think this is White’s most logical plan.

Question: What is the intent?

Answer: White realizes the Black plans to castle long, and prepares a fianchetto so that both
bishops take aim at Black’s future king position on the queenside. 11 h4 is White’s main move.
Following 11 ... a6 12 Nd4 Nxd4 13 Bxd4,
 13…0-0! is correct as we don’t have to castle long 100% of the time. In
this case, Black’s queenside play is adequate. After 14 Be2 Rfe8 15 Bf3 (15
g4 Bf8 16 g5?! can be met with 16 ... Bxa2+! 17 Kxa2 Qe6+ 18 Kb1 Qxe2 19
Qf4 h5 when White lacks compensation for the pawn) 15 ... Bf8 (Black offers
the b7-pawn) 16 Qf4 (16 Bxb7?? completely plays into our hands after 16 ...
Rab8 17 Bf3 c5! 18 Be3 and now Black has 18 ... Bxa2+! with a quick kill),
White plans g2-g4-g5 if given time. The game continues with 16 ... d5 17 c4
(17 g4? is met with 17 ... c5 18 Be5 (18 Be3?? Bd6 traps White’s queen) 18
... f6 19 Bc7 Rac8 and the position is obviously bad for White) 17 ... Bd6 18
Be5 dxc4 (Black is probably also okay taking on an isolani with 18 ... Bxe5
19 Qxe5 c6 20 cxd5 Bxd5 21 Qf4 Qe6 22 Bxd5 cxd5) 19 Bxd6 cxd6 20
Qxd6, S.Zhigalko-S.Mamedyarov, Kocaeli 2014, when the position is
balanced after 20 ... Qe7.
11 ... a6
11 ... Bf6 transposes to the Caruana-Ponomariov note from above.
12 g4 h5!
 A new move and a good one, which reminds White that if you push too
many pawns, too far, you risk overextending. This move loosens White on
the light squares on the kingside. I think Mamedyarov’s move improves upon
12 ... 0-0-0 13 Bg2 g5 14 Nd4 Nxd4 15 Qxd4 c5 16 Qd2, V.Bologan-
V.Belov, Sochi 2007. I slightly prefer White’s side, since f2-f4 may follow.
13 gxh5!?
Karjakin splits his kingside pawns to generate play down the g-file. 13 g5
can be met with 13 ... 0-0!? since kingside attacking lanes are not open for
White. Now the natural sacrifice 14 g6?! looks dubious after 14 ... fxg6 15
Nd4 Nxd4 16 cxd4 Kh7 (intending ... Bf5) and Black hangs on to the extra
pawn.
13 ... Rxh5 14 Rg1 Bf6
Mamedyarov refuses to weaken his kingside dark squares. Riskier is 14 ...
g6 15 Ng5 when I slightly prefer White’s side.
15 Bg5 0-0-0!
Mamed invites doubled f-pawns, which reinforce his control over e5- and
g5-squares, while removing his g7-pawn from the line of fire.
16 Bxf6!?
I think Black stands better after this tempting move. The position remains
in tense balance after 16 h4.
16 ... gxf6 17 h4?!
 The h-pawn can’t be maintained on this square. 17 Nd4 looks like a better
method of giving up the h-pawn. After 17 ... Nxd4 18 cxd4 Bxh3 19 Bxh3
Rxh3 20 Qf4!, White regains the lost pawn and should hold the game after 20
... Rdh8 21 Qxf6 Qe6 22 Qf4 Qd5 23 Rg5 Rf3 24 Qg4+ f5 25 Qg2 Qe4 with
only a tiny edge for Black.
17 ... Ne5

Going after the defender of the h4-pawn, which pretty much wins the
pawn. Black stands better for the following reasons:
1. White’s h-pawn is doomed to fall.
2. Black owns a strong e5-outpost.
3. White’s kingside light squares are weak.
4. White must watch out for sudden lunges like ... Nxf3, followed by ...
Bxa2+, ... Ra5+ and ... Qa4.
18 Qe3
Dual purpose:
1. Karjakin is aware of the ... Bxa2+! ideas and clears the d2-square for his
king.
2. White menaces a Qa7 infiltration.
Instead, 18 Be2 gives Black a pleasant choice of nabbing pawns:
a) 18 ... Nxf3 19 Bxf3 Rxh4.
b) 18 ... Bxa2+ when White is unable to take the offered piece, since 19 Kxa2?? loses instantly to
19 ... Qa4+ 20 Kb1 Nxf3 and if 21 Bxf3 Ra5 when Black forces mate.
18 ... Kb8
Eliminating the a7 infiltration problem.
19 b3
Seizing control over the c4-square.
19 ... Bg4
The h4-pawn is a goner.
20 Be2 Bxf3
Black’s knight is the superior piece, so Black retains it.
21 Bxf3 Rxh4

Black won a pawn, which admittedly is doubled. More importantly, his

knight trumps the White bishop.
22 Bg2 Qf5 23 Rd4!
Preventing the coming ... Rf4.
23 ... Rg4?!
The comp doesn’t like this move and prefers 23 ... Rh2! 24 Rb4 d5 25
Bxd5 Qxf2 (25 ... Rxd5?? 26 Rg8+ would be a tragically stupid way to lose)
26 Rxb7+ Kc8 27 Qxf2 Rxf2 28 Be4 f5 29 Bh1 Rdd2 and the c2-pawn can’t
be defended, since 30 Rc1?? is met with the crushing shot 30 ... Nd3! 31 Bc6
Rxc2! forcing mate.
24 Rxg4 Nxg4 25 Qd4?
 This move allows Black to eradicate his doubled pawn and turn it into a
healthy model. White has much better chances to hold the ending after 25
Qh3! Re8 26 Bf3 Ne3 27 Qxf5 Nxf5.
25 ... Qxf2 26 Qxf2 Nxf2 27 Rf1 Ng4 28 Bh3 Ne5 29 Rxf6 c6
Black, owns a healthy extra pawn and holds excellent winning chances.
30 Rf2 Rg8 31 c4 Rg3
The most active post for Black’s rook is on the 6th rank.
32 Bf5 Kc7
Activating the king.
33 Rh2 Rf3 34 Rh5 a5
Giving his king a dark square path up the board.
35 Kb2 Rg3 36 Rh7 Kb6 37 Bc8
Harassing Black by preventing ... Kc5.
37 ... Rg8 38 Bf5 Kc5 39 Rh4
Cutting off ... Kd4, after which Black’s king could assist his f-pawn up the
board.
39 ... Rg2 40 Rf4

Karjakin defends well, maintaining the f5-blockade, while cutting Black’s

king off from entry to the d4-square.
40 ... Rg3 41 Bc8
The bishop doesn’t meet his target head on. Instead, he prefers to creep closer, silently from behind.
 41 ... b6
A slight concession but otherwise Black has trouble making progress.
42 a3 Rf3!
This move frees Black’s f-pawn to travel forward.
43 Rxf3!
The correct decision. Black wins easily if White keeps rooks on the board
with 43 Rh4? f5 44 Be6 f4 45 Bf5 b5 46 cxb5 cxb5 47 Be4 Rf2 48 Ba8 Re2!
49 Kc1 (49 Rxf4?? walks straight into 49 ... Nd3+) 49 ... Kd4! 50 Kd1 (50
Rxf4+?? Kc3 51 Kd1 Rxc2 is an easy win for Black, who threatens the b3-
pawn, as well as ... Nd3/ ... Rc1 mate!) 50 ... Rf2 51 Rh3 f3 and White can
resign.
43 ... Nxf3 44 Kc3
“Oh please, oh please ... ” White thinks up a sneaky little threat!
44 ... d5
Black would require years of therapy and antidepressant prescriptions if he
played the boneheaded helpmate 44 ... Ne5?? 45 b4+ axb4+ 46 axb4 mate
when his pain is enhanced exponentially by the fact that a pawn delivered the
mate.
45 b4+ axb4+ 46 axb4+ Kd6 47 cxd5
Every pawn swapped increases White’s drawing chances.
47 ... cxd5!
Good judgment. This move seizes control over the centre and gives Black
potential knight outposts on the c4- and e4-squares, while clearing the c6-
squares for his knight to attack White’s advanced b-pawn.
 48 Kd3 Ne5+ 49 Ke3 Nc6 50 c3 Ke5
Add ‘superior king position’ to Black’s advantage list. Black must play
exceedingly careful to win, since he has only three pawns remaining on the
board – a fact which greatly enhances White’s chances to hold the game,
since if White manages to swap queenside pawns, the game is drawn.
51 Kd3 Ne7 52 Bb7
The resentful bishop departs the knight’s company after only the most
perfunctory bow allowed by court protocol.
52 ... Nf5 53 Bc8
Most definitely not 53 c4?? as 53…dxc4+ 54 Kxc4 Nd6+ picks off the
bishop.
53 ... Nd6 54 Bg4 f5
It’s time for the f-pawn to move.
55 Bf3 f4 56 Bh5 Nc4 57 Bf3 Ne3 58 Be2 Ng2
 Exercise (critical decision): Can White draw if he plays 59 c4?

59 Bg4
Answer: It’s a trap! Black wins after 59 c4?? f3! 60 Bxf3 (60 Bf1 dxc4+! and White is unable to
recapture, since Black has a …Ne3-fork) 60 ... Ne1+ 61 Ke3 Nxf3 62 cxd5 Kxd5! (the simplest. Black
forces a won king and pawn ending) 63 Kxf3 b5 64 Ke3 Kc4 65 Kd2 Kxb4 66 Kc2 Ka3 67 Kb1 Kb3
and we learned as kids, the pawn down side loses in this king and pawn ending.
59 ... Ne3 60 Be2 Nc4 61 Bf3 Nd6 62 Bh5 Ne4
 Exercise (critical decision): Same question as
last time: Can White get away with 63 c4?

Answer: Timing is everything in chess. This time it works. White activates his queenside pawn
majority and should hold the game.
63 c4! d4
Black’s only prayer to achieve the full point. After 63 ... dxc4+ 64 Kxc4
Nd6+ 65 Kd3 (this is almost a certain draw) 65 ... Nf5 66 Kc4 Nd4 67 Kd3, I
don’t see a way for Black to make progress.
64 Bf3 Ng5
 Exercise (planning/critical decision): Which diagonal should the bishop remain on?
He can play 65 Bb7, or 65 Bg4. Be careful. Only one of the diagonals
holds the draw.

65 Bg4??
The once stalwart defence begins to show signs of decay. Karjakin, after
prolonged heroic defence, goes astray.
Answer: The h1-a8 diagonal is the drawing key. 65 Bb7! holds the game after 65 ... f3 66 c5 bxc5
67 bxc5 and Black is in an odd zugzwang. After 67 ... f2 68 Ke2 Nh3 69 Bg2 Nf4+ 70 Kxf2 Nxg2 71
Kxg2 Kd5 72 Kf2 Kxc5 73 Ke2 Kc4 74 Kd2, White holds the draw by a tempo.
65 ... f3 66 c5 f2!
Pawn promotion is the hub of both sides’ strivings.
67 Ke2
Also insufficient is 67 Be2 bxc5 68 bxc5 Ne6 69 Bf1 (69 c6 Nf4+ 70 Kd2
Kd6 allows Black to win a second pawn, and with it, the game) 69 ... Nf4+ 70
Kc4 d3! 71 Kc3 Kd5 and Black wins.
67 ... bxc5 68 bxc5 Kf4!
The key to Black’s win: he gains a precious tempo on the bishop.
69 c6
White’s best chance as 69 Bc8 Ne4 (threat: ... d4-d3+) 70 Bh3 Kg3 71 Bf1
Nxc5 wins for Black.
 69 ... Kg3!
White reels from the impact from this sudden blow from behind. Most
certainly not 69 ... Kxg4?? (a move made with the philosophy: When a shady
person tries to sell you stolen property, sold at 10 cents to the dollar of its
actual worth, then it’s best not to ask where the item came from) 70 c7 d3+
71 Kxf2 Ne4+ 72 Ke3 Nd6 73 Kxd3 and it’s a draw.
70 c7
Otherwise, 70 Kf1 d3 71 c7 Ne4! 72 c8Q Nd2 is a sweet mate.
70 ... Kg2
White promotes first, but Black does it with check.
71 c8Q f1Q+ 72 Kd2 Ne4+
Black’s queen, knight and pawn make a deadly attacking trio.
73 Kc2 d3+ 74 Kb2

Exercise (combination alert/planning): Work out a mating pattern for Black.

74 ... Kg3?
Serious doubts begin to colour Black’s decisions. This is a slip which
allows White very real chances to hold the game.
Answer: 74 ... Qf2+! wins. The trick is to weave the queen to the a-file after which White’s king is
cornered:
a) 75 Ka3 Qa7+ 76 Kb2 Qb6+ 77 Kc1 Qa5! (threatening mate on the
move) 78 Kb1 Qb4+ 79 Ka1 (79 Ka2 Nc3+ wins) 79 ... Qa3+ 80 Kb1 Nd2
with mate!
b) 75 Kb3 Qb6+ (this pushes the king to the a-file, as Kc4 allows a …Nd6
fork) 76 Ka2 Qa5+ 77 Kb2 Qb4+ and we reach the above winning line.
c) 75 Kc1 Qd2+ 76 Kb1 Qb4+ and we return to the winning line once
again.
75 Qf5! Qg2+
The queen’s face registers her utter astonishment of her f5-sister’s
intrusion by the fact that her left eyebrow arced one-sixteenth of an inch.
There is still room to bungle with 75 ... Qf2+? 76 Qxf2+ Kxf2 77 Kb3 Ng3
78 Kc3 Ke3 79 Bd1 Ne2+ 80 Kb2 Kd2 81 Ba4 is drawn, despite Houdini’s
irrational insistence (sans tablebase) that it is a win for Black!
76 Ka3 Qh1 77 Bd1!
The bishop’s lifelong guiding principles have always been humility and
brazen self-interest. In this case, he decides that spiritual values must be
subordinated to political opportunism.
77 ... Qxd1
Forced, since 77 ... Nf2 78 Qg5+ is perpetual check.
78 Qxe4 Qe2

Black’s queen calls the meeting to order with a high degree of impatience.
 Question: Is this position a win for Black, or a draw?

Answer: According to the omniscient endgame databases, the position is drawn. But I should add,
that it is drawn for comps but not necessarily so for humans. I had just such a pawn up ending a few
weeks ago, and my 2460-rated opponent was unable to hold the game in the time scramble. I felt like I
just made a bunch of random moves until he fell into a trap. Such endings are a nightmare for the
imperfect human mind, since after every move it feels as if our opponent has access to dozens of
variables, which almost renders our analysis meaningless.
79 Qg6+ Kf2 80 Qf6+ Qf3 81 Qd4+ Ke2 82 Qe5+ Kf2 83 Qd4+ Kf1 84
Qa1+ Kg2 85 Qg7+ Qg3 86 Qb7+ Kf2 87 Kb2
This is not really necessary. After 87 Qb6+ Qe3 88 Qd6 and if 88 ... d2+
89 Kb2 Ke1 90 Qb4 Ke2 91 Qb5+, White holds the game.
87 ... Ke2 88 Qe4+ Qe3 89 Qg2+ Qf2 90 Qe4+ Kd2 91 Kb3?
I suppose irony is perhaps nature’s way of denying us absolution via a
cruel practical joke. We tend to come to decisions on the chess board based
upon the following factors:
1. Data.
2. Inference.
3. Paranoid suspicions, both founded and unfounded.
It feels to me like White based his last move on number three in the list. In
some games, luck is as valuable an adjunct as raw skill. In games this long,
our mental process feels bogged in a viscous fog which caffeine is unable to
lift. “There were lots of moves which would save the game, for example 91
Qc4, 91 Qe6, etc. However, it’s hard to see what’s wrong with White’s move
– perhaps his queen was not that well positioned on e4?” writes GM Alex
Baburin.
 91 ... Qb6+! 92 Kc4 Qa6+ 93 Kb3
White puts up greater resistance after 93 Kb4 Kc2 94 Qh7 Qc6 95 Qf5
Qc3+ 96 Ka4 Qc4+ 97 Ka5 Kb3 98 Qf3 Qc3+ 99 Kb5 d2 100 Qd5+ Kb2 101
Qg2 Qd3+ 102 Ka5 Ka3 and Black queens the pawn.
93 ... Qb5+!
The queen enlarges her large, moist cocker spaniel eyes, beguiling White’s
king, who suspects nothing.
94 Ka2 Kc3!
Threatening mate.
95 Qe1+ Kc2 0-1
The most terrifying sight to a stage actor or actress is to witness the
audience leaving in mid-performance. Black’s king slips away in just such a
fashion. When we just barely win such an epic game against a very strong
opponent, we enter both heaven and nirvana. After 96 Qf2+ (96 Qc1+ Kxc1
is unfortunately for White, not stalemate after 97 Ka3) 96 ... d2, White is out
of checks, while his king is out of luck.
Summary: In the Shirov Attack line, one of our weapons is to keep our opponents wondering
which side we eventually castle. So in some cases, allow your king to bide his time and keep to the
shadows for as long as possible.

Game 33
V.Ivanchuk-Wang Yue
MTel Masters, Sofia 2009
1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Be3 Nc6 8 Qd2 Be6 9 0-0-0 Qd7
10 Kb1
After 10 h3 (we will look at 10 h4 in the next game) 10…a6 11 b3 0-0-0
12 Ng5 Bxg5 13 Bxg5 f6 14 Be3 Rhe8 15 c4 Qf7 16 Qc3 Bf5 17 Be2 Be4 18
Rhg1 Kb8 19 Kb2 h6 20 Bd3 Qg6 21 g4 Re7 22 Rge1 Rde8, White played
the overoptimistic 23 b4?! (the position remains balanced after 23 Bxe4 Rxe4
24 c5 dxc5 25 Bxc5 Rxe1 26 Rxe1 b6 27 Rxe8+ Qxe8 28 Bf8! Qxf8 29
Qxc6) 23 ... Bxd3 24 cxd3, “kromos”-C.Lakdawala, Internet (blitz) 2014.
White looks a shade overextended after 24 ... Ne5.
10 ... a6

Black cuts off Bb5-pin ideas.

11 Be2
White refuses to rush into a committal move and simply continues to
develop. The alternatives may weave and merge into one another:
a) White’s main line is 11 h4 and after 11 ... h6 12 Nd4 Nxd4 13 Bxd4 0-
0!, we reach S.Zhigalko-S.Mamedyarov, Kocaeli 2014, a note we covered in
the previous game and Black achieved fully equal chances (instead, 13 ...
Rg8 14 c4 0-0-0 15 Qc3 looks slightly more comfortable for White,
V.Bologan-A.Motylev, Poikovsky 2009).
b) We also saw examples of 11 h3 in the previous game, with the idea of
playing g2-g4, followed by Bg2. Following 11 ... 0-0-0 12 Ng5 Bxg5 (as I
mentioned before, I’m not intimidated by White’s bishop pair in any way,
since in this semi-symmetrical structure, they just don’t shine) 13 Bxg5 f6 14
Be3 Rhe8 15 b3 Bf5, a bishop pair lover may disagree, but chances look
perfectly even to me, V.Bologan-V.Gashimov, Poikovsky 2008.
c) White can pick up the bishop pair with 11 Ng5 Bxg5 (please see the
above note about my disdain for White’s bishop pair!) 12 Bxg5 f6 13 Be3 0-
0-0 14 b3 Kb8 15 h3 Bf5 16 Be2, E.Inarkiev-V.Gashimov, Elista 2008 when
Black is in good shape after 16 ... h5 17 Rhe1 g5, securing the e5-square for
the knight.
d) White can also put the question to the c6- and e6-minor pieces with 11
Nd4 after which we immediately address the situation with 11 ... Nxd4 12
Bxd4 0-0! (we are unintimidated by the prospect of opposite wings attack,
since ours comes quickly) 13 h4, I.Chirila-N.Abasov, Bucharest 2009. Here I
would continue with 13 ... Rac8 14 f3 c5 15 Bf2 Rfe8 16 g4 d5 and if given a
choice, I prefer Black, since he controls the centre.
11 ... Bf6!?

Question: Doesn’t this move lose a tempo if White plays Ng5 next?

Answer: It does, but Wang Yue is so unintimidated by the Ng5 idea that he actually encourages it.
If the tempo loss bothers you, then you have two perfectly viable alternatives:
a) 11 ... 0-0-0 12 Nd4 Bf6 (risky is 12 ... Nxd4!? 13 Qxd4 Bf6 14 Qa7 Qc6
15 Bf3 d5 16 Bd4 Bxd4 17 Qxd4 g6 and even here, Black’s position looks
fully playable, but I prefer White) 13 Nxe6 fxe6 14 Rhe1 and in this version,
I do slightly prefer White due to the potential of the bishop pair, K.Bhatia-
O.Smith, London 2012.
b) Black can invest a move to eliminate Ng5/Bg5 issues with 11 ... h6 12
Nd4 Nxd4 13 Bxd4 and here I would castle kingside, with dynamically
balanced chances, R.Edouard-V.Ivanchuk, Belfort 2012.
12 Ng5
Believe or not, this is a new move in the position. The move makes perfect
sense, now that Black is willing to hand over a tempo.

Question: Does White achieve anything with 12 Bg5?

Answer: I don’t think so.

Black looks even after 12 ... Bxg5 13 Nxg5 0-0-0, M.Olszewski-
B.Macieja, Chotowa 2009. Once again, I don’t believe White’s bishop is
better than Black’s knight after 14 Nxe6 fxe6 15 Bg4 Kb8 16 Rhe1 Rde8 17
Qe3 Rhf8! and now if 18 Bxe6 Qd8 19 f4 Rf6 20 Qh3 (20 f5 Rxf5 21 Bxf5
Rxe3 22 Rxe3 Qg5 23 Re8+ Ka7 24 Bh3 Ne5 is about equal. I think Black’s
queen and knight, a harmonious pair, can hold their own against White’s two
rooks and bishop) 20 ... Rh6 21 Qg4 Rg6 with a repetition draw.
Also, after 12 Rhe1, C.Eluekezie-H.Ble Kouko, Lagos 2014, it’s logical to
play 12 ... 0-0! since now if White wants to attack with a kingside pawn
storm, his rooks are misplaced in the centre.
12 ... Bxg5 13 Bxg5 f6 14 Be3 0-0-0
 Question: Don’t you think White has some winning chances with his bishop pair?

Answer: As mentioned above, my feeling is that White is hard-pressed to prove that his bishop pair
is meaningful in this close-to-symmetrical structure. I feel like Black’s drawing chances are far higher
than White’s winning ones.
15 Rhe1 Rhe8 16 b3 Bg4
Wang Yue, perhaps secretly playing for a win, encourages Ivanchuk to
push his pawns forward. Also logical is 16 ... Re7 intending to double on the
e-file.
17 f3 Bf5 18 Bf1 Bg6 19 Kb2 Re5
Now he encourages f3-f4, which would weaken White’s control over the
e4-square.
20 Bf2 Qf5!?
The beginning of a risky plan. I would just play 20 ... Rde8.
21 Bg3 Rxe1 22 Bxe1 Qg5!?
 This radical move – the idea behind his 20th move – is played with the
philosophy: If you ask for nothing from your position, then that is what you
get. So Wang Yue decides that just because the position is bland, this doesn’t
give his side license to think the lethargic thought: “Any move will do”. So
he decides upon a risky, disruptive idea to remove queens from the board. By
altering the structure like this, Black slightly enhances the latent power of
White’s bishop pair. I would have left matters alone and just shuffled about
with 22 ... Kb8 23 Bf2 Re8.
23 Qxg5
The queen tosses the g5-square’s previous occupant overboard.
23 ... fxg5 24 c4
Clearing the way for Bc3. I slightly prefer White’s position here.
24 ... Ne7
The knight decides upon a circuitous route to cover the g7-pawn.
25 Bd2
Encouraging ... h6 to place Black’s kingside pawns on vulnerable dark
squares.
25 ... h6 26 Bc3 Nf5 27 Re1
Covering against ... Ne3. Less accurate is 27 g4? Ne3 28 Re1 Nxf1 29
Rxf1 Rg8 and the opposite coloured bishops virtually insure a draw.
27 ... Re8 28 Rxe8+ Bxe8 29 Bd3
 White’s ‘advantage’ is theoretical, more than actually real. After 29 g4
Nh4 30 Bxg7 Nxf3 31 h3 h5 32 Bf6 hxg4 33 hxg4 Bd7 34 Be2 Bxg4 35
Bxg5 Nxg5 36 Bxg4+, there is a 99.99% probability of a draw.
29 ... g6 30 Kc1 Nh4
The knight dances about, harassing White at his g2- and f3-bases.
31 Bf1
Also, after 31 Bg7 Nxg2 32 Bxh6 Nh4 33 Be4 Bc6 34 Bxg5 Nxf3 35 Bxc6
bxc6 36 Bf4 g5 37 Bg3 Kd7, Black stands no worse.
31 ... Kd7 32 Kd2 Ke6 33 Ke3 Nf5+ 34 Kf2 Kf7 35 Bd3 Ne7 36 g4 Nc6
37 Be4 Nd8 38 Bd5+ Ke7 39 Bg7 h5 40 Bd4 Bc6!
Black’s indolent bishop always appears exhausted, which is a neat trick,
since he has never put in a hard day’s work in his entire life. Principle:
Eliminate your opponent’s bishop pair by swapping one of them away if
possible.
41 Be3
Following 41 Bg8 Ne6 42 Be3 hxg4 43 fxg4 Be4, if anyone stands worse,
it is White.
41 ... Kf6 42 Kg3
Obviously White has a draw if he wants one with 42 Bd4+ Ke7 43 Be3
Kf6.
42 ... Ne6 43 a4
 Winning a pawn with 43 Bxe6 Kxe6 44 Bxg5 hxg4 45 Kxg4 b5 leaves
White with no real chances to win the pawn up opposite coloured bishops
ending.
43 ... Ke5

Exercise (critical decision): White can be on the symbolically better

end of a draw with 44 Bxe6, followed by 45 Bxg5. But there also
arises a tantalizing winning-try sacrifice. Does 44 Bxg5 work for White?

44 Bxg5??
Answer: It fails miserably, turning a drawn game into a lost one in the space of a single decision.
44 ... Bxd5 45 f4+ Ke4 46 cxd5 Nxg5 47 fxg5
After 47 gxh5 Nh7 48 hxg6 Nf6, White’s pawns are going nowhere and he
will lose this.
 Exercise (combination alert): Both sides had foreseen this position,
but Ivanchuk clearly misread the portents. Black spotted a subtle
defensive breach in White’s defensive barrier. What did he play?

Answer: Deflection/zugzwang.
47 ... h4+!!
A time of sudden strife and bloodshed suddenly descends upon the board.
Sacrifices tend to be habit-forming. When we make one, we crave the feeling
so much, that we want to make another. Black’s last move is made under the
theory: Every thief becomes an honest person when there is nothing to steal.
The devious h-pawn decides to bung in an entire set of monkey wrenches into
the works. If not for this move, White would have won.
48 Kxh4
The king owns more wealth than he can spend in a lifetime. But what good
is it if he is diagnosed with a terminal illness? All he can do now is to meekly
observe the proceedings with a joyless face. Following 48 Kf2 b5! 49 axb5
axb5 50 Ke2 b4 51 Kf2 Kxd5 52 Ke3 Ke5 53 h3 d5 54 Kf3 Kd4 55 Kf4 c5
56 Kf3 c4 57 Kf4 cxb3 58 cxb3 Kc3 59 Ke5 d4 60 Kf6 d3 61 Kxg6 d2 62
Kh7 d1Q 63 g6 Qd3, the race isn’t even close. When we say the position is a
queen and pawn ending, we usually mean both sides have a queen!
 48 ... Kf3
“In the interests of procedural harmony, the council – namely, myself –
has decided to ‘excise’ you from your position as monarch. I will rule in your
stead as indefinite interim leader,” recites Black’s king. We witness an
incredibly rare example of a king and pawn ending, where the side with two
pawns extra loses.
49 b4
After 49 Kh3 (White’s king grudgingly complies with a muffled curse
under his breath) 49 ... b6 50 c3 Kf2 51 b4 b5 52 a5 Kf3 53 Kh4 Kg2 54 h3
Kf3, it is zugzwang. Unfortunately for White, he isn’t stalemated and
following 55 c4 bxc4 56 b5 c3 57 bxa6 c2 58 a7 c1Q 59 a8Q, 59…Qe1
delivers mate!
49 ... b5 50 a5 Kg2 51 h3 Kh2 0-1
Futile is 52 c3 Kg2 53 c4 bxc4 leading to the above mating line.
Summary: Don’t be afraid of White’s bishop pair in Ng5, …Bxg5 lines. In this game, Wang Yue
allowed it a full move down and still achieved a near-equality position.

Game 34
R.Ponomariov-Wang Hao
Kings Tournament, Bucharest 2013

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Be3 Nc6 8 Qd2 Be6 9 0-0-0 Qd7
10 h4
 This line is merely a tributary of a much larger river.

Question: What is the point of h2-h4, which doesn’t seem to threaten anything?

Answer: The main idea is that after Ng5, if Black plays ... Bxg5, then White can greatly enhance his
structure by recapture with the h-pawn with hxg5, which pries open the h-file.

Question: But can’t Black easily prevent this by tossing in ... h7-h6?

Answer: Correct, but this is what White wants, to create some pawn concessions by Black on the
kingside, ... h7-h6 can be tricky thing if Black later castles short, as he did in this game.
10 ... h6 11 Kb1
The usual cautionary move, which covers the a2-pawn. More
straightforward is 11 Nd4 Nxd4 12 Bxd4 Rg8 13 c4 and White’s last move
has a dual purpose:
1. It gives White’s bishop the c3-square, in case Black later tosses in ... c7-
c5.
2. It increases White’s control over the d5-square.
a) Here, 13 ... c5!? is very committal. After 14 Bc3 0-0-0 15 Qe3 Qa4?! (I
think this is a waste of time and Black would be better off playing 16 Kb1 as
the c4-pawn is protected tactically) 16 ... d5!? 17 cxd5 Bxd5 18 Qxe7??
(White believes his opponent miscalculated, when in reality, the opposite just
happened and instead, 18 b3 Qc6 19 Bd3! is promising for White) 18 ...
Qxa2+ 19 Kc1 Qa1+ 20 Kd2.
 Exercise (combination alert): Black to play and force the win:

Answer: Queen sacrifice/attraction. 20 ... Qxd1+!, Li Chao-Bu Xiangzhi, Jiangsu 2009. Black
mates after 21 Kxd1 Bf3+ 22 Kc1 Rd1.
b) Here, my preference is for the immediate 13 ... 0-0-0 when the a7-pawn
really isn’t hanging since 14 Qc3 (14 Bxa7 is met with 14 ... Qa4! 15 Bd4
Qxa2 with a good position for Black) 14 ... g6 (White stands a bit better, but
nothing Black can’t handle) 15 g3 Qa4 (15 ... d5 16 c5 looks better for
White) 16 Kb1 c5 17 Bf6 Bxf6 18 Qxf6 h5 19 Rg1 Rge8 20 b3 Qb4 21 Kb2
Qb6 22 Bd3 d5 23 cxd5 Rxd5 24 Be4 Rdd8 25 Rge1 Rxd1 26 Rxd1 c4 has
Black no longer looking worse, L.Dominguez Perez-Wang Yue, Linares
2009.
11 ... Bf6
The alternative is 11 ... a6 12 Nd4 Nxd4 13 Bxd4. Here I would risk 13 ...
0-0! and we once again reach S.Zhigalko-S.Mamedyarov, Kocaeli 2014,
which we covered in the notes to the first game of the chapter (Game 32).
These positions are transpositional nightmares, with one line weaving in and
out of the other, through various move orders. Instead, Black can deviate with
13 ... Rg8 14 c4! 0-0-0 15 Qc3 g6 with a passive but solid position.
12 Nd4 Nxd4 13 Bxd4 Bxd4 14 Qxd4
 Question: Am I correct in stating that one of Black’s biggest hurdles in this
chapter is how to cover both the a7- and g7-pawns from simultaneous
attacks by either a white bishop or queen posted on the d4-square?

Answer: I think your statement is one of Black’s key concerns in this chapter. If you successfully
deal with the issue, in most positions you equalize.
14 ... 0-0
This looks scary, since ... h7-h6 has been tossed in, but Black has a few
factors in his favour:
1. Three sets of minor pieces have been swapped away, which has the
effect of diffusing White’s attack.
2. Black’s position remains inherently solid. Instead, 14 ... Rg8?! just
brings unnecessary suffering for Black, since his king remains in the middle
for now after 15 Be2 b6 16 Ba6 when the tempo loss for White is
inconsequential compared to the annexation of queenside light squares.
15 Be2
An attempt to improve upon 15 Rg1 intending g2-g4-g5, which was first
seen in F.Caruana-K.Landa, Reggio Emilia 2010. White seeks to procure new
armaments in preparation for a kingside assault. Meanwhile Black must
prepare a central counter as a deterrent. An effective solution was found in 15
... Qc6! with the idea of 16 g4. It is human nature to seek to destroy an object
we yearn for but cannot have. After 16 ... Qc5! (principle: meet the
opponent’s wing attack with a central counter), White’s attack is disrupted
and following 17 Qf4 Rae8 18 Bd3, Hervet-Blitsko, correspondence 2012,
Black’s king is secure after 18 ... Re7 19 g5 (19 Qf3 is simply met with 19 ...
Qc6) and Black can play the bypassing 19 ... h5.
15 ... Rae8
Threat: ... Bxa2+ and ... Rxe2. Every black piece is transferred to the
defence of Black’s king.
16 Bf3 b6 17 g4
Here he comes.
17 ... Qb5!
Intending ... Qc4 to remove the queens from the board. The queen’s outer
expression of vacuous stupidity serves her well, since her victims never
suspect her.
18 g5!
White’s best shot at an advantage. Of course Ponomariov isn’t afraid to
sacrifice a pawn to open kingside lines. Instead, 18 Qd2 went nowhere after
18 ... f6 19 Rhe1 Bf7 20 Be4 d5 21 Bd3 Qa5 22 Rxe8 Rxe8 23 b3 Qc5 24 h5
Kf8 25 Kb2 and Black equalized, Ni Hua-A.Giri, Wijk aan Zee 2010.
18 ... Qc4!
Queens come off the board and White’s would-be kingside saturation
bombing campaign is as effective as water soaking solid rock. However,
White’s chances to win remain, since he holds a slight but nagging edge in
the ending.
19 gxh6
White resolves to practice patience, despite the black queen’s incitements
and provocations.
19 ... Qxd4 20 Rxd4!
 This isn’t just outer decoration. The move contains a hidden bite to it and
is an improvement over the pawn recapture. White hopes to make good use
of his rook on the fourth rank, with swings to both the a4-square and kingside
possible. Black must play accurately to hold the ending. After the less
ambitious 20 cxd4 gxh6 21 Bc6 Rd8 22 Rde1 Bd7! 23 Rhg1+ Kh7 24 Be4+
f5 25 Bd3 Rde8 26 Rxe8 Rxe8 27 Kc1 Rg8 28 Rg3 c5 29 dxc5 dxc5 30 Kd2
Rxg3 31 fxg3 Kg6 32 Ke3 Kf6, the players agreed to a draw, L.Czegledi-
A.Husemann, correspondence 2011.
20 ... gxh6 21 Bc6 Rd8
21 ... Re7 and Black may have feared the line 22 Re4 f5 23 Re3 Kf7 24
Rhe1 Kf6 25 Bd7. But there is no problem after 25 ... Bxa2+ 26 Kxa2 Rxd7
27 Re6+ Kg7 28 Rg1+ Kh7 29 Rgg6 Rg7 30 Rxh6+ Kg8 and Black is just
fine.
22 Ra4!
Ponomariov squeezes the maximum out of the position, hoping to induce
pawn weaknesses on the queenside.
22 ... a5 23 b4 axb4 24 cxb4 Bd7!
White’s powerful bishop must be jarred from its c6-perch.
25 Bxd7 Rxd7
The ending is dynamically balanced. White has the potential to create a
future passed a-pawn with a2-a4-a5, while Black can go after White’s slightly
weakened kingside pawns.
26 Re1!
Black still has some problems to solve, like how to activate his stranded
d7-rook.
26 ... Kg7
The king emerges from his convalescent ward and takes a few shaky steps.
GM Alex Baburin suggests passive defence with 26 ... Rc8!?, intending ...
Kf8. Sometimes we temporize with the hope that a long-festering problem
will resolve itself. After 27 Re4 Kf8 28 Kb2 Re7 29 Rf4 Rce8 30 f3
(covering against ... Re4), I don’t envision any real problem from this point
for Black.
27 Kb2 Kg6!

It feels as if the two parties play the same game, but with different rules. In
a difficult position, it’s easy to get overwhelmed and simply allow the wave
of events to pass over us. In this case, Black takes direct action. Principle: In
the ending, the king is a fighting piece. Use it. With this ambitious move,
Wang Hao gives warning that he may later go pawn hunting with his king.
28 Ra3
A cautious response. After 28 Ra7 Kh5 29 a4 Kxh4 30 a5 bxa5 31 bxa5
Kg4 32 a6 Kf3 33 Rb7 Kxf2 34 Ra1 f5 35 a7 Rdd8 36 Kc3 Ra8 37 Rxc7 f4
38 Kd3 f3 39 Rh1 Rg8 40 Rxh6 Raf8 41 Rch7 Kg3 42 Rh2 f2 43 R2h3+ Kg4
44 R7h4+ Kg5 45 Rh5+, Black must take a draw, since 45 ... Kf4? is met
with 46 Ke2 when White is the one with all the winning chances.
28 ... Kh5!
Black’s king is safe, while it worries White about a future ... Kxh4, should
one pair of rooks come off the board.
29 Rg3 f5?!
When we lack defensive coordination, it’s as if our war council is driven
by inner treacheries and petty jealousies, rendering it completely ineffective.
This move gives White access to the e6-square and thus weakens the h6-
pawn. This gets Black into serious difficulties. Safer and stronger is 29 ... c5
30 a3 Rc8 with a dynamically balanced game.
30 Re6!

Target: the h6-pawn. Also, the move discourages ... c5, since the d6- and
b6-pawns could become weak.
30 ... b5 31 Kb3 f4 32 Rgg6 Rh7 33 f3
White may have had better practical chances by creating a passed a-pawn
with 33 c4! bxc4+ 34 Kxc4 f3! 35 a4 Rf4+ 36 Kb3 c5 37 bxc5 dxc5 38 a5
Rb4+ 39 Kc3 Ra4 40 Rg8! Ra1! (40 ... Rxa5?? walks into mate after 41 Re5+
Kxh4 – Black’s beaten down king reminds us of King Lear on a really bad
day – 42 Re1! – “My reign will be glorious under my kindly and wise
tutelage, with feasting and folk dancing each night by with happy subjects”,
boasted the king when ironically, this was the final thought of his life – 42 ...
Kh3 43 Rh1 mate!) 41 a6 Rc1+ 42 Kd2 Ra1, leaving White with all the
winning chances, but I think the game is still drawn at this stage.
33 ... Rf5 34 c4 bxc4+
More accurate is the immediate 34 ... Re5! 35 cxb5 Re3+ 36 Kc4 Ra3 37
b6 cxb6 38 Rxd6 Rxf3 39 a4 Ra3 40 Kb5 f3 41 Rdf6 Kxh4 42 Rf4+ Kh3 43
Kxb6 h5 44 b5 h4 45 a5 Rh5 46 a6 Rb3 47 Kc6 Rc3+ 48 Kb6 Rb3 with a
drawn ending.
35 Kxc4 Re5!
Principle: In rook endings, activity is worth more than material. This is a
pawn sacrifice to reactivate the rook.
36 Ref6 Kxh4 37 Rxf4+ Kh3 38 Rfg4
A better try may have been 38 a4! Re1 39 Re4 Rd1 (threat: ... d5+) 40
Kc3! Rf7 41 f4 d5 42 Re5 Rxf4 43 Rh5+ Rh4 44 Rxh4+ Kxh4 45 Rxh6+
Kg5 46 Rh7 Kg6 47 Rh4 Rb1 48 Kc2 Rg1 49 a5 Kf5 50 Kb2 Rg6 51 b5 Ke5
52 a6 Kd6 53 Rh8 Rg2+ 54 Kb3 Rg3+ 55 Kb4 Rg4+ 56 Ka5 Rg1 57 Rh6+
Ke5 and Black should hold the draw.
38 ... h5
Black generates counterplay by utilizing his trump, his passed h-pawn.
39 Rg3+ Kh2 40 Rg2+ Kh1 41 Rg1+ Kh2 42 R6g2+ Kh3
The king bolts from the white rook’s odious company with a clocked
speed which would bring him deep admiration from even the most athletic of
cheetahs.
43 Rg7
White decides to make it a race between his passed a-pawn and Black’s h-
pawn in a pure rook ending.
43 ... Rxg7
The long ostracized rook is finally able to re-enter society.
44 Rxg7 Re3
Also possible is 44 ... c6 45 a4 Rf5 46 a5 Rf4+ 47 Kb3 Rxf3+ 48 Ka4 Rf1
49 Kb3. The game should end in a draw.
45 a4 Ra3 46 Kb5?!
White’s only winning chances lay in 46 a5! and now 46 ... c5 47 f4! cxb4
48 Kxb4 Ra1 49 Rg5 Rb1+ 50 Kc4 h4 51 f5 lands Black in trouble.
 Exercise (planning): Black to play and force a drawn position:

Answer: Simplification. Black’s last move forces a clearly drawn rook and pawn ending.
46 ... c5!
Now Black finally solves all his problems.
47 bxc5 ½-½
Wang Hao, after lengthy suffering, agrees to the draw in a joyous burst of
bonhomie. Black holds after 47…dxc5 48 a5 (48 Rf7 Kg3 is also drawn) 48
... Rb3+ 49 Kxc5 Rxf3 with a completely drawn ending.
Summary: 20 Rxd4! is an improvement for White, who gets a long ‘+=’ pull in the ending. In a
way, this is a very typical Petroff’s game, where Black gets the draw at the end after prolonged
suffering.

Game 35
F.Caruana-Wang Hao
Kings Tournament, Bucharest 2013

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Be3 0-0

Question: In all the previous games, Black stalled castling as long as possible to keep White in the
dark as to our king’s intent. Why does Black castle so early in this case?

Answer: Castling early is our feisty, nuclear option. In this case Black refuses to play move-order
games. We in essence declare to our opponent: If you castle long, don’t be so sure that it will be
Black’s king who is the one to get mated! You can play this high risk way if you don’t mind chancing a
loss to achieve greater chances to win.
8 Qd2 b6!?

A consensus doesn’t automatically mean that the majority is correct. This

is Wang Hao’s invention, which has taken off in popularity by the top GMs.

Question: What is the idea behind it?

Answer: Apart from blocking the a7-pawn from the Be3’s gaze, Black’s main idea is that his light-
squared bishop is best placed on the a8-h1 diagonal, where it:
1. Simultaneously fights for control over key central d5- and e4-squares.
2. Discourages White from the g2-g4 pawn storm which is one of his main
attacking ideas.
In the next game, we will look at 8 ... Nd7.
9 0-0-0 Bb7 10 Nd4
White’s most logical move. He evades any possible ... Bxf3 ideas, while
simultaneously eyeing Black’s two weakest squares, f5 and c6. Alternatives:
a) 10 Bc4 Nd7 11 Rhe1 Ne5! (this induces White to swap, since both the
f3-knight and c4-bishop are simultaneously attacked) 12 Nxe5 dxe5 13 Qe2
Bd6 14 h4 Qf6 15 Bg5 Qg6 16 g4, M.Vachier Lagrave-E.Safarli, Doha 2014,
looks to me like Black equalizes with 16 ... h6 17 Be3 Rad8 18 g5 h5 with a
sharp and perfectly playable position from Black’s side.
b) 10 h4 Nd7 11 Bd3, L.Dominguez Perez-W.So, Wijk aan Zee 2014,
doesn’t seem advantageous for White after 11 ... Nc5 12 Bf5 Re8.
10 ... Nc6
This was played by Wang Hao in all the three games he reached in this
position.
11 Nf5
White logically avoids swaps and his knight hovers over Black’s king on
the weakened f5- square. 11 Ba6 gets White nowhere after 11 ... Bxa6 12
Nxc6 Qd7 13 Nxe7+ Qxe7 with full equality.
11 ... Bf6
Clearing the way for ... Ne7, challenging White’s knight.
12 Ng3!?

The idea is to transfer the knight to the h5-square. Wang Hao faced 12 h4
in his other two games from this position:
a) 12 ... Re8 13 Bg5 Ne5 14 f4 Ng4 15 Bb5 Re4 16 Ng3, H.Nakamura-
Wang Hao, Stavanger 2013. Houdini rates the game at dead even after 16 ...
Re7 17 Rhe1 h6 18 Bxf6 Nxf6 19 Nf5 Rxe1 20 Rxe1 Be4 21 Bd3 Bxf5 22
Bxf5 h5! 23 g3 Qf8! with ... Re8 to follow.
b) 12 ... Qd7 is an unplayed suggestion of GM Mikhalevski. After 13 Nd4
Ne7! (seizing control over both the d5- and f5-squares, while sidestepping a
potential Bb5 pin) 14 Bg5 Bxd4! 15 cxd4 Rfe8 and I don’t believe White’s
bishop pair means all that much in this symmetrical Exchange French
structure.
c) 12 ... Ne7 13 Ng3 Nd5 14 Bg5 h6 15 Bc4 hxg5 16 Bxd5 Bxd5 17 Qxd5
gxh4 18 Ne4 Re8 19 Qf5 Re5 20 Nxf6+ Qxf6 21 Qxf6 gxf6 22 Rxh4 Rae8
has White’s structural advantage proving to be merely symbolic and Black
held the game without much trouble, S.Karjakin-Wang Hao, Beijing 2013.
12 ... Ne7
Mikhalevski suggests a pair of options:
a) 12 ... g6 cuts off Nh5 and gives Black’s bishop air, at the cost of
allowing White play on the kingside dark squares.
b) 12 ... Qd7 13 Nh5 Be7 14 Be2 Nd8 15 Rhe1 Re8 and now 16 Nxg7
Kxg7 17 Bh6+ Kg8 18 Qf4 Ne6 19 Qg4+ Kh8 20 Bg7+! Kg8 21 Bf6+ Kf8
22 Bg7+ is drawn.
13 Nh5 Nf5 14 Bf4
Caruana wasn’t satisfied with 14 Nxf6+ Qxf6 15 Bg5 Qe6 when once
again, I’m not convinced White’s bishop pair means anything.
14 ... Be5 15 h4!?
A typical advance of the kingside pawns in the Petroff.
15 ... g6?!
I don’t see a good reason to weaken the dark squares like this. Instead, 15
... f6! yields Black full equality.
16 Bd3! gxh5!?
 Wow. This is radical stuff. When we partake in such suicidally risky ideas,
we in a sense declare: “I prefer to harm myself, rather than risk being harmed
by the world.” Black was obviously nervous about the line 16 ... Ng7!? 17
Nxg7! (17 Bg5 Qd7 18 Nf6+ Bxf6 19 Bxf6 Nh5 20 Bg5 Bxg2 21 Rhg1 Bf3
22 Rde1 Rfe8 and I have doubts about White’s full compensation for the
pawn) 17 ... Bxg7 18 Rh2 and if 18 ... h5 19 g4! Bf3 20 gxh5 Bxh5 (taking
the exchange is way too risky) 21 Rg1 and White’s attack looks ominous.
17 Bxf5 Qf6 18 Bxe5 dxe5 19 Bd3
Matters have obviously not gone well for Black, whose structure appears
distressingly wonky on the kingside.
19 ... Kg7!?
I don’t understand this move. 19 ... Rad8 is Black’s most harmonious
move.
20 Qe2
Mikhalevski suggests 20 Rh3 Rg8 21 Rg3+ Kh8 22 Rg5! with advantage
to White after 22 ... Bxg2 23 Rf5 Qe7 24 f3 Rg7 25 Qe2 f6 26 Rxh5 Bh3
with a continuing edge for White, although I would be a bit concerned about
that h5-rook’s position.
20 ... e4!
Wang Hao activates his kingside pawn majority via a little tactical trick.
21 Bc4
 Of course not 21 Bxe4?? which hangs a piece to 21 ... Qf4+.
21 ... Qh6+!?
The idea is to play ... Rad8 and ... f7-f5, without hanging the h5-pawn.
However the move is awkward. Defensive principle: Centralize your forces
when under assault. Therefore, maybe Black should consider 21 ... Qe5.
22 Kb1 Rad8 23 a3 f5
Black’s main trump in this position is that his majority (although
hampered by the doubled h-pawns) can still produce a kingside passed pawn,
while White’s crippled majority on the other side effectively makes him
down a pawn. Of course, his promising attacking chances against Black’s
aired-out king easily make up for this factor.
24 Bb3
Houdini suggests 24 Rxd8 Rxd8 25 Rd1 Rxd1+ 26 Qxd1 Qd6 27 Qxh5 Qg6 28 Qe2! and 28 ...
Qxg2? isn’t possible due to 29 Qd1! with decisive threats down the d-file.
24 ... Kg6?!
Logic is often unable to follow where faith leads. The passage of many
years brings changes for Black’s siege-plagued king, and not for the better.
Attempting to find a coherent plan in an intrinsically chaotic situation is an
attempt to predict that which is inherently unpredictable. The only
explanation of this bizarre-looking move is the approach of time trouble. 24
... Qg6 is more natural and most certainly superior.
 25 Rh3
Caruana again rejects the plan 25 Rxd8 Rxd8 26 Rd1.
25 ... f4!
Wang Hao hopes to insulate the coming blow to his king by cutting off
Rg3+. Black’s king’s position must have been galling to Caruana, who now
goes after it with all he has.
26 Rxd8 Rxd8 27 g4?!
We sense the approach of the rogue elephant, maddened by the midday
sun, about to take it out on the world by trampling a few randomly innocent
villagers. This is a serious slip. Caruana refuses to yield to reason and pushes
on, with no fear of potential backlash. The comp found the shocking 27
Rd3!!, a move which gets Black’s attention:
 a) 27 ... exd3?? 28 Qe6+ Kg7 (the king quivers and jumps like an overcaffeinated frog at even the
mention of his sister) 29 Qe7+ Kh8 30 Qxd8+ Kg7 31 Qe7+ Kh8 32 Qe5+! Qg7 33 Qe8+ Qg8 34 Qxg8
mate!
b) 27 ... Qf8 28 Bf7+!! (overloaded defender/deflection) 28 ... Kg7 (28 ...
Kxf7?? 29 Qxh5+ Ke6 30 Qg4+ Ke7 31 Qg5+ and White wins) 29 Rxd8
Qxd8 30 Qxh5 and the kingside promises to develop into an ugly bar brawl
favouring White (who is close to a winning position).
c) Black’s best chance of survival is to trade rooks with 27 ... Rxd3 28
cxd3 e3 29 fxe3 fxe3 30 Qf1 also favours White. After 30 ... Qg7 31 Qf4 Qe7
32 Bf7+! Kg7 33 Bxh5 e2 34 Qd4+ Kg8 (34 ... Kh6?? 35 Bxe2 and if 35 ...
Qxe2 36 Qf6+ Kh5, 37 Qg5 mate!) 35 Qg4+ and White picks up the e2-
pawn.
27 ... c5?!
Black returns the favour. Houdini found the tricky 27 ... Qg7!! (preventing
Rd3 and opening an air hole for Black’s errant king on the h6-square) 28 Rh1
Kh6! with full equality according to the comp.
28 Rh1
Correct is 28 gxh5+! Kf6 29 Rh1 when White stands better.
28 ... f3
Black again missed 28 ... Qg7!.
29 Qc4
 Instead, 29 gxh5+! Qxh5 30 Qe3 Kf6 31 Qf4+ Qf5 32 Qc7 Rd7 33 Qb8
looks very difficult for Black to hold. If 33 ... e3 34 Qh8+ Ke7 35 Ba4 e2 36
Qg7+ Kd6 37 Bxd7 Qxd7 38 Qxd7+ Kxd7 39 Kc1 and White wins the
ending, since Black’s passer is no problem.
29 ... Bd5 30 Qa4 Bxb3 31 cxb3
Caruana was understandably worried about Black’s pawns in the line 31
Qc6+!? Kg7 32 Qc7+ Bf7 33 Qxd8 hxg4 34 Qc7 Qf6.
31 ... Qf4 32 Re1
32 Qc6+ Rd6 33 Qe8+ Kh6 34 Re1 Rd2! 35 Rxe4 Rxb2+! and the rook,
whose hurt feelings are not easily soothed, displays his vindictive nature
toward all who oppose him. After 36 Kxb2, 36…Qd2+ leads to perpetual
check.
32 ... hxg4 33 Rxe4 Qf5 34 Ka2 Kh5! 35 Re1 h6 36 Qxa7 g3!?

Exercise (critical decision): Should White play 37 fxg3,

or should he go for 37 Qe7, intending Re5 next?

37 fxg3?
Now Black’s f-pawn is deep and White may be lost.
Answer: 37 Qe7! and the game is even after 37 ... gxf2 38 Re5 Qxe5 39 Qxe5+ Kg4 40 Qe4+ Kg3
41 Qg6+ Kh2 42 Qc2 Kg1 43 Qg6+, and the game should end in a draw.
 37 ... f2 38 Rf1 Kg4!
The zombie king arises from the grave and begins to give chase to White’s
rook on wobbly, rotten legs. The critical difference is the fact that Black’s
king helps in the f-pawn’s promotion effort.
39 Qg7+ Kf3 40 g4
White’s g-pawn is actually a hindrance to his perpetual check attempts.
40 ... Qf4 41 Qb7+ Ke2 42 Rb1 Rd3
Also winning is 42 ... f1Q 43 Rxf1 Qxf1 44 Qxb6 Rd1 45 Qb5+ Kd2 46
Qxf1 Rxf1 47 a4 Kc2! (White’s pawns are no match for Black’s rook) 48 a5
Rb1 49 a6 Rxb2+ 50 Ka3 Rxb3+ 51 Ka4 Kxc3 52 a7 Rb1 53 Ka5 Ra1+ 54
Kb6 Kd4 55 g5 hxg5 56 hxg5 Rxa7 57 Kxa7 Ke5 when Black stops White’s
passer, while his own promotes.
43 Qg2

White’s queen eyes the f1-square with a worried brow of visible concern.

Exercise (planning): Come up with a clear winning plan for Black:

Answer: Removal of the guard. White’s remaining lone queen is unable to prevent Black from
promotion.
43 ... Rd1!
The rook shows up with a happy whistle on his lips.
 44 Rxd1 Kxd1 45 Qf1+ Kd2
The f-pawn will promote and White lacks a perpetual check mechanism.
46 g5 hxg5 47 hxg5 Qg3
Also equally effective is 47 ... Qe3.
48 Qb5
Also, 48 g6 Qg1 49 g7 Qxf1 50 g8Q Qd3 51 Qg5+ Kc2 52 Qg2 Kc1! 53
Qg5+ Qd2 wins.
48 ... Qd3 0-1
This had to have been a heart-breaking loss for Caruana. It was one of the
rowdier Roman Emperors, Marcus Aurelius, who wrote in his Meditations:
“Nothing happens to any man which he is not fitted by nature to bear.”
Whenever I lose in particularly nightmarish fashion, the suspicion that
Marcus Aurelius was a damn fool occasionally intrudes into my bitter
thoughts.
Summary: “The new idea of Wang Hao, 8 ... b6, looks interesting and seems to equalize, although
accuracy is required,” writes GM Mikhalevski. Also, castling short early drastically sharpens the
position, since we declare to White our king’s address.

Game 36
V.Gashimov-B.Gelfand
FIDE Grand Prix, Astrakhan 2010

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Be3 0-0
When we castle early like this, we struggle to traverse harsh, unforgiving
terrain, where even the slightest misstep can prove fatal.
8 Qd2 Nd7
 This move and 8 ... Nc6 are Black’s main lines in the early kingside
castling variation.

Question: What is the difference between playing 8…Nd7 and 8…Nc6?

Answer: Not much, since in both lines, Black takes advantage of the fact that White’s bishop went
to the e3-square, rather than the f4-square. So he usually plans ... Ne5, and it does not matter which
route it took. But to my mind, ... Nd7 offers slightly more options, just in case Black needs either ...
Nf6, ... Nf8, or a future ... Nc5.
9 0-0-0 Ne5
Warning White that damage to his structure may follow with ... Nxf3.

Question: But doesn’t this open the g-file for White’s attack?

Answer: It’s a give and take. The good part of the bargain is that we are not obliged to take on f3 if
we don’t want to.
10 h4
Gashimov dares Gelfand to damage his structure but in the process, open
the g-file.
10 ... c6
Opening a path for Black to play …Qa5. Alternatives:
a) 10 ... Bg4 11 Be2 Qc8 (the safest move and the main pick of the top
players) 12 h5 h6 13 Nxe5 Bxe2 14 Qxe2 dxe5 15 Qc4 c6 16 a4 Rd8 17
Rxd8+ Bxd8 18 Rd1 Bb6! (whittling down White’s potential attackers) 19 g4
Qe8 (clearing the way for ... Rd8) 20 Qe4, F.Caruana-V.Gashimov, Khanty-
Mansiysk 2009. At this point, 20 ... f6 intending ... Rd8 looks even.
b) 10 ... Re8 11 h5 Ng4! (taking advantage that White no longer can play
h2-h3 to eject the intruder) 12 Bd3 Nxe3 13 Qxe3 Bf6 14 Qf4 h6,
R.Kasimdzhanov-S.Mamedyarov, Baku (blitz) 2005. I like Black’s chances
on the dark squares and feel they mean more than White’s attacking potential.
c) 10 ... Nxf3!? (this is the kind of move your reckless writer would
probably risk with the assumption that Black won’t get mated in the next
dozen or so moves!) 11 gxf3 Be6 12 Bd3 f5 (limiting the scope of White’s
light-squared bishop, while enabling a second rank defence of the g7-pawn, if
necessary) 13 Bg5 Qd7 14 Rde1 Rae8 15 Rhg1 Bd5 16 f4 Kh8 17 Be2 Bxg5,
T.Radjabov-A.Motylev, Izmir 2004. Houdini assesses at even after 17 ... Bc6
18 Bh5 g6 19 Be2 Bxg5 20 hxg5 a5.
d) 10 ... Be6 ( ... Nc4 is in the air) 11 Kb1 c6 12 Nd4 Bd7 (there is no
reason to give White the bishop pair) 13 Be2 (note that White can’t easily
play f2-f4, since this would leave a gaping g4-hole) 13…Re8 14 h5 Bf8 15 c4
a6! (the coming ... b7-b5 offers Black serious counterplay) 16 Bg5 f6 17 Bf4
b5 18 Nf3 Be6 19 Qc3!? Qc7 (safer than 19 ... Nxc4 20 Nd4 Qd7 21 Nxe6
Qxe6 22 Rhe1 Qf7 23 h6 g6 24 Bxc4 bxc4 25 Bxd6 Rxe1 26 Qxe1 Re8 27
Qh1 Bxd6 28 Rxd6 Qe7 29 Rd1 Qe2 when Black’s activity gives him full
equality) 20 h6 g6 21 Nd2 Rad8 22 Ne4 Qf7, E.Van Leeuwen-H.Rada,
correspondence 2012. I don’t believe Black stands worse.
11 c4
Clamping down on any future ... d6-d5 ideas Black may have harboured.
11 ... Be6 12 Ng5!?
 Such sacrifices represent the sigel of House Gashimov. He goes for the
bishop pair, daring Black to chop the c4-pawn.
12 ... Nxc4!?
This looks pretty risky but the comp tells me it’s fully playable for Black.
Black has tried the more cautious 12 ... Bf5:
a) 13 f3 f6 14 Nh3 Be6 15 b3 a6! (remember this ... a6! and ... b5 idea) 16
Nf4 Bf7 17 Qf2 b5 18 c5 Qc7 19 Nd3 a5 20 cxd6 Bxd6 21 Bb6 Qb8 22 Bc5
Rd8 23 Nxe5 Bxe5 24 Rxd8+ Qxd8 25 f4 Bd6 26 Bd3, S.Karjakin-
B.Gelfand, Monte Carlo 2011. Houdini slightly prefers Black after 26 ...
Qf8!.
b) 13 Kb1 Re8 14 f3 h6 15 Be2!? (the ancient Aztecs believed that a periodic human blood sacrifice
must be made by the high priests for the sun to rise each day), S.Karjakin-V.Kramnik, Moscow 2010.
At this point, I would take the piece and risk 15 ... hxg5 (in the wild, the conscienceless brutality is
nature’s necessary adjunct for survival) 16 hxg5 Qc8! (wisely staying off the d-file) 17 g4 Bg6 18 Rdg1
Qe6 19 f4 Be4 20 Rh3 Nxc4 21 Bxc4 Qxc4 22 Qh2 f5 23 g6 Qxc2+ 24 Qxc2 Bxc2+ 25 Kxc2 fxg4 26
Rxg4 Bf6 and Black stands no worse in the ending)
13 Qd3
Threatening a dirty h7-cheapo, as well as Nxe6, followed by Qxc4.
Otherwise, 13 Nxe6 Nxd2 14 Nxd8 Nxf1 15 Nxc6 bxc6 16 Rhxf1 d5 is an
approximately even ending.
13 ... Bxg5
The bishop snaps out of his spectatorial slumber grudgingly decides to
participate, since he didn’t have much of a choice.
14 hxg5 g6 15 Bf4 Qb6!

Initiating a double attack on the b2- and f2-pawns, and this forces the
queens off the board.
16 Qd4
White clearly didn’t want to swap queens, but he felt he had little choice.
After 16 b3 Qxf2 17 Qf3 Qxf3 18 gxf3 Nb6 19 Rxd6 Nd5 20 Bd2 Rfd8, I
would rather have Black, with his extra pawn, rather than White, with his
bishop pair and dark square control.
16 ... Qxd4 17 Rxd4 d5 18 Bxc4 dxc4 19 Be5 f5!
Black’s king needs help! So Gelfand returns the pawn to secure equality.
Instead, 19 ... Rad8? walks into the swindle 20 Rxh7! f6 21 Re7! fxe5 22
Rxd8 Rxd8 23 Rxe6 Rd5 24 Rxg6+ Kf7 25 Rf6+ with some winning chances
for White.
20 gxf6 Rae8 21 Rh6 c3
 Exercise (combination alert): Black’s king, much like the chronically
misbehaving third grader, lives in fear of the teacher’s rebuke. In this
approximately even position, Gashimov, who may have feared the threat
of a loss from his inferior structure, found a forced drawing combination.
How did White set his attack on Black’s king and force the draw?

Answer: Rook sacrifice/clearance/perpetual check.

22 Rxh7!
A gift may signify both a beginning and an end. In the case of this game, it
is an example of the latter.
22 ... cxb2+ 23 Kxb2 Kxh7 24 Rh4+ Kg8 25 f7+!
This clearance shot secures the draw.
25 ... Kxf7 26 Rh7+
The rook, like Mary’s little lamb (whose fleece was white as snow),
follows Black’s king everywhere he goes.
26 ... Kg8 27 Rh8+
The game ends in perpetual check.
27 ... Kf7
The king’s bloody face is a mass of contusions, yet he remains alive.
28 Rh7+ Kg8 29 Rh8+ ½-½
 Much currency and many goods change hands, with neither side getting
the better of the negotiation.
Summary: I think Black should equalize in the early kingside castling line, with the manoeuvre ...
Nd7 and ... Ne5!, which follows the principle: Counter a wing attack with a central disturbance.

Game 37
P.Negi-Li Chao
Casino Graz Open, Austria 2014

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Be3 Nc6

Question: If Black plans ... Ne5, then what is the

difference between this game and the previous one?

Answer: In this game, Black plans the ... Ne5 manoeuvre, but without castling early. So he isn’t
quite locked into kingside castling just yet, although he often does so later, only if and when White’s
potential attackers feel out of synch.
8 Qd2 Ne5
This is Black’s idea: ... Ne5, without kingside castling. 8 ... Be6 9 0-0-0
Qd7 gets us into positions we examined earlier in the chapter.
9 0-0-0
Just like last game, White invites the structural damage versus open g-file
bargain. More sedate options:
a) With 9 Be2, White prudently takes the sting out of the coming ... Nxf3,
but the move is too passive to yield an advantage. After 9 ... 0-0 (Black got a
normal position but with a slight concession from White, who backed down
with the passive Be2) 10 0-0-0 Ng4! (Black takes advantage of White’s Be2
by attacking the e3-bishop:
a1) 11 Bd3 was L.Nisipeanu-A.Karpov, Vitoria Gasteiz 2007. This bishop
moved twice and so White lost a tempo. After 11 ... Nxe3 12 Qxe3 Bf6,
Black stands solidly, owns the bishop pair and also dark square influence.
a2) Black stands no worse after 11 Bd4 c5 12 h3 cxd4 13 hxg4 dxc3 14
Qxc3 Bxg4 15 Bd3 h5! 16 Qc4 Re8 17 Bh7+ Kf8! 18 Qf4, A.Naiditsch-
V.Kramnik, Dortmund 2005. Black is up a pawn and I don’t believe in
White’s compensatory attacking chances after 18 ... Qa5 19 Kb1 Rad8 20 c3
Bf6 and I don’t see a viable way for White to press his attack. If 21 Rxd6??
Bxf3 22 Rxd8 Qxd8! 23 gxf3 g6 24 Qh6+ Bg7 25 Qf4 Qb6, Black is
winning, since he menaces ... Bxc3, or ... Re2, and also White’s bishop is
stuck in limbo on h7.
b) 9 Nd4 c5! 10 Nb3 0-0 11 0-0-0 Be6 12 f4 Nc4 13 Bxc4 Bxc4,
D.Anagnostopoulos-J.Howell, Hastings 1993. Black already stood better,
with bishop pair and White’s b3-knight out of play.
9 ... Nxf3!?

“The principled follow-up to 8 ... Ne5,” writes GM Mikhalevski.

 Question: In the previous game, Black refrained
from ... Nxf3. Here he plays it. What is the difference?

Answer: The difference is that in this game, Black has yet to castle kingside. So he hopes to inflict
damage to White’s structure, but without paying the penalty of a White kingside attack. The difficulty
with this plan is that Black’s king is hardly safe in the middle, and it won’t be at all easy to castle long.
So he may end up castling short anyway, as he did later in this game. 9 ... 0-0 gets us to the position we
looked in the previous game.
10 gxf3 Be6
This is a continuation of his idea to keep White guessing about the black
king’s residence. Instead, 10 ... 0-0 is answered with 11 h4 Be6 12 Bd3 f5 13
Rdg1 Bf6 14 h5 Qd7 15 Bd4, S.Sevian-D.Ippolito, Los Angeles 2012, when
chances look balanced after 15 ... Rae8.
11 Rg1
The text move is more ambitious than:
a) 11 Kb1 Bf6 12 Bd4 Bxd4 13 Qxd4 Qf6 with equal play.
b) 11 Qd4 Bf6 12 Qb4, L.Dominguez Perez-Wang Yue, Khanty-Mansiysk
2010. Here I would just castle and offer the b7-pawn, with an unclear
position.
11 ... g6
Li Chao takes his non-castling policy to deep extremes. However, 11 ...
Bf6 doesn’t score well after 12 Bd4 Kf8!?.

Question: Why would Black agree to this contortion rather than just castle?

Answer: Oddly enough, Black’s king is safer with his rook on the h8-square, rather than if he
castled. After 13 Be3, A.Motylev-V.Bologan, Moscow 2005, I would actually feel rather comfortable
here with Black after 13 ... g6 since I reach similar positions from the ... Nd7 Smyslov line of the Caro-
Kann.
12 Qd4
A rare continuation. After 12 Bd4 Rg8 (Mikhalevski assesses this position
as equal) 13 c4 Qd7 14 b3, Wang Hao-Li Chao, China 2010, Black can play
14 ... 0-0-0! when I like his chances, since damage has been inflicted to
White’s kingside structure, while the black king is placed on the opposite side
of the board.
12 ... Bf6
A new move, consistent with his non-castling plan. Instead, 12 ... 0-0 is
given an exclam by Mikhalevski, who writes “it’s not easy to find a clear way
for White to create an attack”.
 13 Qb4 0-0
Black finally castles, while offering the b7-pawn to open queenside lines.
14 Qxb7!?
The queen deeply believes in her “Me-first,” Ayn Rand world view, where
unapologetic self-interest is virtuous, while its opposite, the altruistic mind, is
the societal aberration. Some pawns are not so free. White felt he had to take
this pawn, otherwise he had nothing.
14 ... Bxa2!?

One senses that there isn’t a wealth of corroborative evidence backing up

the bishop’s claims of innocence. The straight and narrow honest path grows
wider and crooked! Shades of Bobby Fischer’s ... Bxh2 in his first World
Championship match versus Spassky. This is unbelievably risky, since the
bishop – who believes in his dagger, more than his prayers – checks in but
has no way to check out. I think Black gets full compensation for the pawn
after 14 ... a5! 15 Bf4 Rb8 16 Qe4 Qd7 17 Bc4 Bf5 18 Qd5 Be6 19 Qe4 and
Black can either take the repetition draw, or go for it with 19 ... a4!?.
15 b3
“When the good Lord closes one door, he always opens another,” mutters
the now trapped a2-bishop, who doesn’t really believe in his own sermons.
The principled move, which threatens to trap the wayward a2-bishop, who
now kneels prayerfully, asking God for that which can never be granted: a
second chance.
 15 ... Rb8
Other tries:
a) 15 ... Bxc3?? is met with 16 Qa6 Bxb3 17 cxb3 when Black lacks
compensation for the piece.
b) 15 ... a5 16 Kb2 a4 17 Bc4 axb3 18 cxb3 c6 (intending ... Qa5) 19 Bb6
Qe7 20 Qxe7 Bxe7 21 Ra1 d5 22 Bxd5! cxd5 23 Rxa2 Rxa2+ 24 Kxa2 Bf6
25 Kb2 Rc8 26 Rc1 Be5 27 h3 and Black is fighting for a draw in the ending
a pawn down.
16 Qd5!
Retaining control over the b3-pawn. Now 16 Qa6? was probably the move
Li Chao calculated when he entered the line and after 16 ... Bxb3 17 cxb3
Rxb3, Black’s attacking chances easily compensate for his missing piece.
16 ... Bxc3

17 Bc4?!
Stronger is 17 Rg4! intending Ra4, and White’s advantage remains after
17 ... Qf6 18 Bd4 Bxd4 19 Qxd4 Qxd4 20 Rdxd4 Bxb3 21 cxb3 Rxb3 22
Rgf4. I think Black is the one fighting for a draw in this ending, since
White’s piece is worth more than Black’s unimpressive pawns.
17 ... Qf6
Frustratingly, White has no easy path to capture the smirking a2-bishop.
18 Rd4!?
 White believes the dark squares are a currency more valuable than gold or
silver. White’s last move was flashy, but stronger was the simple 18 Rd3!
Rbe8 19 Rxc3! Qxc3 20 Qd4 Qxd4 21 Bxd4 a5 22 Kb2 Bxb3 23 cxb3 c6. I
like White’s chances with his two bishops versus Black’s rook and two
pawns.
18 ... Ba1?!
Correct was 18 ... Bxd4! 19 Bxd4 Qf4+ 20 Be3 (20 Kb2?? Bxb3! 21 Bxb3
c6! 22 Qc4 (22 Qe4?? loses to 22 ... Qxe4 23 fxe4 c5 24 Bf6 c4) 22 ... c5 23
Rg4 Qxf3 24 Be3 d5 25 Qf4 Qe2! 26 Bxc5 Rxb3+! (now Black’s initiative
threatens to engulf White in a giant tide) 27 Kxb3 Rc8! 28 Qd4 Qb5+! 29
Bb4 Rc4 and Black wins) 20 ... Qf6 with a repetition draw.
19 h4?!
White does better with 19 Rgg4! (intending 20 Rgf4) 19 ... c5 20 Qxd6
Bxd4 21 Rxd4! Qxd6 22 Rxd6 Rb4! (threat: ... Rxc4, which extricates his
bishop) 23 Ba6 Rb6 24 Bxc5 Rxd6 25 Bxd6 Re8 (threat: ... Re6) 26 Bf4
Re1+ 27 Kb2 Bb1 28 Be3 and retains a large advantage. Black’s pieces are
paralyzed, while White will simply chop the a7-pawn, followed by pushing
his b-pawn down the board.
19 ... Rbe8
Black can start cashing in some chips with 19 ... Bxd4 20 Bxd4 Qf4+ 21
Be3 (instead, after 21 Kb2? Bxb3! 22 Bxb3 c6 23 Qc4 c5 24 Rg4 Qxf3 25
Be3 d5 26 Qf4 Qxf4 27 Bxf4 c4 28 Bxb8 Rxb8 29 Rd4 Rb5 30 Kc3 cxb3 31
cxb3 a5, Black’s two extra pawns will be decisive) 21 ... Qxh4 22 Rd1 c6 23
Qxd6 Rbd8 24 Qxc6 Rxd1+ 25 Kxd1 a5! with a position of dynamic
equality. Following 26 Qa4 Qh1+ 27 Ke2 Qb1, White can take a draw if he
wants with 28 Bxf7+ with perpetual check, no matter which way Black
captures.
20 Rdg4
The recurring idea 20 Rgg4! intending 21 Rgf4! was still in White’s
favour.
20 ... Re5
 Exercise (planning): Come up with a plan for White to win the trapped
a2-bishop, which has been left there to decant for quite some time now.

Answer: Trade off Black’s guardian dark-squared bishop, after which White’s king gets access to
the b2-square.
21 Bd4!
After a trade of the dark-squared bishops White’s king gains access to b2.
21 ... Bxd4 22 Qxd4
White simply threatens Kb2 and Ra1, if necessary, winning Black’s
bishop. Li Chao tries a desperate central distraction.
22 ... d5! 23 Bd3 c5?
This is a little too desperate. Black was better off preparing the move with
23 ... Rc8! (intending ... c5) 24 Qxa7! (24 h5 c5 25 hxg6! hxg6 (25 ... cxd4?
26 gxh7+ Kf8 27 Rg8+ Ke7 28 Rxc8 Rh5 29 Re1+ Kd7 30 Ree8 and Black
will lose both rook and bishop in the coming few moves) 26 Qb2 Re6 27
Qxa2 c4 28 Rxg6+ fxg6 29 Rxg6+ Qxg6 30 Bxg6 Rxg6 31 Kd2 a6 and the
game should probably end in a draw) 24 ... Bxb3 25 cxb3 Qxf3 26 Qd4 Rce8
27 Rd1 and White has good chances of consolidating.
24 Qxc5 Qe6
This was the idea behind Black’s previous move, setting up 25 ... Bxb3
followed by 26 ... Rc8. However, the threat is easily parried.
25 Kb2
“Your life is in my hands now, not God’s, as you so naively believe,” the
king informs the doomed a2-bishop.
25 ... Bxb3 26 cxb3 Rc8 27 Qd4 Re1 28 Rxe1?!
Slightly inaccurate, since this allows Black’s queen into White’s camp.
White should play 28 h5! Rxg1 29 Rxg1 which is completely hopeless for
Black.
28 ... Qxe1 29 Bc2!
White’s king is safe and he should win from this point without too many
problems.
29 ... Qe2 30 Qd3 Qxf2 31 h5 Rc6 32 h6!?

In time pressure it is human instinct to be attracted to the lowest common

denominator. Why? Because subtlety in planning requires time, which we
lack. White plays for a clever idea based on Black’s now weak back rank.
The simpler 32 Rg5 d4 33 Rd5 is also winning for White.
32 ... Qh2 33 Rc4!
The weak back rank theme comes into play.
33 ... Rxc4 34 bxc4 Qb8+
Now all that White must worry about is perpetual check.
35 Bb3 dxc4 36 Qxc4 Qe5+ 37 Kc2
 White has to be careful, since there are very few pawns left on the board.
37 ... Qf5+ 38 Qe4
Best is 38 Kd2! Qg5+ 39 Kd3 Qf5+ 40 Ke2 Qe5+ 41 Kf1 (White’s
cowardly king has a gift for disappearing in times of crisis as Black has run
dry on checks) 41 ... Qf5 42 Kf2 Kf8 43 f4 Ke8 44 Qd4 Ke7 45 Ba4 Qa5 46
Qd7+ Kf6 47 Kf3 Qb6 48 Qd5 (threat: Qe5 mate!) 48 ... Qb8 49 Bb3 and it is
game over, since White threatens mate with Qg5 or Qf7.
38 ... Qc5+ 39 Kb2 Qf2+ 40 Ka3
We sense White’s king is getting ready for a journey, since he got his
cholera inoculation, packed his bags and is currently googling foreign
monetary exchange rates.
40 ... Qc5+ 41 Ka4
White’s king treats Black’s queen the way most teenagers treat their
mothers: with eye rolling, grudging obedience.
41 ... Kf8 42 Qa8+?
This decentralizing move doesn’t make sense. White should convert
without too many problems after 42 f4.
42 ... Ke7 43 Qb7+!?
43 Qe4+ was a better chance for a win.
43 ... Kf6 44 Qxf7+?! ½-½

The energy in White’s winning attempts wanes, turning a three

dimensional entity into a caricatured 2D Black and White version of reality.
He agreed to a draw in a position where he still has winning chances.

Question: Why did White agree to the draw here when he still has chances?

Answer: The decision to agree to the draw was a self-inflicted wound, reminiscent of a sloppily
swung hammer over our soon-to-be bruised thumb. Chess is unmerciful in that if we misread reality,
then the interpretation’s accompanying pain is certain to follow. I remember a painful game where my
clock ran low in a terribly complex position and I agreed to a draw against a 2350 player. When I got
home and put it on the computer, it had me up +5.50, the equivalent of a full rook and more! I simply
didn’t understand the position’s most basic elements.
Summary: Mikhalevski called Li Chao’s early ... Ne5 and ... Nxf3 without kingside castling “very
interesting and requires many more practical tests” and my take is that the resulting positions give rich
play though Black has to be defensively alert dealing with White’s potential build-up on the half-open
g-file.

Game 38
V.Anand-V.Kramnik
Corus A, Wijk aan Zee 2008

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Bf4

Question: In the first part of the chapter, we became acquainted

with White’s problems with developing his bishop to the e3-square.
Here, by placing the bishop on the f4-square instead, he covers
against the ... Nd7 and ... Ne5 ideas. What is the downside of 7 Bf4?
 Answer: In the case of 7 Bf4, I see two issues:
1. White may lose a tempo later if Black plays ... Nd7-c5-e6.
2. If White plans a full-fledged kingside pawn storm, then his bishop may
get in the way as it will be hampering the movement of the f-pawn.
7 ... 0-0
No move order stalling this time. We just castle straight away.

Question: In the 7 Be3 half of the chapter, you emphasized the importance
of Black stalling castling as long as possible. Yet in the 7 Bf4 part,
Black castles kingside and early in every game. What has changed?

Answer: My feeling is that Black’s king is safer on the kingside in the 7 Bf4 version, since he has
access to more central counters. For example:
1. Black will soon play ... Nd7 and ... Nc5, which gives White central
headaches, because ... Ne6 may gain a tempo.
2. White has to watch out for ... Ne4 tricks, when his queen often lacks a
safe, stable square. Now this is a problem since White doesn’t want to part
with his light-squared bishop as if he plays Bd3 to cover the e4-square, a …
Nd7-c5 sally might threaten the prelate’s existence. We won’t be covering 7
... Nc6 or 7 ... Nd7.
8 Qd2 Nd7
A note to remember. We always develop …Nd7 in the 7 Bf4 line. The
point is the knight heads for the c5-square, where it creates trouble for White,
via …Ne4 or …Ne6.
9 0-0-0 Nc5 10 Be3
White’s main move. The bishop, having done its job of preventing ... Ne5
from Black, moves it back to e3 to cover the f2-square. The obvious problem:
White took two moves to play Be3, handing our side a precious tempo, just to
keep us away from playing ... Ne5. Let’s look at some alternatives:
a) Preparing g2-g4 with 10 h3 looks too slow to produce any kind of an
edge. After 10 ... Re8 11 Bd3 (Black isn’t intimidated by 11 g4 Ne4 12 Qe2
Bf6 13 Be3 Be6 when White’s queen sits awkwardly on the open e-file) 11 ...
c6 12 Rhe1 Qa5 13 Kb1 Be6 14 a3 Nxd3 15 Qxd3 Qd5! (forcing queens off
the board is nearly always beneficial for Black in this chapter, and in
Petroff’s in general) 16 Qxd5 Bxd5, Black already stood slightly better due to
his bishop pair, Z.Byambaa-D.Frolyanov, Khanty-Mansiysk (blitz) 2013.
 b) The natural square to develop the light-squared bishop is 10 Bc4 since
the d3-square is covered by Black’s knight on c5. However, this runs into
10…Be6 (swaps favour Black) 11 Bxe6 Nxe6 12 Be3 Qd7 (a key manoeuvre
in the 7 Bf4 line: Black activates his queen to d7 to grab light squares via c6,
b5 or a4, after Black’s knight plays to to e6-square) 13 h4 Qb5! (this prevents
the Qd3/Ng5 battery) 14 Ng5, V.Popov-I.Khenkin, Sochi 2005. At this point
White isn’t threatening anything and I would just play 14 ... a5 with
approximate equality.
c) Sakaev calls 10 Kb1 “indifferent”, meaning that White may or may not
need the move in this version, and may well later just be down a crucial
tempo. Black should just play 10 ... Re8 which can get us back to most of the
positions we covered, except we in effect may be up a move.
We will look at 10 h4 and 10 Nd4 in Game 40 and Game 41 respectively.
10 ... Re8
Less accurate is 10 ... Be6 (we will look at 10….c6 in the next game). 11
Kb1 Re8 12 Nd4 Ne4 13 Qc1 Bd7 White clearly profited from Black’s early
... Be6. 14 f3 Nf6 I prefer White, since he can now commence with his
kingside pawn storm, J.Mrkvicka-T.Matheis, correspondence 2009.

11 Bc4
White’s dilemma: deciding where to place his light-squared bishop:
1. Be2 is passive.
 2. The d3-square is covered by Black’s knight.
3. Bc4 gets challenged immediately with ... Be6.
4. Bxc5, followed by Bd3, costs White the bishop pair and some degree of
control over the dark squares.
11 ... Be6
Of course. We peace-loving Petroffers are usually happy to swap.
12 Bxe6 Nxe6 13 h4
Anand begins White’s standard plan, preparing Ng5 and Qd3.
13 ... Qd7
This manoeuvre is important to remember. Black proceeds to engage his
queen on the queenside light squares.
14 Qd5
White’s other tries:
a) 14 Qd3 Qc6 15 Qf5 Qc4 transposes to the game’s position.
b) 14 Kb1 Qb5 15 g4 Qc4 16 g5 a5 17 h5 Qg4 18 Nd4 Nxd4 19 cxd4 d5
20 g6!? (Black has the resources to defend this attack, mainly since White is
nearly out of minor pieces to do the attacking) 20 ... fxg6 21 Rdg1 Qe4 (21 ...
Qf5 is more accurate) 22 hxg6 hxg6 23 f3 Qe6 24 Bf4 Bd6 25 Be5 Bxe5 26
dxe5 Qxe5 27 Rxg6 Re7 28 Rg5 Qe3 29 Qxd5+ Qe6 30 Rxg7+?? (White
overpresses, refusing the probable drawn result after 30 Qxe6+) 30 ... Kxg7
31 Qg5+ Kf7 32 Rh7+ Ke8 33 Rh8+ Kd7 34 Rxa8 Qe1+ 35 Qc1 Qf2! 36
Rxa5 c5 37 Ra3 c4 38 Ra5 Kc6 39 Ra4 b5 40 Qh6+ Kd7 41 Qh3+ Kc7 42
Rxc4+ bxc4 43 a3 c3 44 Qf5.
 Exercise (planning): Black is up a rook for pawns, with an
exposed king. His win is simple if you find the correct plan.

Answer: Simplification. Black can forces queens off the board, with a trivial win after 44 ... Qe1+!
(with the idea of 45…Qe6+) F.Piccoli-T.Tiits, correspondence 2008.
14 ... Qc6
Of course Black would love a queen swap, which would eradicate White’s
attacking dreams.
15 Qf5
After 15 c4 Bf6 16 g4 g6 17 Qxc6 bxc6 18 c3 c5 19 Kc2 Ng7 20 h5 Re4
21 hxg6 fxg6 22 g5 Be7 23 Rh4, the players agreed to a draw in this even
position, V.Anand-V.Kramnik, Nice 2008.
15 ... Qc4
Black can also go passive with 15 ... Nf8 16 h5 Qd7 17 Qd5 Qe6 18 c4 (18
Qxb7 Qxa2 is not bad for Black) 18 ... Qxd5 19 cxd5 Nd7 and White was
unable to exploit his space advantage against Black’s wall of solidity,
P.Svidler-B.Gelfand, Mexico City 2007.
16 Kb1
Instead, 16 Ng5 Bxg5 17 hxg5 Nf8 18 Kb1 is met with 18…Re5
(Principle: Meet a wing attack with a central counter) when Black prepares
to swing the rook into his own potential queenside attack. This induces White
to swap queens, which leaves him with nothing in the ensuing ending. After
19 Qd3 (19 Qf3 is met with 19 ... Qe4) 19 ... Qxd3 20 cxd3, an even ending is
reached, S.Karjakin-V.Kramnik, Wijk aan Zee 2007.
16 ... g6

Played in all 10 games in my database. Black finally agrees to weaken his

kingside pawn front, since now Ng5 is a real concern. Moreover, 16 ... a5
leaves White better after 17 Ng5 Bxg5 18 hxg5 Nf8 19 Rd4 Qe6 20 Qf3
Rab8 21 c4.
17 Qh3
Intending h4-h5 next.
17 ... h5
White induces more weakening. Still I believe in Black’s defensive ability
to absorb the coming attack.
18 Nd2
White’s main move. Chances are equal after 18 Nd4 Nxd4 19 Rxd4 Qe6
20 Qf3 Bf6 21 Rb4 b6 22 Qc6 Rac8 23 Rd1 a5 24 Rc4 Rcd8 25 a4 Re7 26
Rd5 Qd7 27 g3 Qxc6 28 Rxc6 Ra8 29 Ka2 Kf8 30 Ka3 Ke8 31 b4 axb4+ 32
cxb4 Kd7 33 b5 Re4 34 c4, M.Bartel-D.Fridman, German League 2006. I
don’t see how White makes progress after 34 ... Bc3 35 Kb3 Ba5 Now if 36
c5 Rb4+ 37 Ka3 Rc4 38 Kb3 Rb4+ which is a repetition draw.
 18 ... Qe2
An untried idea is 18 ... Qa4 and it looks fully playable. After 19 Rdg1
Ng7 20 g4 Bf8 21 gxh5 Nxh5 22 Rg4 Qb5 23 Rhg1 Bg7 (threat: ... Bxc3) 24
Rg5 Re5 25 Rxe5 dxe5, I don’t believe Black stands worse. His healthy
kingside pawn majority easily makes up for White’s attacking chances, which
grow dimmer with each swap.
19 Rde1 Qg4 20 Qh2
White’s only prayer for an advantage is to keep queens on the board.
20 ... d5 21 f3 Qa4 22 g4 Bd6
Forced. Not 22 ... hxg4?? 23 h5 g5 24 h6 Bf6 25 fxg4 with a winning
attack for White.
23 Qf2
Anand offers his g-pawn to open kingside lines. Instead, after 23 Qg1 Ng7
24 Nb3 Qd7 25 Rd1, S.Karjakin-D.Jakovenko, Foros 2008, Black’s defensive
resources appear adequate following 25 ... c6 26 Nc5 Qe7 27 Bd4 Be5 28
Qg2 Bxd4 29 cxd4 b6 30 Nd3 Qf6 31 Ne5 c5 32 f4 cxd4 33 Qxd5 Rad8 with
even chances.
23 ... hxg4!?

A subversive influence begins to wear away at Kramnik’s mind, until he

indulges in a very risky pawn grab. What incredible self-confidence to play
such a move on one of the world’s premier attackers. Much safer is 23 ... Ng7
24 c4 dxc4 25 Bd4 Qc6 26 Bc3 Bc5 27 Qg3 Bd6 28 Qf2 Bc5 29 Qg3 Bd6
and the players agreed to a repetition draw, P.Svidler-V.Kramnik, Mexico
2007. “Perhaps both players had found an improvement on White’s play,”
writes IM David Vigorito.
24 fxg4
White must make it a real sacrifice:
a) The zwischenzug 24 b3? is met with the counter-zwischenzug 24 ... g3!
25 Qg2 Qc6 when White’s attack is no more.
b) The same applies for 24 h5? g3! 25 Qg1 Ng7! 26 hxg6 fxg6 27 c4 Nf5
when White is the one in trouble.
24 ... Qxg4 25 Reg1 Qh5

Question: Doesn’t this violate the principle: The queen is the worst
possible blockader, since any piece which attacks it, forces it to step aside?

Answer: Correct, yet this may be an exception to the rule, since White has no clear pathway of
attacking the h5-square.
26 Nf3 Re7 27 Bg5 Ree8 28 Be3 Re7 29 Bg5
So a repetition draw?
29 ... Rd7!?
Nyet! Kramnik avoids the repetition, although Anand may well have
declined the repetition himself one move later.
 30 Nd4
This has turned into a battle between White’s pure brawn versus Black’s
ability to elude.
30 ... Nxd4?!
This move swaps away a potential attacker, at the cost of giving White’s
queen a jump into the centre. I think Black has it better with 30 ... Bc5! 31
Re1 Bxd4 32 cxd4 Ng7 33 Re5 Re8 34 Rhe1 Rxe5 35 Rxe5 Qd1+ 36 Bc1
Qg4. I like Black’s chances here and don’t believe in White’s full
compensation for the pawn.
31 Qxd4 Bf8 32 Qe3!

The queen strikes her Charlie’s Angels guns-held-high-in-miniskirt-and-

boots pose. Dual purpose:
1. White prevents ... Re8.
2. White prepares Bf4, followed by Rg5, ejecting Black’s queen from the
h5-blockade.
32 ... c6 33 Qh3
Attacking the d7-rook.
33 ... Rd6 34 Bf4
Making way for the rook.
34 ... Re6 35 Rg5 Qh8 36 h5
This is optically terrifying for Black, yet Houdini assesses at 0.00.
 36 ... Rae8 37 Bd2 Bc5 38 Rg3 Re2 39 Kc1 Qg7 40 a3
Vigorito describes this move as “future-luft for his king on a2.”
40 ... Bd6
Also, 40 ... Qh8 41 hxg6 Qxh3 42 gxf7+ Kxf7 43 Rhxh3 R8e7 is probably
a drawn ending.
41 Rgg1 Bc5 42 Rg3 Bd6?!
Stronger is 42 ... R2e5! 43 hxg6 fxg6 which maintains the balance.
43 Rg4!
Nahi! (Your writer is a linguist, who mastered the word “No!” in both
Russian and Hindi. For the Russian part of the equation, please scroll up to
move 29 to see the answer). Now it is Anand who dodges the repetition.
43 ... R8e6?
Better is 43 ... R2e6 covering the g6-pawn.
44 hxg6! Rxg6
After 44 ... fxg6?? 45 Rh4!, White has a winning attack since ... Be5 isn’t
possible.
45 Rxg6 fxg6 46 Be3!
Intending Bd4. Black’s king is in serious danger now.
46 ... Qe5!
Black’s queen and king’s sense of regret equals Adam and Eve’s
misgivings when they received their eviction notice from the Garden of Eden.
47 Qh7+ Kf8 48 Bd2!
Note that 48 Rf1+ Ke8 49 Qxg6+ Kd8 allows Black’s king to escape to the
safety of the queenside.
48 ... Qf6!
 Kramnik keeps finding only moves. White’s rook must be kept away from
the f-file. After 48 ... Be7?? 49 Qxg6 Bf6 50 Kb1!, White has a winning
attack, since his bishop is covered by Qh6+.
49 Qxb7
Black is probably losing since his king remains insecure and his queenside
pawns have been destabilized.
49 ... Rh2!
Playing for back rank tricks, while ejecting White’s rook from the h-file.
50 Re1?!
White’s rook dismisses the h2-pest with an impatient gesture of the hand.
White can do better or worse:
a) Stronger was 50 Qa8+ Kg7 51 Qxa7+ Kg8 52 Re1 which leaves White
in a winning position.
b) Most definitely not 50 Rxh2?? Qf1+ 51 Be1 Qxe1 mate!
50 ... Qf2!
 Exercise (critical decision): Kramnik somehow contrived an initiative,
while on the defensive! White’s choices are: a) Go on defence with
51 Rd1. Now factor in Black’s potential response 51 ... Bf4. b) Go ballistic
and sacrifice a piece with 51 Kb1 Qxd2 52 Rf1+. White retains winning
chances in only one of the lines. Which one would you play?

51 Kb1?
White’s now dysfunctional attack is similar to the military regiment which
loses contact with its commanding officer. This aggressive attempt allows
Black to escape. It took the Allies two full years to prepare for a full-capacity
assault on the beaches of Normandy. Moral: big projects require time.
Answer: White should have slowly prepped with 51 Rd1! and Black’s pawns are ripe for the
picking. The bungling 51…Bf4?? runs into 52 Qb4+ followed by Qxf4.
51 ... Qxd2 52 Rf1+ Kg8!
The position resembles one of those perilous Coast Guard helicopter sea
rescue attempts, where they battle swelling waves and a capricious wind:
a) 52 ... Ke8?? walks into a mate after 53 Qc8+ Ke7 54 Qf8+ and the
queen warns her brother, “If you choose to jump off a cliff, you will die; if
you wilfully disobey my edicts – either in spirit or the letter of the law – you
will just as surely die. Both methods are equally certain methods of suicide”.
Following 54 ... Kd7 55 Rf7+ Ke6, Black’s king is unable to outrun the page
on which his inevitable end is written and 56 Qe8+ Be7 57 Qxe7 wins.
b) 52 ... Bf4?! declines to address an important footnote. White regains the
lost piece with 53 Qb8+! (defending the b2-pawn) Kg7 54 Rxf4! Qe1+ 55
Ka2 Qe7 56 Ra4 Rh8 57 Qf4 (threat: Qd4+ and Rxa7) 57 ... c5 58 Qg3 and
White still retains some wining chances since his king is safe, while Black
must deal with both king safety and also potential pawn targets.
53 Qf7+ Kh8 54 Qxg6
White covers the c2-pawn, while threatening Qe8+ and Rf7+.
54 ... Qg2! 55 Qe8+ Qg8! 56 Qxc6
The queen finally averts her dark obsession with Black’s king and takes
time out to pick off a pawn, just in case the attack falls flat.
56 ... Bf8
Kramnik feeds every defender he has to the aid of his king.
57 Qa8 Bc5 58 Qxg8+
The ending is a probable draw. 58 Qc6 Bf8 simply repeats the position.
58 ... Kxg8

Exercise (combination alert): White has only two pawns for the piece in the ending, yet he stands
no worse, since he found a way to win a third pawn. How?

Answer: Pin.
 59 Rf5! Rd2 60 c4!
White wins a third pawn for the piece and the game is dead even.
60 ... Kg7 61 b4 Be7 ½-½
After 62 Rxd5 Rxd5 63 cxd5 Kf6 64 c4 Ke5 65 Kc2 Kd4 66 Kb3 Bf8 and
it is White who must play accurately to draw after 67 Ka4! (not 67 a4?? Be7
68 a5 Bf8 69 b5 Bd6 70 b6 a6 when it’s zugzwang and Black wins) 67 ...
Kxc4 68 Ka5 Bd6 69 Ka6 Kxd5 70 Kxa7 Kc4 71 Kb6 Bxb4 72 axb4 Kxb4.
Summary: Some key points to remember in the 7 Bf4 line:
1. We refuse to play move order games and always castle short early on.
2. We always develop …Nd7, where it heads for the c5-square, when White must then watch out
for …Ne4 and …Ne6 tricks.

Game 39
C.Deepan-H.Koneru
Indian Championship, Visakhapatnam 2006

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Bf4 0-0 8 Qd2 Nd7 9 0-0-0 Nc5
10 Be3 c6

Question: What are some of Black’s ideas behind her last move?

Answer: Three potential ideas behind the move:

1. Black opens the possibility of ... Qa5.
2. Black cuts off a future Qd5, which is normally the route the white queen
takes to transfer over to the kingside.
3. Black sets up a potential queenside future pawn storm with ... a7-a6 and
... b7-b5, or ... a7-a5, ... b7-b5-b4, depending on how White arranges his
queenside structure.

Question: Why doesn’t Black simply develop with

10 ... Be6 which adds pressure to the a2-pawn?

Answer: The move is natural but probably inaccurate, since after 11 Kb1, White will play Nd4 next,
going after Black’s valuable bishop.
11 h4
Surprisingly White’s natural last move doesn’t work well against Black’s
early ... c6 move order, since Black’s queenside play self-generates with
alarming speed. “Here this advance is an inaccuracy that passes the initiative
into Black’s hands,” writes GM Maxim Notkin.

Question: Do you believe the move is an error?

Answer: I think the move gives away any hope of an advantage, but I don’t believe White stands
worse just yet. Alternatives:
a) 11 Nd4 Ne4 12 Qe1 Re8 13 f3 Nf6 14 Qf2 (White’s queen is obviously
uncomfortable on the e-file) 14 ... Nd5 15 Bd2, M.Sebag-H.Koneru,
Ekaterinburg 2006. Black attained equality at a minimum and perhaps even
stands better after 15 ... c5 16 Ne2 Be6 17 Kb1 b5 since her queenside play
looks faster than White’s potential kingside assault.
b) 11 Kb1 Re8 12 Qc1 (threatening the cheapo Bxc5, at the cost of placing
the queen on a safe yet lame square) 12 ... Qc7 13 Nd4 a5 14 c4 a4 15 a3 Bd7
16 Be2 Rab8 17 g4 Bf8 18 Bf3 Ne4 with dynamic equality, M.Vachier
Lagrave-D.Jakovenko, Evry 2008.
c) 11 Bxc5 (a steep concession as White hands over the bishop pair to rid
himself of the pesky knight and to remove Black’s control over the d3-
square) 11 ... dxc5 12 Qf4 Qa5 13 Bd3 (intending Qe4, with a double attack
on the h7-pawn and e7-pawn) 13 ... Be6 14 a3 Bf6 with even chances,
P.Leko-V.Kramnik, Wijk aan Zee 2008.
d) 11 c4 Ne4 12 Qe1 Re8 13 Bd3 Bf6 14 Qb4 c5 15 Qb3 b6 16 c3 Bb7,
Black’s knight is an unrepentant pest and Black already looks a shade better,
R.Ponomariov-L.Nisipeanu, Bucharest 2013.
 11 ... Qa5
Black’s queen, in conjunction with a coming …Be6, produces swift
queenside play.
12 Kb1 Be6 13 a3
With hindsight of the game, maybe White should bail out into an even
ending with 13 c4 (based on the knowledge, that in the wild, an injured
animal will go to great lengths to hide its own frailty) 13 ... Qxd2 14 Nxd2,
A.Warakomska-M.Goslawski, Warsaw 2014.
13 ... Ne4

The powerfully centralized knight’s energy projects outward, like spokes

from a wheel’s hub. Instead, 13 ... Na4 is met with 14 Ng5 Bf5 15 Bd3 Bf6
16 c4! Qxd2 17 Rxd2 Bxd3 18 cxd3 Rfd8 with an even ending.
14 Qe1
The queen has no comfortable posts. If 14 Qd3 Bf5 15 Nd4 Bg6, now
White must move his queen again, since 16 h5?? is met with 16 ... Nxf2 17
Qd2 Nxh1 18 hxg6 hxg6 19 Bf4 Bh4. Black’s knight gets out and he is an
exchange up.
14 ... Bf6
Dangerous sacrifices begin to loom over the c3-pawn.
15 Bd4
The defenders are awakened, forewarned of the coming ambush attempt
by the barking of the f6-dog. Other choices:
a) White can centralise the knight with 15 Nd4 Rfe8 16 Bd3 Nc5 17 Nxe6
Nxd3 18 Rxd3 Rxe6 and if any one stands better, it is Black.
b) If White proceeds blissfully unaware with something like 15 h5??,
Black meets it decisively with 15 ... Bxc3! and if 16 bxc3 Qxa3 is curtains for
White, whose king is engulfed by too many threats.
15 ... Rfe8
Protecting the knight tactically.
16 Bd3
Not 16 Bxf6? Nxf6 17 Qd2 (17 c4? Qc5 White will lose material) 17 ...
Ne4 18 Qe1 Bg4 19 Be2 d5 which is obviously unpleasant for White, who is
under pressure down the e-file, while the c3- and f2-pawns remain sensitive
points.
16 ... Bxd4
After 16 ... d5, 17 Ng5! is okay for White and following 17 ... Bf5 18
Nxe4 Bxd4 19 cxd4 Qxe1 20 Rhxe1 Bxe4 21 Bxe4 Rxe4 22 Rxe4 dxe4, a
balanced ending is reached.
17 Bxe4?

Exercise (combination alert): A massive fire can merely be the growing continuation of a once tiny
spark. The arsonist is held responsible for the totality of the damage – not just for lighting the initial
match. White’s last move was a blunder and we sense a newly formed fortuitous geometric conjunction
for Black. Find one deadly idea and you inundate White’s position with overwhelming threats. How?

Answer: Double attack.

17 ... Bc5!
The a3- and f2-pawns warehouse all of Black’s potential, as the bishop
issues a pair of not-so-veiled threats. Black intends ... Qb6! next,
simultaneously attacking both points.
18 Qf1
White can go wrong easily:
a) 18 b4?? loses instantly to 18 ... Qxa3.
b) 18 Qd2 Bxa3 is also hopeless for White.
c) 18 Ng5 h6 19 Bh7+ Kh8 20 Qd2 Qb6 (with dual threats on the a3- and
f2-pawns) 21 Nxe6 Bxa3 22 b3 Rxe6 and Black remains up a clean pawn.
This may be White’s best line out of his many unpleasant choices.
18 ... Bxa3!
The long silence is shattered by a torrent of explosive threats from Black.
The piece sacrifice is far stronger than 18 ... Qb6 19 Ka1 Bxf2 20 Rxd6 when
White is still in the game.
19 bxa3 Qxa3 20 c4
Flashy but futile is 20 Bxh7+ Kxh7 21 Ng5+ Kg8 22 Nxe6 Rxe6 and
White is down a pawn with an endangered king.
20 ... Qb4+
Black had a superior continuation in 20 ... f5! 21 Bd3 b5! 22 Nd2 bxc4 23
Bxc4 Rab8+ 24 Nb3 (24 Bb3? Bxb3 25 cxb3 Rxb3+ 26 Nxb3 Qxb3+ 27 Ka1
Rb8 forces mate) 24 ... d5 25 Bd3 a5 and there is no good defence to the
coming ... a5-a4.
21 Kc1 Bxc4 22 Bd3 b5!
White is unable to trade bishops, since that would open the b-file. Black
has a winning attack.
23 Re1 a5?!
Black cements her advantage with 23 ... Qa3+! 24 Kd1 Qa1+ 25 Kd2
Qa5+ 26 Kd1 d5.
24 Rxe8+?
After this move, handing Black control over the e-file, the defence lapses
into a comatose version, a still living body, drained of spirit and sentience.
Matters are not so clear after 24 Re3! when White has chances to survive,
since his king may now be able to escape to the kingside.
24 ... Rxe8
Now Black is winning again, since White’s king is unable to pass the e-
file.
25 Qd1 Qa3+ 26 Kb1
The king’s humiliation is evident from the fact that his eyes refuse to meet
his sister’s cold stare. It doesn’t help that 26 Kd2 loses quickly to 26 ... Bxd3
27 cxd3 Qb2+ 28 Qc2 Re2+ and there goes the queen.
26 ... Ba2+
“You are charged with perverting God’s law,” declares the bishop to the
b1-prisoner.
27 Ka1 Bd5+ 28 Kb1 b4
Black’s simple plan is to play ... Qa2+, followed by ... a5-a4-a3, when the
a-pawn soon threatens to promote.
29 Nd2 Qa2+
The queen’s up and down moods rule the lives of all those around her.
30 Kc1 a4!
Threat: ... a3 and ... Qb2 mate!
31 Re1 Qa3+ 32 Kb1 Ba2+ 33 Ka1 Be6+ 34 Kb1 Qa2+ 35 Kc1 a3
Threatening mate.
36 Ne4 Qb2+ 37 Kd2
 Exercise (planning): Black to play and force the win:

Answer: Pawn promotion/discovered attack. The key is to seize control over the c3-square, which
in turn wins heavy material.
37 ... d5! 0-1
After 38 Ng5 Qc3+ (the queen asks her younger brother, “And just where
do you think you are going, poppet?” as she soon plans to eradicate him) 39
Ke3 (or 39 Kc1 a2 and the a-pawn’s succession to queen is no longer in
dispute) 39 ... Bf5+, the discovered attack picks off the e1-rook and 40 Kf3
Rxe1 ends the game.
Summary: 10 ... c6 is a dangerous idea. If White isn’t careful, our side can generate quick attacks
based on ... Qa5, ... Be6, ... Ne4 and ... Bf6, when all our pieces take aim at White’s king.

Game 40
S.Rublevsky-A.Shirov
Russian League 2006

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3

So Rublevsky plays the system against its champion. Heritage-wise, my
database has Adolf Anderssen playing it against the Johann Loewenthal in a
London 1851 match of two 19th century chess giants.
5 ... Nxc3 6 dxc3 Be7 7 Bf4 0-0 8 Qd2 Nd7 9 0-0-0 Nc5 10 h4
A logical move. White proceeds with his kingside agenda.
 10 ... Bf6
Black activates his dark-squared bishop, while clearing the e-file. White
can also centralize his pieces with 10 ... Re8 11 Bc4 Be6 12 Bxe6 Nxe6 13
Be3 Qd7 (13 ... c6 14 c4 a6 15 Ng5 b5 16 Qd3 Nf8 17 Qf5?! (17 Rhe1 is
even) 17 ... f6 18 Ne6 Qc8 19 Nd4 bxc4 and I don’t believe in White’s
compensation for the pawn, M.Carlsen-A.Karpov, Moscow (blitz) 2007) 14
Qd5 Qc6 15 Qf5 Qc4 16 Kb1 g6 17 Qh3 h5. Does this position look familiar?
Well, it should, since we covered this in V.Anand-V.Kramnik, Wijk aan Zee
2008 (Game 38).
11 Ng5
We have already seen this idea, over and over in the chapter.
11 ... h6 12 f3!?

Question: Why did White play his last move?

Answer: Dual purpose:

1. White cuts off a future ... Ne4.
2. White prepares a kingside pawn storm with g2-g4 next. Of course, there
is a slight catch: his g5-knight remains hanging! 12 Bd3 is another
incarnation of the piece sacrifice, which really can’t be taken:
a) 12 ... hxg5?? is suicidal after 13 hxg5 Nxd3+ 14 Qxd3 Re8 15 Qh7+
Kf8 16 Qh4! and Black is crushed, since 16 ... Be5 is met with the simple 17
Bxe5 when Black is unable to recapture.
b) 12 ... Nxd3+ 13 Qxd3 g6 14 Nf3 h5 15 Rde1 Bf5 16 Qd2 Re8 17 Bg5
Bxg5 18 Nxg5 Qf6 19 f3 Re5 and Black stood no worse, V.Topalov-
E.Bacrot, Wijk aan Zee 2006.
c) Black can also stall with 12 ... Re8!? 13 Kb1 Bd7 14 f3 Bc6 15 Bc4 d5 16 Qf2 Na4!? (or 16…b6)
with a right mess.

Chances are even, meaning all three results are equally possible!
12 ... Rb8!?
When we openly advertise our intention to attack without any attempt at
concealment of our plan, it is similar to walking up to a pair of police officers
and telling them: “I intend to hold up the bank across the street, at exactly
3:50 p.m. today.” The police may not have the power to arrest you just yet,
but they will certainly do it at 3:51 p.m., the moment you enter the bank and
pull the gun.

Question: What is Black’s idea?

Answer: Shirov isn’t interested in grabbing a piece (which happens to lose!) and plays for his own
attack with ... b7-b5, maybe ... a7-a5 and then ... b5-b4.

Question: Can Black get away with acceptance of the piece?

 Answer: It’s suicidal once again:
a) 12 ... hxg5?? 13 hxg5 Be7 14 g4! and there is no good response to the
coming Qh2. If 14 ... Bxg5 15 Bxg5 f6 16 Be3 Ne6 17 f4, White has a
winning attack.
b) Logical and strong is 12 ... Ne6, eliminating the loitering g5-attacker.
After 13 Nxe6 Bxe6 14 Kb1 Re8 15 Qf2 Qd7, H.Lang-A.Matoussi, Dresden
2008, Black threatens ... Qa4. Now if 16 a3 Qa4 17 Bc1 b5 18 g4 Rab8
(White is unable to play 19 g5?? since Black has 19 ... Bxc3! 20 bxc3 b4!
with a winning attack) 21 Ka1 bxa3 22 gxh6 a2 (threatening …Rb1 mate) 23
Bb2 Rxb2!, Black wins.
c) The untested 12 ... Re8 also looks fine. If 13 Bd3 Qd7! (This time Black
can maybe take the piece. 13 ... hxg5!? 14 hxg5 Be5 15 Bh7+ Kf8 16 Bxe5
Rxe5 17 b4 Ne6 18 f4 Nxf4 19 Qxf4 Qxg5 20 Qxg5 Rxg5 21 Be4 Ke7 22
Rh8 a5 23 Bf3 Re5 and Houdini assesses at even. Black is up a pawn but his
queenside pieces are tangled up and unable to develop properly) 14 Ne4
Nxd3+ 15 cxd3 Be7 16 g4 d5 17 Ng3 d4 18 c4 b5 19 g5 Bb7 and it’s
anybody’s game.
13 g4 b5
Here he comes.
14 Bd3 Re8
a) Trading minors with 14 ... Nxd3+?! helps White by creating weakness
in Black’s camp. After 15 Qxd3 g6 16 Ne4 Bg7 17 Be3 Bb7 18 Kb1 a6 19 h5
g5 20 Ba7 Ra8 21 Bd4, I slightly prefer White’s chances.
b) Black can go all guns blazing with 14 ... b4! 15 cxb4 a5! 16 Ne4! (a
wise precaution, eliminating Black’s c5-attacker) 16 ... Nxe4 17 Bxe4 Rxb4
18 c3 Rxe4! 19 fxe4 Bxg4 20 Rdg1 h5. It feels like Black gets full
compensation for the exchange, with one pawn, an entrenched g4-bishop and
potential e4- and h4-pawn targets.
15 Bh7+ Kf8 16 Nh3!?
Threat: g4-g5. Right piece, but perhaps the wrong square! If a single
problem is ignored and left to fester, the danger is that the one becomes the
many, and it erupts into an infestation. Safer is 16 Ne4!, since it eliminates
Black’s powerful knight. Rublevsky clearly underestimated Black’s coming
attack. In this variation, the attack gets respect and a key attacker is
eliminated. After 16 ... Nxe4! 17 Bxe4 (now g4-g5 is a serious threat) 17 ...
Bxg4 18 Bc6 (18 fxg4 Rxe4 19 g5 Be5 20 Bxe5 Rxe5 21 gxh6 g6 22 Rdf1
Qe7 looks okay for Black) 18 ... Bd7 19 Bxd6+! cxd6 20 Qxd6+ Re7 21
Bxd7, White is up a pawn, although the opposite coloured bishops and solid
position keep White’s edge to a bare minimum.
16 ... b4!
Shirov is more interested in launching his own attack, rather than playing
it safe with a line like 16 ... Be5 17 Bxe5 Rxe5 18 g5 Bxh3 19 f4 Re7 20
Rxh3 h5 which is about even.
17 cxb4 a5!

Shirov calculated the consequences of this tricky move with great

accuracy.
18 g5
The only move. Naturally not 18 bxc5?? Bxb2+ 19 Kb1 Bc3+ winning on
the spot.
18 ... Bxh3 19 gxf6
Let’s look at the other bishop capture with 19 Rxh3 Rxb4!:
a) 20 c3 Rxf4! 21 Qxf4 hxg5 22 Qg4 gxh4.
 Question: How should we assess this position?

Answer: GM Olivier Renet assesses as unclear, while I feel like Black stands significantly better for
the following reasons:
1. Black has two pawns for the exchange, so in a sense he is maybe up half
a pawn in the material count.
2. The black king is absolutely safe.
3. Black controls the e-file.
4. Black dominates the dark squares.
5. Black’s h-pawn is passed and must be watched carefully should more
pieces come off the board.
In a single position, two players may have deep faith in their own
assessments, which is part of the beauty of the game. However, faith doesn’t
mean equivalency of the judgment’s net worth. In fact, sometimes two
players use the same logical reasoning, yet reach opposite conclusions!
b) 20 gxf6 Qxf6 21 Bxd6+ cxd6 22 Qxd6+ Qxd6 23 Rxd6 Re1+ 24 Kd2
Re5 25 Kc1! (25 b3?! Ke7! (White’s rook is nearly out of squares) 26 c3
Kxd6 27 cxb4 axb4 and Black stands better since his b-pawn for now holds
back White’s two queenside pawns. Also, White must nurse a pair of
kingside isolanis) 25 ... Re1+ 26 Kd2 Re5 is drawn by repetition.
 19 ... Qxf6
Shirov continues to ignore the fact he has two loose minor pieces, with an
almost insulting negligence.
20 c3
20 Rxh3 transposes to a position we already looked at in the notes (to
White’s 19th move) after 20 ... Rxb4 21 Bxd6+ cxd6 22 Qxd6+ Qxd6 23
Rxd6 Re1+ 24 Kd2 Re5.
20 ... axb4 21 cxb4
Two black pieces hang simultaneously yet Shirov has everything under
control. Inferior is 21 Rxh3? which is met with 21 ... bxc3 22 bxc3 Rb3! and
White has no choice but to bail out into an inferior ending after 23 Bxd6+
cxd6 24 Qxd6+ Qxd6 25 Rxd6 Rxc3+ 26 Kd2 Ra3 27 Bb1 Rb8 28 Kc1,
leaving White struggling.
21 ... Na6!?
True objectivity is a virtual impossibility for the human mind, which
insists on superimposing its stylistic biases upon the data. In this case, strong
resolution may be pushed too far, to the very brink of obstinacy.

Question: Isn’t this knight completely out of play on the a6-square?

Answer: An idea’s worth is not negated by its initial visual perception. I admit it’s very risky
though. Shirov banks on the fact that he will either sacrifice the knight later on b4, or pry the b-file
open with a timely ... c7-c5, which I might add, is not so easy to achieve. A lot safer is 21 ... Ne6,
keeping the knight centralized.
22 a3
Rublevsky hopes to keep Black’s knight offside and out of play. After 22
Rxh3 Nxb4, Black has a dangerous attack:
a) 23 Kb1?? loses to 23 ... Qe6! with a double attack on the a2-pawn and
the h3-rook. Now if 24 b3 (The simplifying trick 24 Bxd6+?? fails miserably
due to a weak back rank after 24 ... cxd6 25 Qxd6+? Qxd6 26 Rxd6 Re1+ 27
Rd1 Rxd1 mate!) 24 ... Nxa2!, White is busted. Note that Black must avoid
the greedy 24 ... Qxh3?? 25 Bxd6+ and it is White who wins. The bishop’s
rudeness is a shock to Black’s king, as profound as that time when he used a
dessert fork for his entree at the State dinner.
b) 23 Qd4! Nxa2+ 24 Kb1 (the only move) 24 ... Nc3+ 25 Kc2 Nxd1 26
Qxf6 gxf6 27 Kxd1 Rxb2 28 Bxh6+ Ke7 29 Bc2 Rb5 with a very difficult
position to assess. My feeling is that it is probably dynamically balanced.
Let’s go deeper: 30 Be3 Ra8 31 h5 Ra1+ 32 Kd2 Rd5+ 33 Kc3 Ra3+ 34 Kc4
Rda5 35 Bf2 Ra2 36 Kd3 R2a3+ 37 Kd2 Rd5+ 38 Kc1 Raa5! 39 h6 Rh5 40
Rxh5 Rxh5 41 Be3 f5 42 Kd2 Kf6 43 Bf4 Kg6 44 Bb3 Rh1 and I don’t think
either side can make progress.
22 ... Bf5
Principle: Eliminate an opponent’s bishop pair if possible.
23 Bxf5 Qxf5
 Exercise (combination alert): The position was approximately
balanced, until White’s last move. How should Black respond?

24 Rhg1??
White takes for granted a premise – that his king is safe – which has yet to
meet the burden of proof. White was still fine if he played the correct 24
Rhe1.
24 ... Nxb4!!
Black’s knight traffics in a forbidden commodity. Shirov displays a
masterly grasp of the position’s essential elements.
Answer: Piece sacrifice/Annihilation of the defensive barrier. If a powerful person gives you
something (Black’s once out-of-play knight), then it is very easy for him to take it away if he changes
his mind.
25 axb4 Ra8
Threatening mate on the move.
26 b3
Clearing the d2-square for the king with 26 Qd4 is met with 26 ... Re2! 27
Qxg7+ (27 Bxd6+ Ke8! forces mate) 27 ... Ke7 and White is unable to deal
with the dual a1- and c2-mate threats.
 Exercise (combination alert): The white king’s chances are
equivalent to the sparrow’s, who begs the hawk to go vegan
and spare his life. Black to play and force the win:

Answer:
Step 1: Lure the king to the b2-square.
26 ... Ra1+ 27 Kb2 Qf6+! 0-1
Step 2: Check on f6, which wins heavy material in all lines. White has a
choice of losing moves:
a) 28 Qd4 Qxd4+ 29 Rxd4 Rxg1 gives White an easy endgame win.
b) 28 Qc3 Ra2+ and White’s king is deflected from his queen.
c) 28 Kc2 Ra2+ 29 Kd3 Rxd2+ and the rook cites, “If a wealthy person
dies after her legal heirs die, then the wealth reverts to the government”.
Summary: When White sets up kingside assault plans with h2-h4, Ng5 and f2-f3, we can just
ignore the piece offer, as Shirov did in this game.

Game 41
V.Malakhov-D.Frolyanov
Russia rapid championship, Olginka 2011

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Bf4 0-0 8 Qd2 Nd7 9 0-0-0 Nc5
10 Nd4
 White rejects the h2-h4/Ng5 plan as too artificial, and instead, centralizes
his knight to make way for a kingside pawn storm with f2-f3, g2-g4 and h2-
h4 later on.

Question: This sounds logical. Is there anything

within this plan which should worry White?

Answer: One problem for White is that this plan allows Black a quick ... Ne6!, simplifying his
position with swaps, since his d4-knight and f4-bishop are under the e6-knight’s gaze.
10 ... c6
Black opens the path for ... Qa5. Other options:
a) 10 ... Re8 11 f3 Ne6! (this forcing swap is Black’s main theme when
White posts a knight on the d4-square) 12 Be3 Bg5!? (Kramnik provokes f2-
f4, which in turn weakens the e4-square) 13 f4 (13 Bxg5 Qxg5 14 Qxg5
Nxg5 15 Nb5 Ne6 16 Bc4 Bd7 17 Bxe6 Bxb5 18 Bd5 Ba6 19 c4 c6 20 Be4
Bxc4 21 Rxd6 Rad8 is equal) 13 ... Nxd4 14 fxg5 Nc6 (White’s bishop pair is
easily compensated by the fact that Black’s knight has access to the e5-
square) 15 h4 Bg4 16 Re1 Qd7 (16 ... Ne5 is more ambitious) 17 Bb5
(Adams decides to return the bishop pair to secure equality) 17 ... Re5 18
Bxc6 Qxc6 19 Bd4 Rxe1+ 20 Rxe1 Re8 21 b3 b6 22 c4 Rxe1+ 23 Qxe1 Kf8
24 g3 Qf3 25 Kb2 Bf5 and a draw is agreed, M.Adams-V.Kramnik, Wijk aan
Zee 2008.
b) My preference is for 10 ... Ne6! when Black wastes no time in forcing a
relieving swap. Following 11 Be3 Nxd4 12 cxd4 (12 Bxd4 c5 13 Be3 Be6
and if 14 c4 Qb6, Black stands slightly better, since his queenside play is
more developed than White’s on the other side) 12 ... Bf5 13 f3 Re8 14 g4
Bg6 15 Kb1 Bh4! (blockading White’s intended h4 ideas) 16 Bf2 h6 17 f4
(this move comes at the cost of weakening the e4-square) 17 ... Be4 18 Rg1
Qf6 19 Bxh4 Qxh4 20 Rg3 Re7 21 d5 Rae8 22 a3 a6 23 Qf2 f5 24 h3 fxg4 25
hxg4, R.Hübner-V.Kramnik, Frankfurt 1996, White looks a bit loose and is
the one fighting for the draw after 25 ... Bg6 26 Kc1 Re3 27 Rf3 Qxf2 28
Rxf2 Rg3 29 Rg2 Rxg2 30 Bxg2 Re3.
11 f3 Ne6!

This is Black’s main theme (when White’s knight is played to the d4-
square), which is based on the principle: The side which is cramped benefits
from swaps. After 11 ... Re8 12 g4 Ne6 13 Be3 Qa5 14 Kb1 Nxd4 15 cxd4
(15 Bxd4 Be6 16 c4 Qxd2 17 Rxd2 Bg5 18 Rd1 d5 and if anyone stands
better, it is Black) 15 ... Qxd2 16 Bxd2, the ending is equal, I.Denisov-
D.Frolyanov, Moscow 2012.
12 Nxe6?!
Something of great value in one environment can be useless in another, the way a brilliant scholar
may not be the best comrade if you get into a bar fight and count on his help. White maybe should be
thinking about ways to equalize. He would have been better off playing the unambitious 12 Be3 Nxd4
13 cxd4 Be6 14 Kb1, M.Kleinburg-V. Mikhalevski, Ashdod 2013.
12 ... Bxe6 13 a3!?
This is awfully slow.

Question: Doesn’t 13 Bxd6 just win a pawn?

Answer: It isn’t easy to delay gratification of our desires. In this case Black hopes to sell White a
useless trinket for good, hard cash. In the end, though, I think White had to play this way. Black’s
initiative more than compensates after 13 ... Bxd6 14 Qxd6 Qa5 15 a3 Rad8 16 Qf4 Rxd1+ 17 Kxd1
Qd5+ 18 Bd3 Qa2 19 Qc1 c5. I prefer Black’s initiative to White’s rather worthless extra pawn. I still
prefer this line to the way White played in the game.
13 ... d5 14 Bd3 b5!
Black clamps down on the c4-break, while preparing his own, with ... a7-
a5 and ... b5-b4.
15 g4 a5

I already prefer Black, since lines are guaranteed to open when ... b4
arrives. White, on the other hand, has a long way to go before lines open on
the kingside.
16 Bf5!?
A sign of desperation. White allows damage to his structure to dilute
Black’s attacking chances. Black’s attack looks terribly dangerous if White
proceeds normally with 16 h4 Qb6 17 h5 b4 18 a4 bxc3 19 Qxc3. It feels to
me like White won’t survive Black’s coming attack, due to the open b-file.
16 ... b4 17 a4
17 Bxe6?? fxe6 and White is busted, since he faces the dual threats ...
Rxf4! and ... bxa3.
17 ... Bxf5 18 gxf5
White’s alleged attacking chances from the newly opened g-file is an
abstract entity, while the disfiguring contusions to his kingside structure are
of a real and permanent nature.
18 ... Bf6
Black can also speculate with 18 ... Qd7! 19 Rhg1 Qxf5 20 Bh6 g6! 21
Bxf8 Bxf8. For the exchange, Black has a pawn, the dark squares and the
superior structure, which offer him the superior chances.
19 cxb4 axb4 20 b3 Bc3 21 Qd3 Qf6 22 Rhg1 h6?!
There is no need for this time-wasting defensive addition. Black’s
advantage is clear after 22 ... c5!.

Question: What happens if White takes the bait with 23 Qxd5??

Answer: Let’s turn this into an exercise:

 Exercise (combination alert): Black to play and win material:

Answer: Interference/clearance/double attack/pin. The computer shot 23 ... Bd2+!! wins on the
spot:
a) 24 Kxd2 Rfd8 pins and wins White’s queen;
b) 24 Kb1 Bxf4 wins as well.
c) 24 Bxd2?? Qa1 is mate!
23 Be3
Seizing control over the c5-square.
23 ... Rfe8 24 Bd4!
Life tends to fit into harmony when something we need and something we
also desire, merge. The defence’s once numbed faculties slowly begin to
thaw. Malakhov finds a way to loosen Black’s queenside grip, yet still must
struggle to equalize.
24 ... Bxd4 25 Qxd4 Re5 26 Rg4?!
The market is sluggish on White’s attacking chances. He just doesn’t have
the time to double rooks to hit the g7-pawn. Unsettled positions don’t always
clarify into clearly discernible ones. Sometimes they get even more unsettled.
With his last move, White crosses that almost invisible demarcation point of
no return, and his hopes droop like week-old roses in the vase. White should
chance the line 26 Kb2! Rb8 27 Rde1 Rxf5 28 Qxf6 Rxf6 29 Re7. This looks
like a slightly superior version of the ending White got in the game. After 29
... Rxf3 30 a5 Ra8 31 Ra1 Ra6 32 Rb7 c5 33 Rb5 c4 34 bxc4 dxc4 35 Rxb4
Rf5 36 Rba4, White’s dangerous passed a-pawn puts the fate of the game up
in the air.
 26 ... h5!
26 ... Rxf5?! allows White counterplay after 27 Qxf6 Rxf6 28 Rxb4 Rxf3
29 Rb6 Rf6 30 Rd3 intending Rc3, when White’s a-pawn, in conjunction
with pressure on the c6-pawn may allow him to save the game.
27 Rg3
Also, 27 Rf4 Rae8 28 Kb1 Qh6! (threat: ... c6-c5!, overloading White’s
queen) forces 29 Rh4 Rxf5 30 Qxb4 Rxf3 and leaves Black up a pawn,
although White’s passed a-pawn remains a source of concern.
27 ... Rxf5 28 Qxf6 Rxf6 29 Rd4 Rb8 30 f4
 Black is up a pawn, but the win is by no means easy, since White still
owns that passed a-pawn.
30 ... g6 31 Kb2 Kf8
Black activates his king.
32 c3 c5
Stronger is 32 ... bxc3+! 33 Kxc3 Ke7 34 b4 Kd6 35 Re3 c5! 36 bxc5+
Kxc5 37 Red3 Rd6. White’s starving army exists on half-rations. But for how
long? His a-pawn is no longer a danger for Black and he is unlikely to save
the game.
33 Rxd5 Rxf4 34 Rd2
If White gives up the second rank with 34 Rxc5 Rf2+ 35 Kb1 Rxh2 36
cxb4 Rxb4 37 Rgc3 h4 38 a5 Re2 39 a6 Re7 40 Kc2 Ra7 41 Ra5 Rb6 42 Rc4
Rbxa6 43 Rxa6 Rxa6 44 Rxh4 g5, I think Black’s chances of winning are
higher than White’s chances to draw.
34 ... bxc3+ 35 Rxc3
Black should also win after 35 Kxc3 Rfb4 36 Ra2! h4 37 Rd3 Rxb3+ 38
Kc4 Rxd3 39 Kxd3 Ke7 40 Kc4 Kd6 41 a5 Rb4+! 42 Kc3 Kc7 43 a6 Kb8 44
a7+ Ka8 when the a-pawn’s power to intimidate is over.
35 ... Rxa4
What a relief it must have been to clip this dangerous passed pawn.
36 Rxc5 Rab4 37 Rd3 Kg7 38 Rf3 R8b7
 Covering his second rank.
39 Ka3
White’s king, with head clutched in hands, looks like the ‘before’ scene in
an aspirin commercial. It will be very difficult for White to push his passed b-
pawn. Meanwhile, Black’s 3:1 kingside majority soon rolls forward with
ease.
39 ... f5
The kingside majority is activated and White is engulfed in a surging tide
of passed pawns.
40 Rc6 h4
Cutting off Rg3 ideas.
41 Rd3 Kh6 42 h3 R7b6 43 Rc1 Re4
Notice that White’s passed b-pawn hasn’t moved an inch since it went to
b3.
44 Rc8 Kg5 45 Rc1 Kf6 46 Rf3 Rbe6!

Black sets up ... Re3.

47 Rcc3
The third rank has to be covered as 47 b4 Re3+ 48 Rxe3 Rxe3+ 49 Ka4
Rxh3 50 b5 Rh2 51 Kb3 Re2 52 Rc8 Kg5 53 Rc7 (53 b6 Re6 54 Rb8 h3 is
hopeless for White) 53 ... Re1 54 Kc2 h3 55 Rh7 Kg4 56 b6 Re8 is not even a
close race. Black wins.
 47 ... g5 48 b4
Finally, the pawn inches up, but the counterplay is just too slow, and there
are just some problems in life which we are unable to outrun.
48 ... Ra6+ 49 Kb3 Rb6 50 Rc4 Rxc4 51 Kxc4
Black’s plan is to create two passers and sacrifice his rook if necessary
later on for the b-pawn.
51 ... Ke5 52 Rb3 g4 53 Kd3 gxh3 54 Kc4 Kf4 55 Rxh3 Kg4 56 Rh1 h3
57 Rg1+ Kh4 58 Kc5 Rb8 59 Rg7

When White’s king asks for volunteers to step forward to help stop
Black’s surging h-pawn, there is much shuffling of feet and twiddling of
fingers, yet no takers on his request. Black cuts off Rh7+ ideas and wins a
rook.
59 ... Rh8! 0-1
It’s all over as there is no defence after 60 Rg1 (this rook is no match for the featherweight h-pawn
to his right) 60 ... h2 61 Rh1 Kg3 winning.
Summary: White’s Nd4 plan is effectively neutralized with the freeing swap ... Ne6!, which gets us
equality.

Game 42
D.Baramidze-V.Akobian
World Team Championship, Antalya 2013

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3 6 dxc3 Be7 7 Bf4 0-0 8 Qd2 b6!?
 If you come up with an idea which is in conflict with a widely recognized
doctrine, the odds are against you being right. This move, however, may be
an exception. It’s a new move, but not a new idea. GM Varuzhan Akobian
borrows on Wang Hao’s early ... b6 idea, and inserts it in the 7 Bf4 lines.

Question: What does the bishop do on the b7-square?

Answer: Pretty much the same as in Game 35 (Caruana-Wang Hao):

1. Black’s fianchettoed bishop fights for key central d5- and e4-light
squares.
2. Black’s fianchettoed bishop keeps White guessing about ... Bxf3,
although this scenario is unlikely, since it would greatly enhance White’s
light square control.
3. If White tries the standard plan h2-h4 and Ng5, followed by f2-f3, then
the intended g2-g4 is a tricky thing, since the f3-pawn is greatly weakened.
Also, in this plan, if White’s knight is transferred from g5 to e4, than Black’s
fianchettoed bishop pressurizes that square as well.
9 0-0-0 Bb7 10 h4

Question: Is the h4/Ng5 idea White’s only plan in the position?

Answer: No. White can also try 10 Nd4 intending to exploit the newly weakened f5 square –
Black’s only slight downside to the early ... b6 plan. The d4-knight also watches the c6-square, which
could later get weak if White gets a chance to play Bb5 and Nc6 later on. I think Black’s best bet in this
case is to immediately challenge the d4-knight with 10 ... Nc6 and if 11 Nf5 (11 Nxc6 Bxc6 allows
Black easy equality) 11 ... Bf6 12 f3 Qd7! has White finding difficulty defending the f5-knight. For
example, after 13 Bd3 Ne5, Black picks up the bishop pair:
a) 14 Be4?? loses to 14 ... Nc4 15 Qd3 Qa4! (Black simultaneously
threatens a2 and also ... d5!):
a1) 16 Kb1 d5 17 Bxd5 Bxd5 18 Qxd5 Rad8 19 Qb7 Rxd1+ 20 Rxd1
Nxb2! 21 Kxb2 Qb4+ and Black has a crushing attack.
a2) 16 Bxb7 Qxa2 17 Qxc4 Qxc4 and White doesn’t even have time to
pick up the a8-rook, since his f4-bishop hangs with check.
b) After 14 Bxe5 Bxe5, I think Black stands a shade better.

Question: Why? Can’t White eventually play Qf2 and

then f2-f4, Rg1 and g2-g4, with a strong kingside attack?

Answer: Black’s bishop pair, dark square control and central play down the e-file give him the
superior chances. For example, 15 Qf2 g6 16 f4 Bf6 17 Ne3 Rae8 18 Rhg1 Re7 19 g4 Rfe8 20 Rde1
d5! 21 g5 Bg7 22 f5 d4 23 Ng4 Rxe1+ 24 Rxe1 dxc3 25 Nf6+ Bxf6 26 gxf6 cxb2+ 27 Kb1 Rxe1+ 28
Qxe1 Bc6 29 Qe5 Qd6! 30 Qxd6 cxd6 31 Kxb2 leaves Black with excellent chances to win the pawn
up bishop ending.
10 ... Nd7 11 Nd4
A well-timed move. Now Black doesn’t have ... Nc6, as he did on the
previous move. An ineffective try is 11 Ng5?! Nc5 12 Bd3 (I don’t believe
White gets enough for the pawn after 12 f3?! h6 13 Nh3 Bxh4!) 12 ... h6
when Black’s chances are at least even. For example:
a) 13 f3?! Nxd3+ 14 Qxd3 hxg5 15 hxg5 f5! 16 Rh5 Qd7 17 Rdh1 Kf7! 18
Rh7 Rg8 and White doesn’t have enough for the piece.
b) 13 Nf3 Nxd3+ 14 Qxd3 Re8 15 Rh3 Bc8! and the rook must retreat,
since 16 Rg3 Bxh4 17 Nxh4 Qxh4 18 Be3 Bg4 19 Re1 h5 leaves White down
a pawn and awkwardly placed.
c) 13 b4 (this ugly move may be White’s best in the position) 13 ... Nd7
(there is no reason to allow White 13 ... Ne6 14 Nxe6 fxe6 15 Bxh6 gxh6 16
Qxh6 Rf7 17 Qxe6 Bf6 18 Bc4 Qe7 19 Qxe7 Bxe7 20 Bxf7+ Kxf7 21 f3
although, even here, Black may not be worse after 21 ... a5) 14 Bh7+ Kh8 15
Be4 Bxe4 16 Nxe4 a5 17 b5 f5 18 Ng3 Qe8 and Black stands no worse.
11 ... Nc5
Eyeing the e4-square, as well as ... Ne6 possibilities.
12 Re1
I would try 12 f3.
12 ... Bf6
There is no reason to get greedy and allow an inhuman comp line like 12
... Bxh4!? 13 Bd3 Nxd3+ 14 Qxd3 Re8 15 Rxe8+ Qxe8 16 Qh3 g5 17 Bg3
Qe4 18 f3 Qe3+ 19 Kb1 Bc8 20 Nf5 Bxf5 21 Qxf5 Re8 22 Bxh4 gxh4 23
Qg4+ Kf8 24 Qxh4 Qd2 which is still fine for Black.
13 Bb5
Believe it or not, this invasion isn’t nearly as powerful as it appears. Also,
13 f3 a6 14 Bg5 h6 15 Bxf6 Qxf6 is at least even for Black.
13 ... a6 14 Nc6
The logical follow-up as 14 Bc6 Bxd4 15 Bxb7 Bxf2 16 Qxf2 Nxb7 leaves
White down a pawn.
 Exercise (critical decision): Should Black play 14 ... Qc8, or offer
his queen with 14 ... Qd7? One line is stronger than the other.

14 ... Qd7
Akobian, who appreciates the gravity of his situation, takes appropriately
desperate (yet sound) measures with a spectacular queen sacrifice.
Answer: However, stronger was 14 ... Qc8!!. When we rely on intuition over logic, we sometimes
discover that there can be more power in inner silence than in outer perception. This looks suicidal but
is actually very strong after 15 Bg5 axb5 16 Bxf6 (White menaces the dual threats Qg5 and Ne7+) 16 ...
Qg4! 17 Ne7+ Kh8 18 Qg5 Qxg5+ 19 hxg5 Bxg2! (weaker is 19 ... gxf6 20 g6! fxg6 21 Nxg6+ Kg7 22
Nxf8 Kxf8 23 Rxh7 Rxa2 24 Kb1 Ra4 25 Ree7 Rh4 26 Rhf7+ with a likely draw) 20 Rh2 Be4 21 f3!
(seizing control over the g6-square) 21 ... Bxf3 22 g6 fxg6 23 Nxg6+ Kg8 24 Nxf8 Rxf8 25 Be7 Re8
26 Bh4 Rxe1+ 27 Bxe1 g5 and Black’s two pawns for the exchange offer him all the winning chances.
15 Ne7+
White has no choice but to accept the queen sacrifice.
15 ... Qxe7 16 Rxe7 axb5!
Black only has two pieces for the queen, which is probably enough, since
he threatens both e7-rook and a2-pawn. Black doesn’t get enough for the
queen after 16 ... Bxe7? 17 Bf1.
17 Rhe1!
A wise decision. Hoarding material with 17 Rxc7?! Rxa2 18 b4 Ne4 19
Qe1 Ra1+ 20 Kb2 Rxe1 21 Rxe1 Bxc3+ (White’s king feels a shiver of
unease along his spine) 22 Rxc3 Nxc3 23 Kxc3 Rc8+ 24 Kb2 Bxg2 25 Bxd6
f6 leaves Black up a pawn, with winning chances, since White’s remaining
kingside pawns are weak.
17 ... Bxe7 18 Rxe7 Rxa2 19 Qe3 Ra1+ 20 Kd2 Bxg2?
20 ... Ne6 is approximately even.

Exercise (combination alert): Black’s last move gave White access

to a deeply hidden winning idea. How would you continue?

21 h5?!
Answer: White misses 21 b4!! (chess computers, like clairvoyants, have the power to divine a
position’s hidden truths) 21 ... Ne6 22 Bh6! Kh8 23 Qg3 Bf1 (23 ... gxh6 24 Qxg2 leaves White with a
winning position, since he can destabilize Black with a future f2-f4-f5) 24 Rxe6 gxh6 25 Qf4 Bc4 26
Qd4+ Kg8 27 Qg4+ Kh8 28 Re3 and Black won’t survive, since his king is too exposed.
21 ... Rh1 22 h6 Rh3! 23 f3?!
I still like White’s chances after 23 Qe2! Bf3 24 Qxb5 gxh6.
23 ... Rxf3 24 Qd4?
Chances are even after the correct 24 Qg1 Rxf4 25 hxg7 Kxg7 26 Qxg2+
Kh8.
 Exercise (combination alert): How did Akobian halt the
g7-mate threat and win White’s queen to boot?

Answer: Deflection/knight fork.

24 ... Rd3+!
In times of famine, even normally honest people may steal from their
neighbours. We notice a sharp intake of breath, followed by pursed lips from
White’s queen at the rook’s vulgar proposal.
25 Qxd3
Also, 25 cxd3 Nb3+ 26 Ke3 Nxd4 27 Kxd4 Rc8 is similar to the game’s
continuation.
25 ... Nxd3 26 cxd3 Rc8
 Black is up three pawns and the opposite coloured bishops do not suffice
to save White.
27 Re3
Threat: Rg3.
27 ... Bd5 28 Rg3 g6 29 Bg5 Kf8 30 Re3 Be6 31 d4 d5
Disallowing d4-d5 tricks.
32 Re1 Ra8
Akobian seizes the open a-file before White does.
33 Ke3 Ra2 34 Kf4
When you are already down three pawns, what is another one to you?
Black should convert without trouble after 34 Bd8 Rxb2 35 Bxc7 Rc2 36
Bd6+ Ke8 37 Kd3 Ra2.
34 ... Rxb2 35 Ra1
Threatening mate.
35 ... Ke8 36 Ra8+ Kd7 37 Rh8
White’s soon-to-be passed h-pawn isn’t enough to salvage the game.
37 ... Rf2+ 38 Kg3
Perhaps less torturous for White is 38 Ke5?? (sometimes I think a
mercifully quick death is better than the slow, lingering version) 38 ... Rf5
mate!
38 ... Rc2
 The rook picks off the c3-pawn and creates even more passed pawns.
39 Rxh7 Rxc3+ 40 Kg2 Rh3
White has a little combination, but it doesn’t save the game.
41 Rxf7+

Also futile is 41 Rh8 b4 42 h7 b3 43 Rd8+ Kc6 44 h8Q Rxh8 45 Rxh8 b2

46 Rh1 Bf5 and Black regains the lost rook with a won ending.
41 ... Bxf7 42 Kxh3 Bg8 43 Kg4 b4 44 Bd2 b3
“How can you outrun me when you are weighed down by your own
incompetence?” the pawn asks White’s bishop.
45 Bc3 Ke6 46 Kg5
White’s king is lost, having wandered off the trail, where one tree looks
much like the other.
46 ... Bh7!
The humble bishop subjects himself to humiliating penance – as long as in
the end, he wins the argument. Now the passed h-pawn’s hopes, like desert
clouds devoid of rain, are full of undelivered promises.
47 Bb2 c5 0-1
White’s bishop is overwhelmed by Black’s queenside passers and unable
to blockade after 48 Bc3 c4 49 Bb2 b5 50 Bc3 b4. “The blindfold of your
foolish faith is torn from your eyes, and now you clearly see that your God is
unwilling or unable to save you from my wrath,” the b-pawn lectures White’s
long-suffering bishop.
Summary: GM Wang Hao’s early ... b6 works just as well in the 7 Bf4 lines. This relatively new
idea also has the added benefit of perhaps surprising unprepared opponents.
 Chapter Six
The Three Knights Petroff

A typical attempt to sidestep the Petroff sees White bringing forth both his
knights in preparation for a Four Knights Game or Philidor’s Defence.

However, our last move 3 ... Bb4 challenges the assumption that White has successfully evaded our
attempts to play under a Petroff’s banner, and the position once again resumes its default state of our
opening. White’s best line is to capture our e5-pawn and then play Nd3, which picks up the bishop pair.
Now this will probably offer your opponent a tiny edge – perhaps even less than ‘+=’ – with not much
chance for our side to win. But if our overall goal is to draw (remember, this is Petroff’s!), then our line
is very playable. Another big advantage is that we don’t need to learn either The Four Knights or all its
offshoots, like the Scotch Knights (3…Nc6 4 d4 exd4 5 Nxd4) and Belgrade Gambit (3…Nc6 4 d4
exd4 5 Nd5). So from a practical view point, the Three Knights Petroff is a time-saving choice.

Game 43
J.Polgar-V.Ivanchuk
Amber blindfold, Monte Carlo (rapid) 1994

1 e4 e5 2 Nf3 Nf6 3 Nc3

White invites the Four Knights or Philidor’s Defence.

Question: If I don’t want to study those two lines, is

there a way I can remain within the Petroff’s umbrella?

Answer: Please see Black’s next move!

 3 ... Bb4

This line, the Three Knights Petroff, is exclusive to Petroff’s Defence. I

think it’s actually slightly inferior to the Four Knights and even the Philidor,
but it’s not such a bad line if you are okay with a draw:
a) 3 ... Nc6 reaches the Four Knights; while
b) 3 ... d6 4 d4 gets us to Philidor’s Defence.
4 Bc4
White plays it as if it were a reversed Berlin Lopez with the extra move
Bc4 tossed in. In the next two games, we will look at 4 Nxe5 and 4 d3.
4 ... Bxc3
If equality is what you are after, then this move is a good choice.
 Question: But are we not giving away the bishop pair?

Answer: We get a slight devaluation of White’s structure in return. Also, we plan to switch our
structure to favour our remaining bishop with ... d7-d6, and then challenge White’s bishop pair with ...
Be6. I’m just going cursorily examine 4 ... 0-0 5 0-0 d6:
a) 6 Nd5 Be6 7 d4, V.Ivanchuk-Bu Xiangzhi, Nanjing 2008, and Black
equalizes with 7 ... exd4 8 Nxd4 Bxd5 9 exd5 Nbd7 when White’s bishop
pair doesn’t constitute any kind of advantage in such a static structure.
b) 6 d3 Be6 7 Bxe6 fxe6 8 Ne2 Nbd7 9 c3 Ba5 10 a4 c6 11 Ng3 Qe7 12 Qe2 h6, B.Larsen-
V.Smyslov, Las Palmas 1972. Larsen went on to win brilliantly in the ending, which I annotated in
Larsen: Move by Move, but at this stage, the game is even.
5 dxc3

Question: It feels to me like some of the dynamism of White’s

position is removed by this recapture away from the centre.
Why not 5 bxc3, intending to follow with a future d2-d4?

Answer: This is also possible. Play may go 5 ... d6 (5 ... Nxe4 6 d3!? is an untested idea when 6 ...
Nxc3 7 Qd2 d5 8 Qxc3 dxc4 9 Qxe5+ Qe7 10 Qxe7+ Kxe7 11 Ba3+ Kf6 12 Bb2+ Ke7 13 Ba3+ ends
in a draw) 6 0-0 0-0 7 d3 Be6 8 Bb3 h6 9 Nd2 (intending f2-f4) 9 ... Nc6 10 f4, M.Munoz Pantoja-
S.Beltran Rueda, Barcelona 2008, and now 10 ... exf4 11 Rxf4 Ne5 looks dynamically balanced.
5 ... d6 6 0-0
There is no real need to protect the e4-pawn with 6 Qe2 but White’s idea is
to clear the path for castling long. After 6 ... 0-0 7 Bg5 Nbd7 8 0-0-0 Qe8!? 9
g4 Kh8! (for now, Black ignores the g4-taunt and makes room on the g8-
square for the knight) 10 h4! Nxg4! (now that h2-h4 is played, this move is
more tempting, since it will be more difficult to remove the g4-squatter) 11
h5 Nb6 12 Bb3 a5, M.Todorcevic-N.Nikcevic, Igalo 1994. I don’t believe in
White’s compensation after 13 a3 h6 14 Bh4 f5.
6 ... Nbd7
Most definitely not 6 ... Nxe4?? (some ‘combinations’ should be avoided,
even when we spot them) 7 Qd5 Be6 8 Qxe4 d5 9 Qxe5 dxc4 10 Qxg7 and
Black’s dark squares collapse.
7 Re1 0-0 8 a4 Nc5 9 Nd2
White covers the e4-square, hoping for time to play f2-f3 and Nf1-e3.
9 ... a5
Seizing a little queenside space. After 9 ... Be6 10 Bf1 a5 11 b3 Nfd7 12
Nf3 Bg4 13 h3 Bh5 14 Re3 Ne6 15 Qe1 Qf6 16 Nd2, I already prefer Black,
who can continue to build on the kingside. I don’t see a useful plan for White,
B.Pott-M.Lacunza, correspondence 2011.
10 Qf3?!
When we place our faith in an unsound plan, we pledge allegiance to an
essenceless entity. Polgar is an attack-minded player, even in a blindfold
game. She transfers her queen to the kingside to build in that region.
However, she wanders away from the correct plan’s hub, since the queen is
awkwardly placed and only manages to get chased around at a loss of time.
She should probably give up on kingside attacking dreams and make a more
harmonious move like 10 f3 which frees her knight to transfer over to the f1-
square, and then either to g3 or e3, depending on the circumstances.
10 ... Bg4
This urges White to play the queen to g3, a square to which perhaps Polgar
wanted to occupy with her knight.
11 Qg3 Be6 12 Bf1
Principle: The side with the bishop pair should try and hang on to both of
them if possible. However, in doing so, the bishop is forced into a retreat.
Also, 12 Bxe6 fxe6 isn’t a very tempting option, since it opens the f-file.
12 ... Nh5!
 Ivanchuk clears the path for a future ... f7-f5, which opens the f-file.
13 Qf3?
This move accentuates rather than hides White’s strategic flaws. The
queen is not well-placed on the f-file, which soon opens. She had to try 13
Qe3 with a still playable game.
13 ... Nf4 14 Nc4
Going after the f4-knight, but Black’s play arrives first. Instead, after 14
g3 Nh3+ 15 Kh1 Ng5 16 Qe3 f5 17 exf5 Bxf5, suddenly it is very hard to
defend the c2-pawn without a concession of some kind.
14 ... f5!
This move is doubly irresistible with White’s queen on the open f-file.

Question: Here we have a case of conflicting points:

a) ... f5 gives Black the initiative, so it should be played.

b) ... f5 opens the position, which helps White, since she owns the bishop
pair.
How does Ivanchuk know which one to follow and which one to avoid?
Answer: From my experience, short-term initiative tends to trump long-term issues, like opening
the position for the bishop pair, for one simple reason: If your initiative develops to decisive
proportions – as it did in this game – then there is no ‘long term’ for White, since the game is over!
15 exf5
 Most definitely not 15 Bxf4??, which is met with 15 ... fxe4 16 Qd1 Rxf4
winning a key pawn for zero compensation.
15 ... Rxf5 16 Ne3?
White has a bleak position anyway:
a) 16 Bxf4?? hangs a piece to 16 ... Rxf4.
b) Relatively best is 16 Qe3 but it should probably still lose after 16 ...
Bd5! 17 g3 (17 f3? Nxg2! and White can resign, since she either loses a
second pawn, or faces 18 Kxg2 Rxf3 19 Qxf3 Bxf3+ 20 Kxf3 Qh4 21 Be3
Rf8+ 22 Ke2 Qg4+ 23 Kd2 Ne4+ 24 Kc1 Rxf1! 25 Rxf1 Qe2 26 Rd1 Qxc4
and Black wins) 17 ... Qf6! 18 Nd2 Rf8 19 f3 Nfe6 as White won’t survive
the onslaught.

Exercise (combination alert): This is the fairy tale question of the would-be
heroic e3-knight, attempting (and failing miserably, I might add!) to
rescue the f3-princess, who for now is threatened by multiple ogres.
Find one powerful idea and Black’s world will come tumbling down upon
her:

Answer: Discovered attack/double attack/pin.

16 ... Nxg2!
The white queen and king’s perpetually fixed smiles go slack at the sight
of the knight’s rude entry into their bed chamber.
17 Nxf5
Note that 17 Qxg2 Rg5 pins the queen.
17 ... Nxe1 18 Qe2 Bxf5 19 Qxe1
Not only did Black win a second pawn, but the c2-pawn is doomed, and
White’s king is grievously weakened. White can resign here.
19 ... Qf6
The c2-pawn isn’t going anywhere since Black can play ... Qg6+ if White
defends it.
20 Bg2 Qg6 21 Kh1
White’s king walks around with dark rings under his eyes, suffers a queasy
stomach, and pops Xanax tablets from a Pez dispenser, with dark thoughts of
ruin and desolation – pretty much all normal for him when he awaits his
sister’s coming visit.
21 ... Rf8
Ivanchuk recruits his final attacker. There is nothing wrong with the
mundane-looking 21 ... Bxc2 which has the effect of not only winning
another pawn, but also weakening the b3- and d3-squares. But of course, such
blatant greed is psychologically distasteful to most players and is similar to
the billionaire who takes in lodgers in his mansion to make a few extra
dollars each month.
22 b4 Be4
Going after White’s only defender. “Pray for your life’s salvation until
your throat is hoarse. It won’t alter your fate even a notch,” the bishop boasts
to his g2-counterpart.
23 f3
Unsurprisingly, 23 Bxe4 Nxe4 24 Be3 Qh5 obviously isn’t going to save
White.
23 ... Rxf3!
This is a little tricky, since Ivanchuk had to calculate the consequences of
White’s next move.
24 Qe2 Qg4! 25 bxc5
 Why not? Everything else loses.

Exercise (combination alert): Black attackers rain down on White

until they blot out the sun. How did Ivanchuk force the win?

Answer: Clearance/pin/weak back rank.

25 ... Rf7! 0-1
The weakness of White’s back rank allows 26 Qxg4 (26 Qxe4 Qd1+
forces mate as well) 26 ... Rf1 mate! White’s king calls for his pipe, bowl and
fiddlers three – none of which he gets, since all his servants seem to have
abandoned him.
Summary: After 4 Bc4, our clearest route to equality is simply to play 4 ... Bxc3, where White gets
a move-up version of the Berlin Lopez, but no advantage.

Game 44
E.Sutovsky-K.Sakaev
Serbian League, Kragujevac 2009

1 e4 e5 2 Nf3 Nf6 3 Nc3 Bb4

In The Four Knights: Move by Move book, I called this the Four Knights
Petroff. Math is not my strong suit. I only count three knights out!
4 Nxe5!
 In my opinion, this is White’s only way to an edge in this chapter. This is
the only line covered for White in the above-mentioned book, which
advocates the opening from White’s perspective.

Question: What is so great about this move? It doesn’t win a pawn.

Answer: But it does pick up the bishop pair, and in my opinion, a modest ‘+=’ edge for it.
4 ... 0-0 5 Be2
Also quite common is the move order 5 Nd3 Bxc3 6 dxc3 Nxe4 7 Be2 Re8
8 Nf4 which transposes to the game continuation.
5 ... Re8 6 Nd3
This move insures White the bishop pair.
6 ... Bxc3

Question: Can Black avoid giving up the bishop pair

and sacrifice a pawn to tangle White up with 6 ... Ba5?

Answer: Double question mark! Black’s knight has no safe haven after 6…Ba5?? 7 b4! Bb6 8 e5
which steals a piece in broad daylight.
7 dxc3
White owns the bishop pair, so he logically opens central lines. 7 bxc3 is a
bit contorted as after 7 ... Nxe4 8 0-0 Nc6 9 Bb2 d5 10 Nf4, G.Todorovic-
D.Marjanovic, Obrenovac 2013, I actually prefer Black’s game after 10 ...
Qd6.
7 ... Nxe4 8 Nf4!

Slightly more accurate than castling, since White clamps down on the d5-
square, with c3-c4 to soon follow.
8 ... d6
Less accurate is 8 ... c6?! intending ... d7-d5. This doesn’t make much
sense, since White responds with 9 c4 when Black’s d-pawn is now backward
and a potential future target if it plays to the d6-square later on, L.Psakhis-
A.Yusupov, Vilnius 1981.
9 0-0 Nc6
White has the bishop pair and a nagging long term ‘+=’ position. Still, if
your goal is to hold White to a draw, Black’s position is fully playable.
10 a4!?
White hopes to increase queenside space. This new move is no
improvement over the more logical 10 c4 Bf5 11 f3 Nc5 12 Rf2 a5 13 Bf1
when White can hope to milk something from his bishop pair, R.Antonio-
A.Minhazuddin, Guangzhou 2010. Interestingly though, Houdini isn’t
impressed with White’s alleged edge and rates the position dead even. To my
human eyes, I feel like White’s bishop pair, in conjunction with his control
over the d5-square, confers a slight edge.
 10 ... h6
This is also Houdini’s first choice, while I would just develop with 10 ...
Bf5.
11 a5 a6
Otherwise Black has to be on constant lookout for a5-a6, which
undermines support for the c6-knight.
12 c4 Bf5 13 Ra3!?
Now the real idea behind Sutovsky’s ambitious plan is revealed. He plans
to swing his rook over to the kingside at some point to generate an attack on
Black’s king. However, the obvious downside of the plan is that the second
his rook moves off the a-file, White’s a5-pawn hangs.

13 ... Nf6
This cuts off Nh5 ideas from White.
14 Rg3!?
Sutovsky dares Sakaev to grab the a5-pawn.
14 ... Ne4
No thanks! This is the equivalent of a draw offer.

Question: What happens if Black does take the a5-pawn?

Answer: Let’s take a look: After 14 ... Nxa5!? 15 b4! (White’s dark-squared bishop will enter the
attack via the b2-square) 15 ... Nc6 16 Nd5 (now h5-h6 is threatened) 16 ... Ne4 17 Re3 Ne5 18 Bd3
Nxd3 19 cxd3 Nf6 20 Rxe8+ Nxe8 21 Re1 (threat: Ne7+) 21 ... Be6 22 Nf4 Bd7 23 Nd5 Nf6 24 Ne7+
Kh7 25 Qf3 (now Black must find the tricky defence) 25 ... Be6! (25 ... Rb8? 26 Bb2 leaves Black in
real trouble) 26 Qxb7 Qb8 27 Qxb8 Rxb8 28 Bd2 c5, the ending is even. So it appears that Sutovsky’s
initial a5-pawn offer was sound.
15 Re3 Nf6 16 Rg3 Ne4 17 Re3 Nf6 18 Rxe8+!?
Sutovsky continues to play for the win. Another way to avoid the draw
was 18 Nd5! Nxd5! (18 ... Nxa5? 19 Rf3! Re5 20 b4 Nc6 21 Bb2 wins
material) 19 cxd5 Rxe3 20 fxe3 when two Black pieces are attacked, but
there is no problem, since his next move covers both. Following 20 ... Ne7 21
c4 Qd7 22 b4 Re8 23 Bb2 Ng6 24 Qd4 f6, White’s queenside space and
bishop pair are counterbalanced by Black’s control over the e5-square, and
also the fact that White has a potentially weak, backward e-pawn. Houdini
prefers White, and if I had to pick a side, I suppose I would agree with
Houdini.
18 ... Qxe8 19 Re1!?
An innocent habit can easily turn into a crutch if we shun introspection of
the consequences of our risk-taking. Sutovsky continues to offer the a5-pawn,
this time perhaps a tad unsoundly. We can see him slipping into that awful
delirium we all fall into at times, where we attempt to win, when common
sense begins to whisper to us to back down and play it safe. Instead, 19 Bd2
Qe5 is a bit awkward for White though he gets to stay even on material!
19 ... Nxa5!
“Social opprobrium only applies to ordinary people,” declares the knight,
as he ignores the villager’s angry glare, while confiscating food supply
intended to get them through the winter. This time the pawn grab isn’t nearly
as risky as it was a few moves ago. I’m not so convinced by White’s
compensation here, despite his bishop pair and slight initiative.
20 Nd5?!
After this move, Black seizes the initiative. Sakaev recommends
interpolating with 20 g4! before playing Nd5. Following 20 ... Bh7 21 Nd5
Nxd5 22 cxd5 c5 23 b3 b5 24 f4, White seems to have adequate
compensation for the pawn, due to:
1. Black’s offside knight.
2. White may play for f4-f5, shutting out Black’s bishop, or g4-g5, going
for an attack.
3. White owns more space and has the bishop pair.
 20 ... Nxd5 21 cxd5
White menaces b2-b4, as well as Bd3, both which win a piece.
21 ... Qe4!

This powerful centralization swings the game to Black’s advantage.

Sakaev halts b2-b4 and covers against Bd3, while threatening ... Qxc2.
22 Bd2
Let’s see the alternatives:
a) 22 Bd3? fails to 22 ... Qxd5 23 Qh5 Be4! which forces White to swap
queens, two pawns down in the ending.
b) White may have better chances to save the game with 22 Bxh6! gxh6 23
Bd3 Qxd5 24 Qh5 Nc6 25 Bxf5 Qd2 26 Kf1 Qg5 27 Qf3 Rd8 28 c3 d5 when
this won’t be so easy to convert for Black, who must nurse a broken kingside
structure. Also, White’s bishop may be the superior minor piece.
22 ... Qxd5
Pawn number two falls. White’s king orders all the locks in the castle to be
reinforced, a normal precaution, since his thieving d5-sister visits.
23 Bf3 Qb5 24 Bc3
White cannot win one pawn back as 24 Bxa5? Qxa5 25 Bxb7 Rb8 26 Bd5
is met with 26… Bxc2! when White remains down two pawns for zero
compensation.
24 ... Nc6 25 Bd5 Rf8!
 Black’s king gratefully views his defender in the similar light of a
snowbound and exhausted mountain climber, upon seeing the Saint Bernard
approach with wagging tail and tumbler of brandy. Black covers the sensitive
f7-square. White has a clear initiative but is it worth two pawns? I doubt it.
Black’s fate – both present and future – revolves around how well he is able
to protect his king.
26 h3
Council for the defence seems to have strong evidence for their case.
White’s last move is a clear sign that he is unable to find a way to stoke his
initiative. Perhaps the immediate 26 h4 should be played, since it saves a
tempo over the game’s continuation.
26 ... Kh7!
A useful defensive move. Black’s king is safest on the h7-square, which
also allows him to play for the unraveling idea ... f7-f6 and ... Ne5.
27 h4
The rook lift 27 Re3 does not suffice after 27…f6 28 Rg3 Ne7 29 Bb3 d5
and White’s attack halts before it even begins.
27 ... f6 28 f4?!
Now White is just losing after this weakening pawn stab. Instead, 28 Qd2
keeps his disadvantage to a minimum.
28 ... Qc5+!
 Apparently, Time, the great healer, didn’t get the job done, and it will be
quite a while before the queen forgives or forgets past wrongs.
29 Kf1
The other king move 29 Kh1? allows 29 ... Qf2! with a dual attack on the
h4- and c2-pawns.
29 ... Nb4!
Double attack on the d5-bishop and c2-pawn. When we are pushed around
for long enough, we get the compulsion to fulfill a legitimate aspiration: push
back. White’s initiative is fast disappearing.
30 Bb3
The greedy 30 Bxb7?? hangs a piece to 30 ... Qb5+.
30 ... d5
Black hopes to smother the b3-bishop’s influence. Stronger was the plan 30 ... a5! intending ... Bd7
and ... Bb5+.
31 Qh5
Nabbing the d-pawn with 31 Bxb4 Qxb4 32 Qxd5 Qxf4+ loses for White.
31 ... Bg6
Sakaev continues to miss the idea of transferring his bishop to the b5-
square. Stronger was 31 ... Bd7! 32 Bxb4 Bb5+! 33 c4 Bxc4+ 34 Bxc4
Qxc4+ 35 Re2 Qxf4+ 36 Qf3 Qxb4 37 Qxd5 f5 when Black is up three
pawns for nothing.
 32 Qg4 d4
Good enough, but stronger was 32 ... Nxc2! 33 f5 Qb5+ 34 Kg1 Bh5! 35
Qxh5 Nxe1 36 Qd1 Nd3 and Black consolidates.
33 Bxb4 Qxb4 34 Kf2
The careless 34 h5? is met with 34 ... Qb5+ picking off the h-pawn.
34 ... d3!

Principle: Meet a wing attack with a central counter. Multipurpose:

1. Black pins White’s f-pawn.
2. Black halts h4-h5, due to the ... Qc5+ double attack.
3. Black overloads White’s c-pawn and threatens ... dxc2.
35 h5
Hey, I just said “2. Black halts h4-h5, due to the ... Qc5+ double attack!”
35 ... Qc5+
The queen addresses herself to the pleasant task at hand, of correcting her
f2-brother of his many flaws.
36 Kg3 Bf5!
Even stronger than taking the h5-pawn.
37 Qd1
The once powerful queen must adjust to her new job as palace front-gate
hostess.
37 ... dxc2 38 Qd2 Kh8
 More accurate is 38 ... a5!.
39 Kh4!?
The less you have to lose, the easier it is to pick a fight with a well-armed
opponent. Well, nobody can accuse Sutovsky of lacking fighting spirit in this
game! White prepares g2-g4 but Black is not in the least worried.
39 ... Qb6 40 Re3
Also, 40 Bxc2 Qxb2 is completely hopeless for White.
40 ... a5
Both ... a5-a4 and ... Qb4 are in the air.
41 g4

Exercise (calculation): 41 ... Bh7 is winning. But can Black

get away with 41 ... Be4? Calculate the consequences.

41 ... Bh7
Answer: The centralizing trick 41 ... Be4! works due to tactics. If 42 Rxe4?, the deflection 42…
c1Q! tactic allows Black to seize control of the f2-square. After 43 Qxc1 Qf2+ 44 Kh3, 44…Qf3+
picks up the loose e4-rook.
42 Rc3 c6 43 f5 a4! 44 Bxa4 Qxb2 45 Rxc2 Qb1
Black remains up two clean pawns.
46 Rc1
Played to stop ... Qh1+.
 46 ... Qb6 47 Bc2 Rd8 48 Qf4 Bg8
At long last, Black’s pieces coordinate and emerge.
49 Rb1 Qc5 50 Be4 Qe7 51 Re1 Rd4 52 Re2 Kh7 53 Qf3 Qd6 54 Qg3
Qb4
A sign that White’s initiative is no more. Sakaev doesn’t bother to enter
the two pawns up ending with 54 ... Qxg3+.
55 Bf3

Exercise (combination alert): Black to play and win.

Answer: Pin.
55 ... Rd3! 56 Qc7
Wow! Talk about desperation! The queen is a one-woman army,
reminding us of Al Pacino at the end of the movie Scarface. However, 56
Qg2 fails to 56 ... Qf4 57 Rf2 Bd5 58 Bxd5 Qg5 mate!
56 ... Rxf3 57 Re3
 Exercise: I’m going to end the game by tossing you an
easy one. White just offered us a rook. Should we take it?

Answer: No! Don’t touch the rook. It’s a trap! White’s act of generosity is largely of a symbolic
nature, since Black is unable to accept the gift.
57 ... Rf4 0-1
White’s king is unable to throw off his oppressor’s tyrannical yoke. Most
definitely, Black should avoid 57 ... Rxe3?? as White creates an impregnable
defence through which no light penetrates, with 58 Qxg7+! Kxg7 stalemate!
Suddenly, Black’s claim to power is delegitimized. “History is written by the
survivors, not the dead,” gloats White’s king, who is now able to amortize his
past debt and square his account at even.
Summary: 4 Nxe5! leads to a tiny but manageable edge for White, due to the bishop pair.

Game 45
A.Miltner-J.Werle
German League 2004

1 e4 e5 2 d3
Our move order runs 2 Nf3 Nf6 3 Nc3 Bb4 4 d3.
 Question: Why is White playing so passively?

Answer: White simply seeks to play a reversed Ruy Lopez a move up from normal. He probably
isn’t after some large opening advantage, but simply wants a position in which he feels comfortable. I
remember, after writing The Colle: Move by Move, reading a post by a player who said he was
disconcerted by the fact that I wasn’t trying hard enough to get White a ‘+=’ advantage in every
variation. The unfortunate fact is: Some openings like Colle and London System are inherently even,
and with best play from Black, there is no true path to a clear advantage for White from every variation.
If a writer tells you there is, he or she is lying. I still play them because I have decades-long experience
in such structures and count on outplaying my opponents later on in from an even position. So White,
in a sense is doing the same thing. He does not strive for an opening edge in a position in which he
feels comfortable with, and probably plays a move down as Black. After 4 ... Nc6, we transpose to our
game’s position.
2 ... Nc6 3 Nc3 Nf6 4 Nf3 Bb4 5 Bd2
Let’s look at other White tries:
a) 5 Be2 d5 6 exd5 Nxd5 7 Bd2 Bxc3 8 bxc3 0-0 9 0-0 Qd6 10 c4 Nf4
(There is nothing wrong with 10 ... Nf6 when White must be on the constant
look out for ... e5-e4) 11 Bxf4 exf4 12 d4 Bg4:
a1) 13 d5?! (this move allows Black the superior minor piece) 13 ... Bxf3
14 Bxf3 Ne5 15 Qe2 Rfe8 16 Rfe1 Qb4 and White was already in trouble,
with pressure on the c4-pawn and indirectly, the e1-rook, E.Zemantelian-
A.Kosten, Paris 2002.
a2) 13 c3 Rfe8 and Black stands no worse, since his slightly inferior kingside pawn majority is
counterbalanced by White’s somewhat damaged queenside structure.
 b) The 5 Bg5 pin doesn’t bother Black a bit, since he hasn’t yet castled
kingside. Following 5 ... h6 6 Bh4 g5 7 Bg3 d6 8 Be2 Be6 9 a3, D.Anton
Guijarro-K.Georgiev, Dubai (blitz) 2014, I prefer Black’s position after 9 ...
Bxc3+ 10 bxc3 Qe7 since both of White’s bishops are rather unimpressive
for the moment.
c) White can try to earn the bishop pair with 5 a3 at the cost of time. After
5 ... Bxc3+ 6 bxc3 d5 (the more modest 6 ... d6 also looks even to me) 7 exd5
Nxd5 8 c4 Nde7 9 Bb2 f6 10 d4!? (otherwise Black may later clamp down on
the d4-square with ... Bc5 and even ... Nf5) 10 ... exd4 11 Nxd4 Nxd4 12
Qxd4 Qxd4 13 Bxd4 Be6 14 0-0-0 0-0-0, Black’s perfect structure easily
makes up for White’s bishop pair, F.Meyer-E.Chevelevitch, Germany 2007.
d) A g3 Vienna game-like position can occur after 5 g3 d5 6 exd5 Nxd5 7
Bd2 Nxc3 8 bxc3 Bd6 9 Bg2 0-0 10 0-0 Bg4 11 Qb1 Rb8 12 Re1 Qd7 and
Black stands no worse, P.Held-P.Gomolla, Wuerttemberg 1998.
5 ... d5
Black’s most logical response, which frees his position. Of course there is
nothing wrong with 5 ... d6 or 5 ... 0-0.
6 exd5

White agrees to the ‘little centre’ formation.

Question: Why is White willing to give his opponent greater central control like this?
 Answer: For two reasons:
1. The central pawn exchange exd5 enhances his light-squared bishop after
a kingside fianchetto.
2. He doesn’t have much choice, since 6 Be2 d4 7 Nb1 Bxd2+, A.Belov-
Y.Rjabzev, Tula 2002, sticks White with a remaining bad bishop. Black
already stands slightly better,
6 ... Nxd5 7 g3
The fianchetto places White’s light-squared bishop on an active diagonal,
at the cost of weakening the light squares around his king. The passive 7 Be2
transposes to the Zemantelian-Kosten note from above.
7 ... 0-0 8 Bg2 Bg4
This pin is irritating for White, since h2-h3 and g3-g4 weaken White’s
kingside structure.
9 h3 Bh5 10 0-0
10 g4!? is possible and 10 ... Bg6 11 Nxd5 Qxd5 12 Nh4 Bxd2+ 13 Qxd2
Qe6 14 Nxg6 fxg6!? seizes control over the f4-square. The position looks
approximately balanced after 15 c3 Rad8 16 Qe2.
10 ... Nde7

This new move is based on the principle: The side with extra space should
avoid swaps. Instead, 10 ... Bxc3 11 bxc3, A.Onischuk-D.Pavasovic, Halle
1995, looks balanced after the stabilizing 11 ... f6.
11 g4
White refuses to be a prisoner to an endless pin and pushes forward.
11 ... Bg6 12 Nh4 f6 13 f4
White can engineer to win the bishop pair with 13 a3 Bxc3 14 Nxg6 Nxg6
15 bxc3 Qd6 leading to an unbalanced position, where I believe Black’s
knights, at least for now, hold their own against White’s bishop pair.
13 ... exf4 14 Rxf4 Bf7
White had his chance to take the g6-bishop but declined to do so. So Black
removes this option with the game opening up.
15 Rf1 Bxc3!? 16 bxc3 Bd5!
Principle: When your opponent has the bishop pair, eliminate one of them
if possible. This move also removes a strong defender of White’s king.
17 Qe2 Qd7 18 Rae1 Rae8 19 Qf2 b6 20 Nf3 Ne5!
A well thought out embracing of a structural defect. The position remains
dynamically balanced.
21 Nxe5 fxe5

Question: Why did Black willingly take on an isolani?

Answer: The isolani, although perhaps a potential future weakness, for now worries White about
ideas like ... Ng6, when White must be on the constant lookout for ... Nf4.
22 Qg3 Ng6
 White’s position, while photogenically attractive, contains hidden flaws:
1. White, on eternal sentry duty, must keep both eyes open for a prowling
black knight on either the f4- or h4-squares.
2. White must also worry about future ... Qa4 queen raids, terrorizing
pawns on the queenside.
23 c4?!
It’s unwise to cede a precious resource to the opponent without getting
anything in return. Correct is 23 Be3 though Black retains the better minor
piece and pawn structure.
23 ... Bxc4?!
Tempting, but stronger was 23 ... Bxg2! 24 Qxg2 Rxf1+ 25 Rxf1 (25 Kxf1
is met with the pawn-hunting 25…Qa4) 25 ... e4 26 Qg3 exd3 27 cxd3 Qd4+
28 Qf2 Rd8 29 Qxd4 Rxd4 30 Rf3 Ne5 and Black picks up a pawn.
24 Rxf8+ Rxf8
 Exercise (combination alert): White is down a pawn. However, if you spot his
combination, he reaches a position where he gets full compensation. How?

25 h4?
In music, the space between the notes is as important to the finished
product as the notes themselves.
Answer: White missed the umpinning double attack 25 Bb4! which functions as both diagnosis and
cure. White regains the lost pawn after 25 ... c5 26 Bxc5 bxc5 27 dxc4 Qd4+ 28 Qe3 Qxc4 29 Qb3
Qxb3 30 axb3 when Black nurses three isolanis, thus offering White full compensation for the pawn.
25 ... Bd5
Now White is just down a pawn with a loose structure.
26 Be4
This stopgap measure isn’t enough to dull the pain. This is a mistake in an
already lost position. Also futile is 26 Rf1 Rxf1+ 27 Bxf1 Bxa2 28 h5 Nf4 29
Bxf4 exf4 30 Qxf4 Be6 as White is unlikely to save himself.
26 ... Bxe4 27 Rxe4 Qf7!
Menacing the a2-pawn as well as the f1-square.
28 Re1 Nf4!?
Threat: ... Ne2+!. We are reminded of Kurt Cobain’s words: “Hello, hello,
hello, how low.” There was nothing wrong with the slightly greedy 28 ...
Qxa2.
 29 Kh2
White’s inconvenienced king is forced to revise his social agenda for the
evening. Most definitely not 29 a3?? Ne2+! (when we miss the final move of
a sequence, it’s as if we survived the rigors of sailing around the world, only
to crash the boat into the dock at the very end of the trip) 30 Rxe2 Qf1+ and
Black wins.
29 ... Qxa2
White’s king – who regards his sister the way a 1970’s heavy metal rocker
viewed disco – comes to a gentleman’s agreement with her, which she
violates immediately, since he forgot that she is no gentleman – or lady, for
that matter. Okay. Now he decides to take it, when it is even stronger.
30 Rxe5 Qxc2 31 Qe3
White has nothing better to do than to watch all his pawns fall, with polite
attention.
31 ... Qxd3 0-1
Summary: The passive but solid 4 d3 line reaches a reversed Ruy Lopez position, where Black
easily equalizes.
 Chapter Seven
Odds and Ends

We wrap up the book by taking a look at a few eccentric attempts from

White, who reckless to the rules of precedence, decides to block the orderly
procession with a series of attempted theoretical dodges.

In the first diagram, we simply take White’s hanging e-pawn with the
intention of returning it later to achieve equality.
 The second diagram is an inferior version of Chapter Five, where White
captures toward the centre, hoping his or her future kingside attacking
chances make up for his slightly damaged structure.

The third diagram leads to similar positions, when we take the e4-pawn, White responds with Nc3
and bxc3. We note a forced, artificial quality in all of White’s attempts, the way a desperate football
coach attempts to rouse his insecure team with an inspirational speech. In all cases, basic familiarity
with the positions offers us equality without the angst we suffered in previous, more theoretical
chapters.

Game 46
M.Nielsen-G.Sargissian
Politiken Cup, Helsingor 2009

1 e4 e5 2 Bc4
Our move order usually runs 2 Nf3 Nf6 3 Bc4 Nxe4.
2 ... Nf6 3 Nf3

Question: White ignored the e5-pawn and also refused to protect

his e4-pawn. I assume he plans Ng5. How should Black proceed?

Answer: Chopping the e4-pawn solves all of Black’s troubles.

3 ... Nxe4
We take the pawn, not motivated by greed, but more as a cushion. At the
right moment we can return the pawn and attain equality at a minimum. Of
course, there is nothing wrong with 3 ... Nc6 except then you might have to
study the Two Knights Defence with its sharp branches like the Fried Liver
Attack, the Max Lange Attack, the Traxler Counterattack et al.
4 Nc3
White’s idea: he plans to make a gambit of it, banking on his development
lead.
 4 ... Nd6!?
This is only if you are a purist and insist on the independent Petroff’s
move. Your best bet may be to transpose to Four Knights with 4 ... Nc6!
which enters a known good line for Black of the Four Knights. White scores
a depressingly low 28.6 % from this position in my database:
a) The trigger reaction 5 Bxf7+? Kxf7 6 Nxe4 d5 isn’t very appealing for
White. Black dominates the centre, will soon have a lead in development,
owns the bishop pair and may make use of the open f-file. These factors
easily outweigh the fact that Black’s king moved and can’t castle.
b) White can try to gambit a pawn with 5 0-0 White and depend on his
development lead to provide compensation. The safest path for us is to
initially grab the pawn, and then immediately return it for easy equality with
5 ... Nxc3 6 dxc3 Be7! 7 Qd5 0-0 8 Nxe5 Nxe5 9 Qxe5 Bf6 10 Qh5 d6 11
Re1 Bd7 12 Be3 Bc6. White had nothing from the opening, G.Battaglini-
V.Malaniuk, Calvi 2010.
c) Allowing the pawn fork with 5 Nxe4 d5 6 Bd3 dxe4 7 Bxe4 Bd6 8 d4 exd4 9 Bxc6+ bxc6 10
Qxd4 0-0 gives Black’s bishop pair and open lines more than sufficient compensation for his queenside
pawn weaknesses. After 11 Be3 Be6 12 0-0 Bd5 13 Ng5 f5 14 Nh3 Re8 15 Bg5 Qb8 16 Rfe1 Qb6 17
Qxb6 cxb6. Black’s queenside pawn weaknesses have evaporated, yet his bishop pair remains, so it is
White who fights for the draw, J.Koch-M.Adams, France 2001.
5 Bb3 e4 6 Qe2
After 6 Ne5 Nc6 7 d4 exd3 8 Nxd3, L.Havrdova-S.Studeny,
correspondence 1996, I don’t see enough compensation for White with 8 ...
Nf5 9 0-0 Be7.
6 ... Be7 7 Nxe4 Nxe4 8 Qxe4 c6
Intending ... d7-d5 next.
9 d4
This move allows Black his intent. Instead, 9 c4!, J.Vorisek-P.Pokorny,
Czechia 2011, is the critical continuation. I took Black versus Houdini and
the game went 9 ... Na6 10 d4 0-0 11 0-0 Re8 12 Re1 Nc7 (…d7-d5 is in the
air) 13 Bf4 d6 14 d5 cxd5 15 cxd5 Bg4 16 Bg3 Bh5. I would be happy to take
Black here. I get such positions from Alekhine’s Defence, and this one looks
like a version where Black equalized.
9 ... d5
White’s light-squared bishop continues to recite his daily prayers, but
listlessly and with wavering faith. Black gets an Exchange French position
where White’s light-squared bishop is out of play. Of course, this may not
mean much since he can play c2-c3, but it costs time, and I already would
take Black if given a choice.
10 Qd3 0-0 11 Bf4 Nd7 12 0-0-0!?

White confronts a philosophical impasse and takes a path laden with

implication for his king. His last move is based on the philosophy: The safe
path doesn’t necessarily keep us safe. Sometimes we get into trouble, not by
doing something, but by not doing it. So he goes for it. White has a choice:
Turn and face the menace or turn and have it follow him. This move maybe
deserves a dubious mark. White seeks to implement a plan which is riddled
with tribulation and lacks the necessary essentials to build its foundation. He
is clearly not playing for a draw, despite the 650+ point differential in rating.
Of course, the question arises: why did the 2667-rated GM Sargissian allow a
Petroff’s on an opponent rated around 2000?

Question: Why don’t you like White’s plan of opposite wing attacks?

Answer: A weakness is sewn into White’s position, which will be difficult to remove without
tearing the fabric around his king: Black gains time on the b3-bishop with ideas like ... b7-b5 and ... a7-
a5. This forces White to weaken his queenside structure further, meaning Black’s attack will be faster. I
would continue frugally with 12 0-0 which leads to a typically boring and even Exchange French-style
position. White probably thought his much higher rated opponent would eventually ground him down
in such a position.
12 ... Nf6 13 Ng5?!
 White certainly isn’t intimidated by his GM opponent. We have seen this
idea from the previous chapter, where White intends to meet ... h7-h6 with
h2-h4. However here I don’t think the idea works all that well.
13 ... g6
Black should call the bluff with 13 ... h6! and White must retreat, since 14
h4?? doesn’t look sound. (anyway, the ‘normal’ 14 Nf3 Ne4 15 Qe3 Re8 16
Rhe1 Bf5 leaves White’s queen awkwardly placed on the open e-file) 14 ...
hxg5 15 hxg5 Ne4 16 f3 Bxg5 17 Kb1 Bxf4 18 fxe4 Bh6 and White doesn’t
have enough for the piece.
14 h4 Bf5 15 Qd2 a5
Threatening to win a piece with ... a5-a4.
16 c3 b5 17 f3
The belligerent lunge 17 h5? (when we complicate from an inferior
position, we attempt to hide our position’s ugliness through a screen of
smoke) is met with 17 ... Nxh5 18 Rxh5 gxh5 and I don’t see a reasonable
follow up to White’s attack.
17 ... b4
Lines pry open quickly on the queenside.
18 Qf2?
White’s would-be attackers move about in trance-like fashion. He can’t
afford this waste of time. He had to try 18 g4 a4! 19 Bc2 Bxc2 20 Qxc2 b3!
21 axb3 axb3 22 Qxb3 Bd6 when I still like Black’s attacking chances on the
queenside.
18 ... bxc3 19 bxc3 a4
“Servants are invisible to the powerful,” says the a-pawn, as he pulls out a
blade in the bishop’s presence.
20 Bc2 Qa5
The queen condemns White’s king for rumours which she herself created.
21 Bxf5?
White had to try 21 Kd2 Bxc2 22 Kxc2 Nh5 23 Bh2 Rfe8 24 g4 Ng7
though Black’s upper hand is in no doubt.
 Exercise (combination alert): White’s king, who awakens to each day
with the conviction that today will be much like yesterday, is about to
undergo a paradigm shift in his world view. Black to play and force mate:

Answer: Zwischenzug/annihilation of king’s cover.

21 ... Qxc3+! 0-1
The loose c3-pawn is the rallying point for White’s discontentment. After
22 Qc2 (or 22 Bc2 Ba3+ 23 Kb1 Qb2 mate!) 22 ... Ba3+, “There is no reason
for me to pray, since I already attained all that I ever wanted in life,” thinks
the bishop, who fulfils his lifelong dream to topple White’s king. White
succumbs after23 Kb1 Rab8+ 24 Bxb8 Rxb8+ 25 Qb3 Rxb3+ 26 axb3 Qb2
mate!
Summary: The gimmicky 3 Bc4 offers White nothing.

Game 47
T.J.Upton-Y.Bayram
Moscow Olympiad 1994

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nf3 Nxe4 5 Nc3 Nxc3

Question: Doesn’t this game belong in Chapter Five?

 6 bxc3

Answer: Not anymore. This time White recaptures toward the centre.

Question: What are the differences?

Answer:
1. Unlike the 6 dxc3 lines, White’s 6 bxc3 doesn’t provide him with an
easy development lead and open central lines.
2. 6 bxc3 offers White a central preponderance of pawns after d2-d4 is
played.
3. White has an open b-file for a rook. This means Black will not very
tempted to castle long, which in turn means that White knows our king’s
address: the kingside, and may prepare to attack there.
4. 6 bxc3 is a bigger structural gamble for White than in the 6 dxc3 lines.
This means that if queens come off the board, White may suffer in an ending.

Question: Do you believe the 6 bxc3 line is as

dangerous to Black as the 6 dxc3 variation?

Answer: No. I think Black equalizes without too much difficulty against the 6 bxc3 line, while we
must learn a huge number of lines against 6 dxc3.
6 ... Be7 7 d4 0-0
 Unlike the 6 dxc3 lines where we may keep White guessing as to which
side we castle, here our choice is made clear. We must go kingside, since
castling long would be inadvisable, due to White’s open b-file.
8 Bd3 Bf6
Black clears the e-file, and keeps open a future possibility of a manual
fianchetto with ... g7-g6 and ... Bg7 later on. Black has no problems after 8 ...
Bg4 9 0-0 Nd7 10 c4 c5 11 c3 Rc8 12 Bf4, T.Upton-M.Condie, Oxford 1984.
With 12 ... Nb6, White must keep eternal watch over his weak c4-pawn.
Black looks at least even here.
9 0-0 Bg4 10 Rb1 b6
Black is forced to weaken his queenside light squares.
11 Re1 Nc6
The knight can either be transferred to the kingside via the e7-square, or
play to the a5-square, followed by the ... c7-c5 pawn break.
12 Bf4 Re8 13 Re3 Ne7
The rook trade 13 ... Rxe3 follows the principle: Swaps help the cramped
side. But he probably avoided it because he didn’t want to allow the opening
of the f-file after 14 fxe3 Ne7 15 e4 Ng6 16 Bg3 Qe7 with a solid but passive
position.
14 h3 Nd5
This forces rooks off the board. 14 ... Bh5 intending a future ... Bg6 is also
about even.
15 Rxe8+ Qxe8 16 Bd2 Bd7!
 An alert move. Black avoids the trap 16 ... Bh5?? 17 g4 Bg6 18 c4 Ne7 19
g5 trapping the bishop.
17 Qf1!
Planning Re1, followed by Be4.
17 ... g6 18 Re1 Qf8 19 Be4 c6 20 c4
Infiltration with 20 Qa6 doesn’t lead anywhere for White. After 20 ... Nc7
21 Qb7 Qd8! and if 22 Bxc6?! Rb8, White is forced to give up his queen for
rook, bishop and pawn after 23 Qxb8 Qxb8 24 Bxd7 Qb7. White may have
enough material for the queen, but Black has all the winning chances, since
the queen has access to many white pawn targets in the future.
20 ... Nc7 21 d5!?
White grabs some central space but at the cost of fixing his c4-pawn as a
future target.
21 ... c5
I like this better than 21 ... cxd5 22 cxd5 which gives White’s pieces
access to the d4-square.
22 Qd3 Re8 23 g4!?
White, who clearly harbours kingside ambitions, fumes with impatience
and believes now is the time to implement his kingside push. However, after
23 Qa3 a6 24 Qb3 b5 25 Ba5 Bd8 26 Bd3 Rxe1+ 27 Bxe1 Qe8, Black stands
no worse anyway.
 23 ... Bg7 24 Nh4!?

Clearing the way for his f-pawn to move forward. When we prepare to
attack a well defended, weakness-free opponent, we are like one spider
attempting to trap a neighbourhood spider rival in our web. Odds are, it’s not
going to work since the rival spider is well aware of our plans.
24 ... Re7 25 Qg3
This may be inaccurate. 25 Ng2 is safer and probably better.
25 ... Be5
I prefer Black’s game after 25 ... Qe8! 26 f3 Be5 27 Bf4 b5.
26 f4
White continues to usher his kingside pawns forward. The attackers are
only half there, just partially formed shapes of demons seeking to enter
Black’s realm of flesh-and-blood.
26 ... Bd4+ 27 Kh1 Qe8 28 Qg2??
White stands only slightly worse after 28 Bd3 b5.
 Exercise (combination alert): We witness the classic overpress.
Excess ambition impelled White into taking an appalling risk to
continue his attack. With his last move, White volunteered himself
into a deadly pin on the e-file. How did Black refute the idea?

Answer: Pin/Interference.
28 ... Bf6!
The bishop reminds us of that Twilight Zone episode where the
ventriloquist’s dummy attains sentience and takes control over the act. This
move probes his opponent’s vulnerabilities past the point of White’s
salvation. Now White’s would-be attack vanishes. This move forces White to
weaken f5, which in turn allows ... Bf5.
Answer#2: Also strong is 28 ... Ba4 29 Bc1 Bf6 30 g5 Bc3 31 Re2 b5 32 Bd3 bxc4 33 Bxc4 Rxe2
34 Qxe2 Qxe2 35 Bxe2 Nxd5 36 Bd1 Nb4 and White can resign.
29 g5
When the damage has already been done and it’s too late to take remedial
action, what else is there but to watch the enemy with a wary eye and be
ready when the inevitable strike occurs. This move is forced, but the trouble
is it opens the f5-square for a black pawn or bishop, which increases the
pressure on the pinned e4-bishop to unbearable levels.
29 ... Bd4 30 Kh2 Bf5
 Winning material. All of White’s joy was earmarked for his attack, and
with its absence, so goes his happiness. Even stronger was 30 ... f5! 31 gxf6
Bxf6 when White’s knight was unable to retreat. Nevertheless, after 32 Nxg6
hxg6 33 Bxg6 Qf8, White lacks a good discovery and doesn’t have
compensation for the piece.
31 Bxf5 Rxe1 32 Bd7 Bg1+! 33 Kg3 Qe2
“A beautiful evening gown and sparkling jewels cannot disguise and undo
nature’s handiwork,” Black’s beautiful queen rudely tells her homely g2-
sister.
34 Bxe1 Qxe1+ 35 Kg4 0-1

There is no point continuing with 35…Bf2 36 Nf3 Qe2 37 a4 Qxc2,

leaving White a pawn down with a hopeless position.
Summary: The 6 bxc3 line is not as dangerous as its Chapter Five cousin, 6 dxc3.

Game 48
M.Carlsen-V.Kramnik
Amber Rapid, Nice 2008

1 e4 e5 2 Nf3 Nf6 3 Nxe5 d6 4 Nc4

 Question: What is the point of playing the knight to the c4-square?

Answer: White plans to later transfer it to the e3-square, where it watches over the f5-square,
perhaps in conjunction with attacking ideas like Qf3.

Question: Is it effective?

Answer: I think Kramnik’s plan essentially neutralizes the idea by depriving White of the f5-square.
Still, it must be taken seriously since it has been tried on players of the calibre of Carlsen and
Morozevich.
4 ... Nxe4 5 Nc3 Nxc3 6 bxc3
White’s main move. 6 dxc3 gets White into a version of Chapter Five,
where he wasted time with his knight. The following game is a good example
of how to play Black: 6 ... Be7 7 Bf4 Bg5! Black either gets a swap, or
transfers his bishop to a better diagonal. 8 Qd2 Bxf4 9 Qxf4 0-0 10 0-0-0
Nd7 11 Ne3 Qf6 12 Qg3 Nc5 13 Rd4 Re8 14 Rf4 Qg6 15 Qxg6 and a draw
was agreed. If anyone is better, it is Black, A.Sokolov-J.Degraeve, Aix les
Bains 2007.
6 ... g6!

Question: Why an exclamation mark?

 Answer: By fianchettoing on the kingside, Black gets the following:
1. He makes his king safer.
2. White’s entire plan revolves around his playing Ne3 with the idea of
Nf5. By playing ... g6, Black deprives White of his intent.
7 d4 Bg7 8 Bd3 0-0 9 0-0 Nd7
Kramnik prepares to eject White’s knight with ... Nb6, while keeping his
c-pawn’s options open for a ... c5-break, which challenges White’s central
domination. Also possible is 9 ... Nc6 10 Rb1 Rb8 11 Qf3 Bd7 12 Ne3 b5! 13
Nd5 b4! (this plan effectively breaks up White’s central pawns) 14 Nxb4
Nxb4 15 cxb4 Bxd4 with an equal position, D.Kokarev-A.Gubajdullin,A
Togliatti 2007.
10 Qf3 Nb6 11 Ne3 c5!

Kramnik effectively chips away at White’s centre.

12 dxc5 dxc5
The c3-pawn is now chronically weak.
13 a4!?
A move made with the philosophy: When you have money, it’s better to
spend it or give it away, because as the old saying goes: You can’t take it with
you. Now the queenside begins to snarl with aggressive intent. A foe who
doesn’t fear death is to his enemy, the most dangerous person in the world.
Carlsen decides he has had enough of this interminable adagio pace and
decides to step it up a notch. He buffs away the cosmetic flaws in his
structure by giving pawns away. This is in effect a two pawn offer for
initiative.
13 ... Bd7!?
The pool of distrust continues to rise between the two armies, both seeking
to strike first. Kramnik isn’t a pawn grabber and prefers initiative to material
gain. He probably feared White’s initiative after 13 ... Bxc3! 14 Rb1 Nxa4
(the invasion of the pawn snatchers!) 15 Nd5 Bg7 16 Bf4 with a scary
looking position. If we analyze a bit further, we see that Black neutralizes
White’s initiative effectively with the forcing swap 16 ... Nc3! 17 Nxc3 Bxc3
18 Be4 Qf6 19 Bxb7 Bxb7 20 Rxb7 Bd4 and White’s pawn-down position
sags in debility.
14 a5 Na4 15 Qxb7 Nxc3 16 Bd2 Be6
Threat: ... c5-c4.
17 Bxc3
A clear concession in the open position, but that c3-knight was just too
strong.
17 ... Bxc3 18 Ra4 Qd7!

Kramnik forces an ending where only he has the winning chances. He had
the bad luck to face the world’s premier endgame specialist defending with
the white pieces.
 19 Qxd7 Bxd7 20 Ra3 Bb4 21 Raa1 Bc6 22 a6 Rad8 23 Rfc1
Threatening the cheapo c2-c3 which will trap the dark-squared bishop.
23 ... Bc3 24 Rab1 Rfe8 25 Bb5 Re6 26 Bc4

Question: Doesn’t it follow principle to swap

with Bxc6, eliminating Black’s bishop pair?

Answer: Theoretically, yes. However, in this case there is a side issue: with the swap of bishops,
White’s a6-pawn is chronically weakened. Black would eventually win a pawn after 26 Bxc6 Rxc6 27
Rb5 Rd2 28 Nc4 Re2 29 Kf1 Re4 30 Ne3 Rxa6 31 Rxc5 Bd2 32 Rb1 Bxe3 33 fxe3 Rxe3 with an extra
pawn and winning chances.
26 ... Re4
A slightly inaccurate move which Carlsen soon exploits. Better is 26 ...
Red6! when Black has good chances to convert with his bishop pair.
27 Rd1 Rxd1+ 28 Rxd1 Bd4 29 Bd5!
Now White gets away with the swap.
29 ... Bxd5 30 Nxd5 Re2 31 c3?!
White’s follow-up is inaccurate and now he must work hard to secure the
draw. Apparently even Carlsen’s almost supernatural defensive/anticipatory
abilities don’t work 100% of the time. 31 Ne3! should hold the position.
After 31 ... Bxe3 32 fxe3 Rxe3 33 Rd7 Rc3 34 Rxa7 Rxc2 35 Rb7 Ra2 36 a7
Kg7 37 Rc7 c4 38 Rxc4 Rxa7, a theoretically drawn three versus two rook
and pawn ending is reached.
31 ... Bxf2+ 32 Kf1 Ra2 33 Rb1 Bh4!?
Now White has excellent drawing chances. Black may have better chances
to win by returning his extra pawn with 33 ... c4! 34 Rb4 Bc5 35 Rxc4 Ra1+
36 Ke2 Bg1 37 h4 Rxa6.
Black stands better but I believe White’s chances to hold the game are
higher than Black’s to win.
34 g3 Bg5 35 h4 Bd2 36 Nf6+! Kg7 37 Ne4 Bh6 38 Rb7 Rxa6 39 Nxc5
 Black remains up a pawn with chances to convert.
39 ... Rc6 40 Ne4 Re6
Stronger is 40 ... Be3! to hang on to the passed a-pawn. This gives Black
better chances to win.
41 Nc5 Re5 42 Nd7 Re3
A double attack on the g3- and c3-pawns. The trouble with this plan is
Kramnik allows his passed a-pawn to fall, and with its loss, Black’s winning
chances decline steeply.
43 Rxa7
With the removal of this passed pawn, a great weight has been lifted from
Black’s shoulders.
43 ... Rxg3 44 Ne5 Kf6
After 44 ... Re3 45 Rxf7+ (45 Nxf7?? walks into a trap, which is sprung
with 45 ... Rf3+, nabbing the knight) 45 ... Kg8 46 Re7 Bf8 47 Re8 Kg7 48
Re6 Rxc3 49 Ke2, White should draw, now that all the remaining pawns are
on the same side of the board.
45 Nxf7 Be3 46 Rb7 h5
Fixing the h4-pawn has a permanent target, though weakening the g6-
square in the process. The latter will play a significant role in the game’s
outcome.
47 Ke2 Bf4 48 c4 Re3+ 49 Kf1 Rc3
 The rook’s plan is clear: intercept and destroy.

Exercise (planning): Snowfall may be a miracle for someone who lives in the desert and an
ordinary occurrence for another person living in a Northern climate. Moral: One person’s miracle is
another’s everyday occurrence. Carlsen, who to my mind is currently the best defensive and endgame
player in the world, routinely saves a wretched position against the world’s best. It appears as if White
is lost. Innocent targets tend to beckon the wicked. After all, the c4-pawn is falling and the h4-pawn
looks sickly as well. How did Carlsen force a drawn position?

Answer: Target Black’s only weak point: the g6-pawn and arrange a wrong-coloured remaining
bishop ending.
50 Nh8!
Threat: Rb6+ and Nxg6. A hidden strength – unnoticeable to the eyes at
the surface-level glance – enhances its power, since a concealed weapon is so
much more dangerous than one wielded openly by the enemy. Carlsen’s
apparent negligence is proven illusory, as he divines the source of his
troubles and makes the appropriate repairs.
50 ... Rxc4
Miraculously, there is nothing to be done about the coming Rb6+. For
example:
a) 50 ... Be3 51 Rf7+ Ke6 52 Rf3 and the g6-pawn falls all the same.
b) 50 ... Rg3 51 Rb6+ Kf5 52 Nxg6! Rxg6 53 Rxg6 Kxg6 is drawn, since
Black is left with the wrong-coloured remaining bishop for his h-pawn.
 51 Rb6+ Kf5 52 Nxg6
The plan’s thread is finally unspooled. Black has the advantage on paper,
but the game is clearly drawn with correct play from White.
52 ... Bg3 53 Kg2 Kg4 54 Rb2!
Covering his second rank.
54 ... Re4 55 Ra2 Bd6 56 Ra5!
Threat: Rg5+ and Rxh5.
56 ... Re2+ 57 Kf1 Re6 58 Rg5+ Kf3 59 Rf5+ Kg4 60 Rg5+ Kh3 61 Kf2
Rf6+ 62 Ke2
Most definitely not 62 Kg1?? which would be a mega blunder after 62 ...
Bh2+. “It is the guilty who fear exposure to the light,” the bishop declares to
White’s cringing king. Both are on their knees for different reasons: White’s
king, because he is mortally wounded; the bishop in gratitude, giving thanks
to the Almighty for fulfilling his wishes. White gets mated after 63 Kh1 Rf1+
64 Rg1 Rxg1.
62 ... Bg3
Black can win a pawn, but not the game.
63 Ke3 Re6+
Also, 63 ... Bxh4 64 Nxh4 Kxh4 65 Rg1 is a drawn rook and pawn ending.
64 Kf3 Rf6+ 65 Ke3 Re6+ 66 Kf3 ½-½
The players decide to call it a day after a hard day’s labours.
Summary: Kramnik’s early ... g7-g6! idea seems to effectively neutralize the 4 Nc4 variation.
Index of Complete Games
Anand.V-Kramnik.V, Corus A, Wijk aan Zee 2010
Anand.V-Kramnik.V, Corus A, Wijk aan Zee 2008
Baramidze.D-Akobian.V, World Team Championship, Antalya 2013
Bruzon.L-Ivanchuk.V, Carlos Torre Memorial, Merida 2007
Camano.C-Ginzburg.M, Argentine Championship, Buenos Aires 1994
Carlsen.M-Kramnik.V, Amber Rapid, Nice 2008
Caruana.F-Wang Hao, Kings Tournament, Bucharest 2013
Deepan.C-Koneru.H, Indian Championship, Visakhapatnam 2006
Dvoirys.S-Mamedyarov.S, Aeroflot Open, Moscow 2004
Gashimov.V-Gelfand.B, FIDE Grand Prix, Astrakhan 2010
Gashimov.V-Kramnik.V, Azerbaijan vs. the World match, Baku 2009
Giaccio.A-Zarnicki.P, Buenos Aires Zonal 2000
Guseinov.G-Kramnik.V, Azerbaijan vs. the World match, Baku 2009
Inarkiev.E-Kasimdzhanov.R, FIDE Grand Prix, Jermuk 2009
Ivanchuk.V-Wang Yue, MTel Masters, Sofia 2009
Jakovenko.D-Wang Yue, Pearl Spring Chess Tournament, Nanjing 2009
Kamsky.G-Karpov.A, World Championship (Game 6), Elista 1996
Karjakin.S-Mamedyarov.S, FIDE Grand Prix, Beijing 2013
Kasparov.G-Karpov.A, X3D Match (Game 2), New York (rapid) 2002
Leko.P-Gelfand.B, Tal Memorial, Moscow 2009
Leko.P-Kramnik.V, World Championship (Game 1), Brissago 2004
Macieja.B-Rozentalis.E, US Open Championship, Reno 1999
Malakhov.V-Frolyanov.D, Russia rapid championship, Olginka 2011
Maslowski.J-Sarwinski.M, Bydgoszcz Open 2001
Miltner.A-Werle.J, German League 2004
Morozevich.A-Wang Yue, Amber Blindfold, Nice (rapid) 2009
Naiditsch.A-Kramnik.V, Dresden Olympiad 2008
Naiditsch.A-Kramnik.V, Dortmund 2008
Navara.D-Kramnik.V, CEZ Trophy match, Prague (rapid) 2008
Negi.P-Li Chao, Casino Graz Open, Austria 2014
Ni Hua-Sandipan.C, Fujairah Masters, United Arab Emirates 2012
Nielsen.M-Sargissian.G, Politiken Cup, Helsingor 2009
Polgar.J-Ivanchuk.V, Amber blindfold, Monte Carlo (rapid) 1994
Ponomariov.R-Wang Hao, Kings Tournament, Bucharest 2013
Reinderman.D-Bosboom.M, Groningen 1995
Rublevsky.S-Shirov.A, Russian League 2006
Samolins.V-Jakovenko.D, Istanbul Olympiad 2012
Shirov.A-Adams.M, European Team Championship, Leon 2001
Shirov.A-Karpov.A, Corsica Masters, Bastia (rapid) 2002
Short.N-Shirov.A, FIDE Grand Prix, Dubai 2002
Smeets.J-Gelfand.B, Amber rapid, Nice 2010
So.W-Giri.A, Corus B, Wijk aan Zee 2010
Sutovsky.E-Sakaev.K, Serbian League, Kragujevac 2009
Svidler.P-Anand.V, Linares-Morelia 2007
Topalov.V-Gelfand.B, Corus A, Wijk aan Zee 2008
Topalov.V-Kramnik.V, Linares 1999
Upton.T-Bayram.Y, Moscow Olympiad 1994
Volokitin.A-Dominguez Perez.L, Bermuda 2005