There’s No Such Thing as Enough Pitching

Aside from the Carlos Correa free agency saga, the most consistent rumor theme this offseason has been the Twins’ purported interest in acquiring starting pitchers that would seemingly fit in the middle or back end of the rotation.

That has caused more than a few followers of the team to scratch their heads because, generally speaking, the consensus across the baseball blogosphere is that the Twins have solid starting pitching depth, with numerous homegrown fourth and fifth starter types. What they seem to lack is a playoff game one caliber ace (or two) that would be reasonably expected to be better than Sonny Gray or Joe Ryan (or a fully healthy and conditioned Kenta Maeda).

In Dan Szymborski’s 2023 ZiPS projections of the Twins at FanGraphs, he wrote:

The Twins’ pitching projections are far from actually being poor, but there’s just such a blandness about it that it’s hard to conjure up much gusto. Joe Ryan, Sonny Gray, and Tyler Mahle are all serviceable more than scintillating, and the team has a bevy of perfectly reasonable fourth and fifth-starter options available. This group would look a lot better if it had a bonafide ace at the helm…

Mahle, Gray, and Ryan all project between 2 and 3 WAR by ZiPS. Maeda and Bailey Ober are both between 1 and 2 wins. Josh Winder, Louie Varland, Cole Sands, Ronny Henríquez, and Simeon Woods-Richardson are also on the 40-man roster and poised to provide depth to deal with injuries and the inevitable need for spot starts for doubleheaders. Prospects Jordan Balazovic and Brent Headrick, who haven’t debuted in the majors yet, are also there.

Photo by Omar Rawlings/Getty Images

And yet, there have been persistent rumors going back to last summer about a trade deal for Miami’s Pablo López, a decidedly mid-rotation option. There have been more recent rumors of interest in right-hander Michael Wacha, a decidedly back-of-the-rotation or depth option. Moreover, the Twins did not appear to make any serious pursuits of front-of-a-rotation pitchers, despite the availability of a few in free agency.

Given all that, the popular reaction to the rumblings about signing Wacha or trading for López has been… “why?”... often followed quickly with concerns about how making such a move would block one of the younger arms from getting an extended opportunity or push another pitcher to a less valuable bullpen role.

There Is No Such Thing As Enough Pitching

Those are understandable perspectives, especially in the offseason when we’re pretty much only able to look at the roster on paper. The on-paper situation can sometimes be misleading because it is full of assumptions about player health and expected workloads.

Yes, right now, the roster comfortably runs eight deep with starting pitchers that we’d generally feel good about taking the ball on a given day.

The problem is, the data from recent history suggests that’s probably not nearly enough.

Data from FanGraphs

Last season the average MLB team used 12.9 starting pitchers across their 162 games. Colorado and Houston used the fewest different starters (8), Tampa Bay used the most (18), and the league median was 12. The average team required 37.9 starts from pitchers that weren’t among their top five most commonly used starters and those depth arms worked an average of 167 innings at the team level.

There is some noise in the above data because of the increased use of openers and bullpen games, but these numbers track with what FanGraphs’ Jeff Sullivan found back in 2014.

Even if those counts are a bit overstated because of the modern starter usage patterns, it’s still the case that teams, on average, only get about four rotation spots worth of work from their top five most-used rotation members (which is likely a different group of names than their top five on-paper starting rotation going into the season). That means that something north of 10% of the season’s total innings have to be absorbed by pitchers ostensibly not part of the primary plan.

While that data does not explain away the Twins' reluctance to seriously pursue any front-line starting pitchers, it does help to explain why they might hold interest in additional, higher-quality starting pitching depth.

In a close playoff race, like the Twins are expected to be in next season, the quality of those marginal innings can be a significant factor in making the tournament or not. Last season’s twelve playoff teams got 4.68 ERA work from their “depth starters” – defined as their #6 most commonly started pitchers and below – while the depth guys for teams that did not make the playoffs worked to a 5.28 ERA.

Forecasting Pitching Injury Risk

Predicting injuries, especially when a player might get injured and how seriously, is not something that we’re able to do with any real repeatable success. That hasn’t stopped people from trying, however, and their past work is informative for understanding where teams and players might have increased exposure.

Injury research published at FanGraphs, mostly by Jeff Zimmerman, has shown repeatedly that the greatest predictor of future pitcher injury is past injury, especially within the past two seasons. The next best indicator is fastball velocity. We learned that more velocity tends to lead to more pitcher injuries than lower velocity when I wrote about Jhoan Duran and the max velocity trend last month. Velocity has shown to be even more impactful for injury than pitch type, despite the stubbornly persistent thinking that breaking pitches cause more arm injuries.

Jeff Zimmerman | FanGraphs | Accessed From: LINK

You can see in the table above, using those two variables, Zimmerman found before the 2020 season that a starter without any prior injury history and an average fastball below 93 MPH has about a one in four chance (28%) of landing on the injured list, with an average stay of 17 days. If his fastball averages more than 93 MPH, the chances of an IL trip increase to 45%, and the average length of stay goes up to 28 days.

If the pitcher has an injury history, represented by prior trips to the injured list, the chances of a future injury were 49% in the low-velocity bucket, with an average length of 31 days, and 57% and an average length of 34 days in the higher velocity bucket.

With that in mind, we can quickly assess the Twins' current starting options by those two primary variables:

Based on some of Zimmerman’s earlier work, Eno Sarris wrote in 2014 that an average team faced a 65% likelihood that at least 2 starters will be hurt at any given point in the season, a 32% likelihood that 3+ starters will be hurt and a 10% likelihood that 4+ starters will be hurt at the same time.

Given the injury data of the Twins’ current starting candidates above, it’s probably reasonable to think they might be somewhat more likely than those averages. Even in this era of diminished starters’ workloads, every team had at least 750 innings thrown by starting pitchers last season, making that a minimum target for what the Twins need to have covered in 2023.

It’s probably not a great way to look at it, but for comparison’s sake, in 2021 it would have taken the actual workloads (including minor league innings) of the first seven names (through Winder) in the table to reach that benchmark. Last year, it took the first nine (through Henríquez) to crack 750 innings.

Chris Paddack is likely out for 2023 recovering from elbow surgery. Maeda is returning from his own elbow injury and probably should not be counted on for a full season’s workload. Gray and Ober seem bankable for a couple of IL trips each. Mahle had a somewhat odd shoulder issue that ended last season early, despite MRIs showing no structural issues. Winder has dealt with recurring shoulder issues dating back to his time in the minors.

It is not hard to look at the Twins’ setup here and see a scenario where they are counting on Varland and Woods-Richardson to regularly take the ball. That’s not to knock on those well-regarded prospects. They may well prove perfectly capable of taking hold of those opportunities and running with them.

But, that’s probably not the most likely outcome. The Twins need only look within the AL Central to Kansas City and Detroit for recent cautionary tales about how quickly well-regarded pitching prospects (in those cases, prospects that were even more highly-regarded than the Twins’ current bunch) transition into consistently effective big-league starters.

Summing Up The Point

I want to be clear that none of this is intended as an argument that making the trade for López or signing Wacha (or similar) is the best course of action. They both have their own notable injury histories, with time missed due to shoulder injuries being a common connection, and there are opportunity costs that have to be considered.

The point I’m trying to suggest is that such “depth” moves should not be dismissed out of hand because the pitchers in question don’t surpass Gray or Ryan on the depth chart or because the Twins have other good internal options. Those things should not be mutually exclusive.

That’s especially the case because of where the Twins are in the competitive landscape. As a projected 85-ish-win team, the Twins are squarely on the part of the win curve where each marginal addition is incredibly valuable because of how it could affect making the playoffs or not. Even just avoiding having a handful of games that have to be started by mid-season, replacement-level waiver claims could make the difference.

If nothing else, all we have to do is refer back to the last two seasons for examples of how not having enough quality pitching depth can sink a season.

---

John is a writer for Twinkie Town and Pitcher List with an emphasis on analysis. He is a lifelong Twins fan and former college pitcher. You can follow him on Twitter @JohnFoley_21.