Police Practice and Research

An International Journal

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/gppr20

A scientific examination of the 21-foot rule

William L. Sandel , M. Hunter Martaindale & J. Pete Blair

To cite this article: William L. Sandel , M. Hunter Martaindale & J. Pete Blair (2020):
A scientific examination of the 21-foot rule, Police Practice and Research, DOI:
10.1080/15614263.2020.1772785

To link to this article: https://doi.org/10.1080/15614263.2020.1772785

Published online: 31 May 2020.

Submit your article to this journal

Article views: 78

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=gppr20
POLICE PRACTICE AND RESEARCH
https://doi.org/10.1080/15614263.2020.1772785

ARTICLE

A scientific examination of the 21-foot rule

a b
William L. Sandel , M. Hunter Martaindale and J. Pete Blairb
a
Department of Criminology and Criminal Justice, Missouri State University, Springfield, MO, USA; bALERRT, Texas
State University, San Marcos, TX, USA

ABSTRACT ARTICLE HISTORY

The purpose of this study was to scientifically assess the long-standing 21- Received 13 February 2020
foot rule. There are several anecdotal publications looking at the 21-foot Accepted 4 May 2020
rule as a standard in policing. This study uses experimental design to KEYWORDS
examine whether this standard should continue in modern-day policing. 21-foot rule; safe distance; 21
The 21-foot rule was tested in three independent experimental design feet; Tueller drill; police
studies. The first study measured the average speed at which a person training
could run 21 feet. The second and third studies tested the speed at which
an officer could draw and fire their weapon with no stress and under stress
respectively. The final study examined methods for increasing survivabil­
ity for the officer (movement). The findings show the 21-foot rule to be an
inadequate standard for officers to safely draw and fire their weapons
when being charged by a suspect who’s intent it to cause harm.
Additionally, different strategies of moving can increase the officer’s
ability to survive.

Introduction
Officer involved shootings have been an issue of great concern to the public. We have seen riots and
protests (such as those in Ferguson, Missouri; Sacramento, California; Chicago, Illinois) in response
to shootings that the public (or at least some segments of the public) thought were unjustified (Shin,
2017; Simon et al., 2018; Sobol et al., 2018). This is not only a problem in the United States. Officers
from around the world also experience use of force encounters involving a suspect with an edged
weapon (Associated Press, 2008). Pundits often take to the 24-hour news channels to pronounce
whether or not a particular shooting was justified, often before the facts are known, and sometimes
seemingly with a desire to do nothing but incite the community. Other media outlets publish story
after story with their take on whether or not the officer(s) involved should have fired. High profile
shootings have also led at least one state, California, to enact legislation that would increase the legal
threshold necessary for officers to use deadly force (CA AB-392, 2019).
At the same time, police officers are also greatly concerned about deadly force encounters. Police
departments place a high priority on protecting their community. They want deadly force to be used
as infrequently as possible and only when necessary to protect lives. Not only can the use of deadly
force negatively impact public perceptions of the police, but many officers also experience sig­
nificant mental trauma and leave policing after they are involved in shootings (Follette et al., 1994).
Police departments also place a premium on the safety of their officers. Policing is a dangerous
profession where suspects have been known to assault police officers and sometimes these assaults
are potentially deadly to officers. Departments seek to provide their officers with the equipment,
training, and skills to successfully avoid using deadly force where possible, but also to effectively use

CONTACT William L. Sandel [email protected] 901 S. National Ave., Springfield, MO 65897, USA
© 2020 Informa UK Limited, trading as Taylor & Francis Group
2 W. L. SANDEL ET AL.

deadly force when it is necessary to protect themselves or others. The following story demonstrates
such an encounter.
On 17 December 2018 a man armed with a knife and screwdriver began chasing people at a gas
station in Oklahoma City, Oklahoma. Upon arriving at the scene, officers commanded the suspect
to drop the weapons. The suspect then charged the officers and the officers used a taser and beanbag
gun in an attempt to stop the suspect’s assault. The less-lethal force had no effect on the suspect,
who continued to move aggressively toward the officers. The officers then used their handguns,
which resulted in the death of the suspect (KOKO, 2018). Was this reasonable?
While there has been substantial research into police use of force in general, and shootings in
particular, there has been relatively little research into the human performance factors involved in
police shootings. Little is known about how quickly and effectively an officer can respond to different
types of threats; yet, this is exactly the type of research that should inform public, court, and police
perceptions of use of force events. Better understanding of these factors can help protect the police by
helping them to understand and avoid dangerous situations. The public is also protected when police
have a clear understanding of the dynamics of these situations and their actual response capabilities,
resulting in better decisions about when force should be utilized and when it should not.
This paper seeks to fill some of this knowledge gap by examining the ‘21-foot rule’ that is taught
across the country to police officers as either a) the minimum distance that a suspect can cross and
attack an officer with a knife before an officer can draw and fire his or her weapon, or b) a safe
distance to deal with potentially dangerous suspects armed with knives. We report on three studies
that explore how fast a person can cross a 21-foot distance, how quickly and accurately officers can
draw and fire their weapons, and some defensive tactics that officers can use to further protect
themselves in potentially deadly situations. We turn now to our literature review. We begin with an
overview of use of force by and against police, then turn to reasonableness, followed by research on
the 21-foot rule, and finally end with a discussion of reaction time.

Literature review
Use of force
By officers
It has been well established that officers rarely use force in their daily job duties. A report by the U.S.
Department of Justice stated that of an estimated 40 million instances of police/citizen contact, only
around 1.4% involved use of force (Eith & Durose, 2011). It is difficult to say what percent of those
encounters involve the use of a firearm by the officer, but federal records show a total of 435
justifiable homicides by law enforcement officers in 2016 (Federal Bureau of Investigation, 2017).
While police shootings are rare, as we discussed above, they can have a substantial impact on the
suspect, community, and the officers. It is, therefore, critical that police training for these situations
be validated in as thorough a manner as possible.

Against officers
According to the National Law Enforcement Officers Memorial Fund (2017, 2018)), there were 33
officers killed in the line of duty between 2007 and 2017 as a result of close-quarters physical
encounters. Of these 33 officers, 13 were stabbed, 16 were beaten, and four were strangled. Though
this may seem like a relatively low number of deaths, violent close encounters occur much more
frequently. The Law Enforcement Officers Killed and Assaulted (LEOKA) data from 2007–2016
provided by the Federal Bureau of Investigation (FBI) reports the number of officers assaulted while
on duty. It was found that 9,586 officers were assaulted with an edged weapon (Law Enforcement
Officers Killed and Assaulted (2018). Of those 9,586 officers, about 1,200 were injured as a result of
the encounter. When considering physical assaults with no weapon, 444,741 officers were assaulted
and nearly a third were injured (Law Enforcement Officers Killed and Assaulted (2018).
 POLICE PRACTICE AND RESEARCH 3

Use of force and reasonableness

An officer’s decision to use force, including lethal force, is constrained by the legal standard of
objective reasonableness. The Supreme Court of the United States held in Tennessee v. Garner
(1985) that a police shooting was considered a seizure under the Fourth Amendment and must be
‘reasonable.’ The Court later added specificity to the standard of reasonableness in Graham
v. Connor (1989). The Court held that any force used by law enforcement must be ‘objectively
reasonable’ based on the totality of the circumstances and the facts known to him or her at the time.
Additionally, the Court opined that this standard of objective reasonableness should be judged
‘from the perspective of a reasonable officer on the scene, rather than with the 20/20 vision of
hindsight,’ (Graham v. Connor, 1989, p. 396). There have been several other court decisions that
have involved the standard of objective reasonableness (see for example, Johnson v. Glick, 1973;
Illinois v. Gates, 1983; Sharrar v. Felsing, 1997), but the underlying logic of Graham is the law of the
land.
Under Graham, and other court rulings, an officer’s actions can be deemed reasonable if the
actions in the moment, with the same facts available to the officer, would be considered reasonable
by another officer. This standard of objective reasonableness is used by law enforcement agencies to
establish use of force policy (Terrill, 2009). For example, it is reasonable for an officer to utilize
deadly force if the suspect is attempting to kill or seriously injure the officer or another civilian.
However, deadly force would not be considered proportional, and therefore would not be reason­
able, if the suspect is verbally assaulting the officer.
Some scholarship informs courts about the reasonableness of police shootings. For example,
a line of research on perceptual distortions has found that police commonly experience tunnel
vision, auditory blunting, and time dilation both immediately before and during shootings
(Campbell, 1992; Klinger & Brunson, 2009; Solomon & Horn, 1986). This information can be
important both for a court trying to understand an officer’s account of a shooting and for officers to
better understand what they are likely to experience during a shooting.
When one is attempting to determine objective reasonableness of a use of force incident, it is
imperative to take into consideration what the officer was experiencing at the moment of the
incident. Klinger and Brunson (2009) established a snowball sample of 80 officers (113 total
incidents) that had shot citizens in the line of duty. They found varying levels of auditory blunting
in 82% of cases. Fifty-one percent reported tunnel vision, while 56% reported heightened visual
acuity. Slow-motion time was experienced by 56% of the officers and fast-motion time occurred in
23% of the cases. Of the officers experiencing some form of perceptual distortion, 87% reported the
perceptual distortion before shooting and 92% reported experiencing distortions during the
shooting.
These distortions may have a profound impact on how the officer perceives the behaviors of
a suspect due to the amount of stress an officer is under when deciding to use force. Keinan (1987)
found that people made decisions before considering all available data when a stressor was present
during decision-making. Therefore, understanding these distortions is relevant to assessing the
objective reasonableness of police shootings. In the next section, we consider the 21-foot rule.

The 21-foot rule

Lieutenant John Tueller laid the framework for the 21-foot rule in the 1980 s when conducting
firearms training with the Salt Lake City Police Department. Lieutenant Tueller had one officer play
the role of a suspect with an edged weapon while another officer stood about 21 feet away with
a holstered training weapon. The ‘suspect’ was instructed to charge the officer in attack mode to
simulate a real-world attack scenario and the officer was instructed to draw and fire before the
suspect reached him or her (Martinelli, 2014). This exercise was designed as a reactionary gap drill
to teach officers how quickly these encounters unfold. After running this drill for several years, Lt.
4 W. L. SANDEL ET AL.

Tueller found that the average person could run 21 feet in approximately 1.5 seconds (Tueller,
1983). Additionally, Tueller found officers were able to draw and fire a holstered sidearm in
approximately 1.5 seconds (Tueller, 1983). Lieutenant Tueller published these findings in SWAT
Magazine. He considered 21 feet to be the ‘danger zone’ for a police officer (i.e., an officer will be
unable to defend him/herself if a suspect charges from 21 feet or closer).
However, the information in this article was interpreted by some agencies as 21 feet being a ‘safe
distance’ for police encounters involving a charging assailant (i.e., the officer will be able to defend
him or herself if a suspect charges from 21 feet). Twenty-one feet as a safe distance became even
more prominent with the distribution of a training video called Surviving Edged Weapons (Smith,
2018). This training video emphatically stated that 21 feet was far enough for officers to be able to
(1) draw their weapon, (2) aim and fire two shots at center mass, and (3) move away from the
suspect after firing. Across the country, there are departments that base their training and use of
force policies (at least partially) on the 21-foot rule (Police Executive Research Forum, 2015, 2016).
This is despite the fact that peer reviewed research on the 21-foot rule is lacking. Most reported
data on the 21-foot rule are published in non-academic, trade-type publications (Adams, McTernan
& Remsberg, 2009; Blake, 2016; Force Science Research Center, 2005a, 2005b; Martinelli, 2014,
2015; Smith, 2018; Tueller, 1983). Generally, these inquiries simply rerun the original Tueller drill
with imprecise measures of speed (e.g., a smartphone stopwatch). While this does not mean that
these data are incorrect, more controlled, peer-reviewed, empirical research is needed to ensure that
both police officers and those who will judge the actions of police officers have reliable data. Next,
we consider the 21-foot rule as a specific type of reaction time experiment.

Reaction time research

The scenario laid out by the 21-foot rule is essentially a reaction time experiment. At some point
during an interaction with a police officer, a suspect decides to charge with a knife. The officer must
perceive that the attack is happening and react accordingly. There are three common types of
reaction time experiments (Luce, 1986). Simple reaction time experiments feature only one
stimulus and one response (e.g., press the button when the light turns on). Selective response
studies require the participant to respond to only some types of stimuli and ignore others (e.g., press
the button when you see a red light, but not when you see a green light). Participants in this type of
study must react to the relevant stimulus and inhibit their response to the irrelevant. Choice
reaction time experiments require the participant to give a specific response to a specific stimulus
(e.g., press the right button when you see the red light and the left button when you see the green
light). The situation of a suspect charging an officer is a choice reaction time problem for the officer.
The officer must detect the stimulus (e.g., the suspect charging the officer) and then choose
a response (e.g., shoot, move, give commands). Research has shown that reaction time increases
as the complexity of the task increases (Brebner & Welford, 1980; Luce, 1986). Reaction times are
the fastest in simple reaction time studies and slowest in choice reaction time experiments.

Law enforcement specific reaction time research

Lewinski and Hudson (2003a) found that the average reaction time for police officers to pull the
trigger of a gun in response to a light was 0.31 seconds. Three-quarters of that time (0.23 seconds)
was taken up with processing and one fourth (0.08 seconds) with the actual physical motion of
moving the finger from the resting position and firing. This finding was consistent with non-law
enforcement reaction time research, which found that reaction times to simple visual tasks were
around 0.20 to 0.30 seconds (Eckner et al., 2010; Welchman et al., 2010). In a more complex
scenario where officers had to process information from a number of lights in different rows when
making the decision to shoot, the reaction time almost doubled to 0.56 seconds (Lewinski &
Hudson, 2003b). Again, this was also consistent with general reaction time research which indicates
that complexity slows reaction time (Brebner & Welford, 1980; Luce, 1986).
 POLICE PRACTICE AND RESEARCH 5

In a more recent study, Blair et al. (2011) studied reaction times of officers when forced to
make a deadly force decision in a dynamic scenario rather than a laboratory setting. The suspects
in the study were armed with a pistol either pointed to the ground or to their own heads. The
officer would turn and face the suspect with his or her gun drawn and issue compliance
commands. The suspect either complied or attempted to shoot the officer. The officer would
attempt to fire on the suspect once he or she recognized the suspect’s intent to fire at the officer.
Suspects fired in 0.38 seconds on average while officers fired in 0.39 seconds on average. Thus, the
process of observing a suspect’s hostile movement, interpreting the movement, deciding to take
lethal action, and acting on the decision resulted in the officer firing after the suspect even though
the officer already had his or her gun aimed at the suspect before the hostile movement (Blair
et al., 2011).
Blair and Martaindale (2014) studied law enforcement officers’ ability to shoot before suspects
while performing room entries where a student volunteer was waiting to ambush the officer. Blair
and Martaindale (2014) found that the officers fired first in only 43% of the runs. In a series of
follow-up studies, Blair and Martaindale (2017) found that deploying a simple distraction device
(e.g., tossing a chair in the room) slowed suspect reaction times while officers performed room
entries. In this experiment, officers utilizing a simple distraction device had a slight reaction time
advantage. As a whole, it appears that in at least some common law enforcement situations, police
officers are at a reaction time disadvantage to suspects, but there are techniques to alleviate this
disadvantage.
The current manuscript will assess the issues of reaction time by breaking down the 21 foot rule
into manageable research questions.

Research questions
The literature shows that law enforcement officers rarely have to use force. However, they are
constitutionally required to make objectively reasonable force decisions when force is necessary.
Some previous research shows that an officer is at a time disadvantage when reacting to the action of
a hostile suspect. However, it appears that strategies can be used to give the officer an advantage
(e.g., utilizing distraction techniques) and hopefully keep the officer safe.
The current paper seeks to evaluate the 21-foot rule by addressing four questions:

(1) How fast can a suspect run 21 feet?

(2) Is the officer able to draw and fire before the suspect can move 21 feet?
(3) How fast can an officer draw and fire his or her weapon at a charging suspect?
(4) Is there anything that an officer can do to provide more reaction time?

Methods
Study 1
Study 1 was designed to examine how quickly a suspect could cross 21 feet.

Design
Study 1 gathered baseline data to determine how fast individual participants ran 21 feet.

Sample
Study 1 consisted of students recruited from criminal justice classes at a large southwestern state
university as part of course credit. A total of 76 students participated in study 1. The mean age was
20.79 (SD = 2.8) years. Of the 76 participants, 50% were Hispanic, 34.2% were Caucasian, 13.2%
were African American, and 2.6% were Asian or other. Most of the participants were female (54%).
6 W. L. SANDEL ET AL.

Four students had only prior military experience and one participant had prior law enforcement
experience and military experience.

Method
To measure how much time is required to run 21 feet, we used purpose built laser timing gates.
Participants ran individually and were instructed to run as fast as they could down a hallway as if
they were charging a police officer with the intention of harming him or her. Simply put, the
participant would line up on the starting line. The first laser was placed at the starting line so the
initial forward movement would start an electronic timer. As the participants hit the 21-foot mark
at the end of the hallway, they would cross the second laser, which would stop the timer. This gave
us an assessment of how quickly the participants could run 21 feet.

Results
Participants ran 21 feet in an average of 1.5 seconds (SD = 0.14). The run times ranged from
1.24 seconds to 1.83 seconds.

Discussion
The results for Study 1 were similar to other studies that examined the run speed over 21 feet (Force
Science Research Center, 2005a, 2005b; Martinelli, 2014, 2015; Tueller, 1983). Individuals charging
an officer can cover a distance of 21 feet in 1.5 seconds on average. The next step was to test the
average draw speeds of the officers.

Study 2
Study 2 was designed to address part of research question 2: How quickly can an officer draw and
fire his or her weapon?

Design
Study 2 gathered baseline response data where each participant drew and fired their weapon at
a silhouette when a light was turned on.

Sample
Study two consisted of police officers who were attending classes at a law enforcement training
center and officers from a municipal police department located in a mid-sized central Texas town.
The sample included local, state, and federal officers from a number of jurisdictions throughout the
U.S. A total of 152 officers participated in study 2. The mean age was 39.02 (SD = 9.31). Most of the
officers were Caucasian (75%) and a vast majority were male (92.2%). The demographics of this
sample closely match the demographic breakdown of law enforcement in the US where 88% of
officers are male and 72% are Caucasian (Haley & Davis, 2019). The mean number of years of
experience was 13.21 (SD = 9.33) and ranged from ½ year to 40 years. Most of the sample were
patrol officers (55.9%). Only nine of the 152 police officers were on a Special Weapons and Tactics
(SWAT) team. They had a mean of 17.67 (SD = 7.89) years of experience. Additionally, the
retention level of the individual officer’s holster was reported. As this study involved the officer
drawing their weapon, the retention level of each officer’s holster was recorded. There are three
levels of retention with level one being the easiest to remove a weapon from and a level three being
the most difficult. Officers used a level two holster in study two 56.6% of the time.

Methods
This study was designed to test the draw speed and accuracy of officers in a sterile environment.
Simply, the officers were told to draw and fire a pistol in response to a simple stimulus. This
stimulus for this study was a lightbulb turning on. Before the start of the study, all officers removed
 POLICE PRACTICE AND RESEARCH 7

live weapons and were patted down for safety by the researchers. The officers were told that their
draw speed and accuracy were being researched as part of an examination of the 21-foot rule.
Officers were tested one at a time during this study. The participating officer was given a Glock 17 T
training pistol, which fires Simunition marking rounds that are filled with a mixture of paint and
detergent (Blair et al., 2011). These rounds are propelled by gun powder and break when the plastic
shell hits a hard surface. Officers were provided with a single Simunition round and told to holster
the training pistol in their actual duty holster. The retention level for each individual holster was
recorded prior to the study. Across from the officer was a silhouette target pinned to a whiteboard.
The silhouette was black to ensure the white paint rounds would be visible when the target was hit.
The officer was instructed to stand at an interview stance, which means that their hands could not
be resting on their weapon in anticipation of drawing. This was done in order to get accurate draw
speeds. The light was on a pillar down range and was connected to a switch held by the researcher
behind the officer. Once the light turned on, the officer would draw and fire the training pistol as
quickly and accurately as possible at the silhouette.
Each officer was recorded using a 30 frame per second (fps) handheld camera mounted to
a tripod. The camera was in position to capture the training pistol being drawn and pushed out
towards the target. As the silhouette was down range, the camera was also able to capture the entire
target. Hits on the target were able to be recorded during the study and the camera footage was
analyzed later to get speed data. Using video editing software allowed the researchers to move
through the videos frame by frame (30 fps) to get accurate timing from when the light turned on to
when the officer fired their one and only round. These times were calculated by individually coding
the frame in which the light turned on, the officer first touched their weapon or holster, when the
weapon cleared the holster, and when the shot was fired. Knowing that the camera was recording at
30 fps it was a matter of simple math to get the corresponding amounts of time for each event.
Twenty percent of the cases were coded by another researcher for reliability with a reliability of
100% (Intraclass Correlation Coefficient (ICC) = 1).

Results
Study two was designed to test the speed at which an officer can draw and fire their weapon under
sterile conditions. It took the officers an average of 1.80 seconds to draw and fire a single shot at
a silhouette (SD = 0.46, 95% CI = [1.73, 1.87]). During study two several officers had trouble getting
their weapons drawn causing several outliers. The median for the study two sample was 1.73 sec­
onds. The speeds ranged from 1.03 seconds to 3.4 seconds. During study two the hit percentage of
the officers was also captured. Of the 152 officers, 14% missed the silhouette altogether.

Discussion
Recall that it took an average of 1.5 seconds for someone to run 21 feet while it took 1.8 seconds for
an officer to draw and fire his or her weapon. There was a significant difference between the run
speeds and the time it takes to draw and fire a weapon (t = −5.53, p < 0.001, Cohen’s d = 0.88).
Meaning, participants were covering the distance significantly faster than officers were able to draw
and fire their weapon in Study 2. Being that these types of encounters are stressful, study three was
designed to include stress as a factor for officer draw speeds.

Study 3
Study 3 was designed to address research question 3 fully: How quickly can an officer draw and fire
his or her weapon at a charging suspect?

Design
Study 3 gathered baseline data regarding how fast each participant drew and fired his or her weapon
in response to a suspect (actor) charging them.
8 W. L. SANDEL ET AL.

Sample
The sample consisted of police officers who were attending classes at a training center and officers
from a municipal police department located in a mid-sized central Texas town. A total of 57 police
officers participated. The sample included federal, state, and local law enforcement from around the
US. The mean age was 37.95 (SD = 8.67). Most officers were Caucasian (77.2%) and male (86%). Once
again, these demographics are fairly close to national breakdown regarding gender and race. The mean
number of years of experience was 11.98 (SD = 9.26) and ranged from ½ year to 31 years. The primary
job duty for 42 of the 57 officers was patrol. This sample was smaller than the officer sample from
study 2 because there were less training classes taking place during study 3 data collection. Study 2
data collection occurred during the peak training season for the law enforcement training center.

Method
This study was designed to capture an officer’s draw speed and accuracy when the officer was
exposed to a stressor. Stress can alter a person’s fine motor skills, which can greatly affect his or her
ability to perform normal tasks (Grossman & Christensen, 2007; Martaindale et al., 2017). The
stressor in this study was a suspect armed with a shock knife standing 21 feet away. A shock knife is
a training tool designed to look like a knife. Instead of having a sharp edge, electricity arcs around
the edge of the knife where the blade would be. The knife makes popping noises like a Taser when
activated and delivers an uncomfortable shock if touched.
The participant officers were given a training pistol that fired force-on-force training rounds. The
training pistol was loaded with one training round. All officers placed the training pistol in their duty
holsters. The officers were told that their draw speed and accuracy were being researched as part of an
examination of the 21-foot rule. Additionally, the officers were told that they would be going through
a series of scenarios that may or may not involve the use of the training weapon. This was done so
that officers would not be primed to always draw and fire. In reality, each officer completed only one
scenario, and in this scenario, the attacker wielding the shock knife always charged the officer.
At the start of the scenario, the officer was told that he or she was responding to a disturbance
outside of a bar and that the suspect needed to be interviewed. The officer was placed inside of
a square marked with tape on the floor and instructed to stay inside of this box. The suspect was
played by an assistant wearing a force-on-force protective mask, a baggy jacket, and protective
gloves. The shock knife was concealed in one of the jacket pockets. We discovered during pilot
testing that officers would often try to game the system by keeping their hands near their holsters
and their body tensed to react. To counter this, the suspect would engage in a few minutes of
conversation with the officer to allow the officer to acclimate to the scenario and behave more
normally. At some point during the interview, the suspect would charge the officer while pulling the
shock knife out of his pocket. It was not the goal of the assistant to reach the officer and make
contact with the shock knife. The suspect would charge the officer straight on until the last second
when he would veer off to one side of the officer. The initial movement of the suspect was the
stimulus for the officer to draw and fire his or her weapon, and the use of the shock knife was
designed to provide some stress to the officer.
A GoPro camera running at 30 frames per second was used to record each exchange. The camera
was set at hip level next to the officer’s holster to capture the entire scenario. Hits and misses were
recorded during the individual runs of the scenario. The suspect’s initial movement, the point at
which the officer touched his or her weapon or holster, and the frame where the officer fired the
weapon were coded using Adobe Premiere video editing software. Using these frame counts, the
time for each action was calculated. Twenty percent of the cases were coded by another researcher
for reliability with a reliability of 100% (ICC = 1).

Results
Study 3 was designed to test officer draw speed and accuracy under stressful conditions. Of the 57
officers participating, seven did not draw their weapon successfully and, therefore, did not shoot.
 POLICE PRACTICE AND RESEARCH 9

These officers had difficulty clearing the retention systems on their holsters. The average time it took
an officer to touch their weapon/holster once the suspect began to move was 0.32 seconds
(SD = 0.09, Range = 0.1–0.63 seconds). It took the officers an average of 0.68 seconds (SD = 0.24,
Range = 0.3–1.27 seconds) after that to draw their weapon. Once their weapon cleared the holster it
took an average of 0.43 (SD = 0.10, Range = 0.17–0.78 seconds) seconds to raise and fire their
weapon. Of the 50 officers who drew their weapons, the average reaction time to draw and fire was
1.43 seconds (SD = 0.26, Range = 0.93–2.4 seconds). Of the 50 officers who fired, 38 (76%) struck
the charging suspect.

Discussion
Study 3 revealed that the reaction time of police officers was slightly, but significantly, faster than the
1.5 seconds that it took suspects to run the same distance in Study 1 (t = 2.27, p < 0.05, Cohen’s
d = 0.38). Figure 1 presents the distributions for both the suspect run times and officer draw times.
Seventy-eight percent of the distributions from Study 1 and Study 3 overlap, showing that if officers
were randomly paired with runners, they would not shoot before the suspects reached them in
many encounters. Recall that in 7 (12%) of the runs, the officers failed to draw and fire their
weapons. These officers could not be analyzed using time data as they never completed the task.
These cases demonstrate the fact that some officers may never successfully draw their weapon
resulting in a ‘loss’ for the officer. This is a whole new topic that could be analyzed and discussed in
future research. Additionally, officers hit the charging suspect in only about 3/4 of the runs. Unless
these shots hit the suspect in the central nervous system they would likely be ineffective at stopping
the suspect from continuing to charge the officer. Also keep in mind that these scenarios were only
moderately stressful. Real life and death encounters would be substantially more stressful and would
likely result in higher failure to fire and miss rates. Taken together, these results clearly show that 21
feet is not a safe distance. Our results suggest that the answer to research question 3 (Can officers
draw and fire before a suspect moves 21 feet?) is no.
Our calculations suggest that in order for 95% of the officers who successfully drew and fired
their weapons to fire before the suspect reached them would require the officer and suspect to be
about 32 feet apart. Clearly many police contacts must occur at distances that are closer than 32 feet.
This suggests that the police should utilize tactics to mitigate this potential threat. Many tactics are

Figure 1. 21 foot run times v. shooting under stress.

10 W. L. SANDEL ET AL.

currently used. For example, having suspects face away from officers and put their hands on their
heads would provide officers with more reaction time. Approaching a suspect so that there are
obstacles between the suspect and the officer is another tactic. We wanted to address a different
suggestion that seems to be popular in training circles. The tactic involves the officer moving from
the location where he or she was standing when that attack began.

Study 4
Study 4 assessed what impact officer movements had on suspects’ ability to charge and touch an
officer with a marking knife.

Design
Study 4 was a 1 × 4 independent groups design with random assignment to test conditions. The test
conditions were no movement, a 45-degree angle towards the suspect, sidestepping, and
backstepping.

Sample
Participants were recruited from criminal justice classes at a large southwestern university as part of
course credit. A total of 137 students participated in study four. The mean age was 21.56 (SD = 3.73)
years. Of the 137 participants, 40.9% were Caucasian, 39.4% were Hispanic, 15.3% were African
American, and 4.4% were Asian or other. Most of the participants were male (55.5%). Three
participants had prior law enforcement experience, one of whom also had prior military experience.
Seven other individuals had only prior military experience.

Method
Participants played the role of suspects who were being questioned by a police officer. They were
told that, as the officer questioned them, they should charge the officer and try to ‘cut’ him with
a chalk knife whenever they felt ready. The chalk knife is a plastic knife-shaped training tool that has
felt around the edges. This felt holds chalk and leaves a mark when it touches something.
Participants were also told to get to the officer as quickly as they could from 21 feet away. The
same highly experienced officer completed all runs of the study. Based on a fixed condition rotation
schedule, the officer either did not move at all, sidestepped, moved at a 45-degree angle towards the
suspect, or backpedaled away from the suspect while attempting to draw and fire his pistol. The
pistol fired security blanks and was used simply as a stimulus. Participants were randomly assigned
to these movement conditions and were blind to what actions the officer would take. The
participant playing the suspect was instructed to stop the charge if the officer was able to draw
and fire his or her weapon before the participant touched the officer with the chalk knife. The officer
reported if he was touched with the knife and was inspected for marks after the run was over. The
exchanges were captured by GoPro cameras recording at 60 frames per second and examined by
using Adobe Premiere video editing software. Twenty percent of the cases were coded by another
researcher for reliability with a reliability of 100% (ICC = 1).

Results
The suspect was able to reach the officer and make contact with the chalk knife 33% of the time
when the officer did not move. The officer was touched with the knife 25.6% of the time when he
moved in a 45-degree angle toward the suspect. When the officer backpedaled away from the
suspect, he was touched 7.7% of the time. The officer was only touched 5.3% of the time when he
sidestepped. Using Fisher’s Exact test there was a significant difference reflective of a moderate
effect size (p = 0.021, Cohen’s d = 0.66) when comparing how many times the participants made
contact with the knife in the different conditions. When the officer did not move, he was touched
33% of the time. The three types of movement can be aggregated into one large movement group.
 POLICE PRACTICE AND RESEARCH 11

For movement overall, the officer was touched 12.9% of the time. There was a significant difference
reflective of a moderate effect size (Fisher’s Exact p = 0.046, Cohen’s d = 0.56, OR = 0.30) between
the movement and non-movement group regarding the officer being stabbed.

Discussion
Simple movement techniques allowed the officer to avoid being touched by the knife more than
simply standing still and drawing. Moving sideways and backwards in particular were effective. It
should be noted, however, that moving backwards is often tactically frowned upon because the
officer cannot see what is behind him or her and is at an increased risk of falling. Additionally, this
was only an initial look into how movement affected the safety of an officer being charged by a knife
wielding suspect.

General discussion
Recall that the average run speed was 1.5 seconds (Study 1), the average draw and fire speed was
1.80 seconds (Study 2), and the draw and fire speed under stress was 1.42 seconds (Study 3; see
Figure 2 for a comparison of these three studies). As previously stated, the difference between run
speeds and an officer’s ability to draw and fire their weapon under stress was significant. However,
this is a case of statistical significance not being of practical significance. Practically speaking, the
difference of 0.08 seconds translates to 13.4 inches from the chest of the officer to the chest of the
suspect. When considering the median run speed of 1.48 seconds this distance drops to just
10 inches. This does not consider if the suspect has an outstretched weapon or if the officer’s
weapon is fully extended. Once this is considered, the distance between the officer and suspect
disappears. The distributions of the officer draw speeds and suspect run times have significant
overlap (see Figure 2). This overlap shows that 21 feet should not be considered a safe distance.
The current studies eliminate some of the decision-making process for the officers. Their
movements and options were constrained, so there were fewer options to consider. The decision
to shoot was devoid of many of the consequences that would accompany an actual shooting. No one

Figure 2. 21 foot run v. shooting with no stress v. shooting under stress confidence intervals.
12 W. L. SANDEL ET AL.

would be injured; the officers’ actions would not be reviewed potentially resulting in disciplinary or
criminal sanctions. Some research has suggested that 75% of the time it takes an officer to fire an
already drawn weapon in response to a simple stimulus is the decision process; and the remaining
25% is the physical action of pulling the trigger (Lewinski & Hudson, 2003a). Taken together, we
believe that the laboratory conditions in these studies present an optimistic picture of officers’
ability to respond in real-life, high stress situations. We expect that officers in real shooting
situations would perform more poorly than they did here. For example, the accuracy that we
observed for officers in Study 3 was 75%. Research into the accuracy of police officers during actual
shootings shows that their accuracy is often substantially below 50% (Copay & Charles, 2001; Geller
& Scott, 1992; Matulia, 1985; White, 2006). In an evaluation of the New York City Police
Department, Rostker et al. (2008) found an average of only 18% of officer shots hit their target.
Additionally, it should be noted that one round may not instantly stop an attacker. There are
numerous examples of officers hitting a suspect several times and still not stopping the suspect’s
movement. This suggests that officers may often need to hit assailants with multiple shots to stop an
attack. These additional shots would take more time, suggesting that officers would be at even more
of a disadvantage. It should also be noted that the samples used in this study were convenience
samples. That being said, the police samples were closely in line with the national demographic
breakdown of law enforcement in the US.
Given that officers can be at a reaction time disadvantage when making contact with suspects, it
is important to develop officer safety techniques that can mitigate this disadvantage. Study 4 is an
initial look into simple movement techniques that can help mitigate the risk to officers. Other
techniques (such as shooting from the hip instead of a full ‘pushed out’ shooting position, falling to
your back, or rolling at a 45-degree angle past the suspect) might also mitigate risk. That being said,
more training would need to be conducted for officers to become comfortable with multiple
response tactics.
Training on different starting positions could also be useful as each use of force encounter is
unique. Dysterheft et al. (2013) discuss how different starting positions and target focusing can
significantly change a person’s sprint velocity over short distances. Lewinski, Dysterheft, et al.
(2015) go further by examining the effect of an officer’s equipment on movement and discovered
that the weight of a duty belt significantly reduced velocity. These studies indicate a number of
factors that could reduce an officer’s reaction time in a use of force situation. Nieuwenhuys et al.
(2017) found increased levels of anxiety among police officers when they had to fire their weapon
before moving compared to firing their weapon after moving away from the suspect. Training that
covers both tactics might reduce this anxiety and better prepare the officer for instances where one
method might be superior to the other given situational factors. Additional research is needed to
assess these alternative strategies for all aspects of an officer’s reaction time in a use of force
encounter.

Policy implications
There are two major implications based on the findings of this research. First, the idea that 21 feet is
a safe distance for an officer to stop a charging suspect does not appear to be supported. This is
especially true when considering that officers tend to have a high rate of missing the target and that
one shot rarely stops a suspect’s forward movement. It is also important to remember that this study
took place in a laboratory setting, which gave the officers a best-case scenario and the greatest
chance of success. No matter its statistical significance, a distance of about one foot on average
between the officer and suspect is not a practical distance when considering officer safety. The term
‘safe distance’ has allowed the 21-foot rule to become a standard in the field, but it places officers in
danger.
The second implication comes from the final study. Officers are frequently trained to use their
firearms in a very sterile environment when compared to the environment of actual shootings.
 POLICE PRACTICE AND RESEARCH 13

Officers generally practice shooting from a static position on the firing line and shoot a known
course of fire at a stationary target (e.g., shoot three rounds at center mass; Adams et al., 2009;
Aveni, 2003; Lewinski, Avery, et al., 2015). Study 4 found that not moving resulted in a higher
percentage of the attackers hitting the officer with the chalk knife. While this is only an initial look
into the effect of movement on officer safety, it implies that having officers move while drawing and
firing their weapon might be a better method of training. Officers are likely to revert to whatever
training they have had when a sudden, stressful situation arises. Having officers only train in
a stationary position means they are more likely to remain stationary when being charged by
a suspect.
This is often referred to as a ‘training scar’ in police training circles. A training scar occurs when
something done in the training environment (usually for the purpose of expediency) causes the
trainee to perform poorly in real-life situations. Most live firearms training is done on a flat range at
known distances with fixed courses of fire because this type of training can be done safely with many
participants on the training line at the same time (i.e., it is expedient). Having officers move and fire
variable numbers of rounds means that far fewer officers can be safely trained in the same area at the
same time. Even though trainers know that the training environment does not match the actual
shooting environment, they persist in the mismatched training because they can more efficiently
move students through the training. This is also one explanation for the low accuracy of police
officers in actual shootings. While the officers might have extensive training, the training does not
match the real-world shooting task that the officers face. Extensive research in the area of sports
training finds that matching training to the specific task produces much better results in actual
competitive situations than mismatched training (see, for example, Burroughs, 1984; Christina
et al., 1990; Singer et al., 1994; Williams & Burwitz, 1993).

Conclusion
The 21-foot rule seems to have been set as a standard in law enforcement, both socially and legally,
with very little evidence or understanding of its consequences. The results from this study show 21-
feet to be an ineffective distance for officers to draw and fire their weapon at a charging assailant.
Policing involves highly fluid encounters that can change in fractions of seconds. By assigning
a specific distance to when it is safe to use a firearm, police officers are limited in their ability to act
when necessary. This study points to the potential for removing such anecdotal standards and
moving towards effective training for officers who encounter such situations. Recall that for 95% of
officers to be able to draw and fire their weapons at a charging suspect successfully, they would need
to be 32 feet apart. It is not practical for officers to maintain a distance of 32 during all encounters.
However, effective movement techniques could mitigate this distance. The importance of training
to improve muscle memory could be highly effective compared to the traditional static shooting
positions officers are typically trained in.
Future research into this topic could focus on the ability of officers to move in a variety of
environments. As stated earlier, the environment can dictate an officer’s ability to move in regards
to the suspect. For this reason, different movement strategies in different spaces, apart from those
examined in this study, could be researched. This study limited the space used to a flat open area
with assigned movements. Future research could set up different environments for officers to train
around so that the efficacy of movement training in general could be assessed rather than specific
movement methods.
As a final note of caution, we do not intend for this work to be utilized to justify decisions based
on ‘he was within 21 feet, so I shot him,’ thinking. While the distance between a police officer and
a suspect is an important consideration in use of force situations, it is only one factor. Police use of
force thinking is often much more nuanced than this. Use of force encounters are dynamic and each
one presents different situational factors that must be considered (e.g., subjects age, capabilities,
position, demeanor, etc.).
14 W. L. SANDEL ET AL.

We hope that our work will dispel the belief that 21 feet is ‘safe’ and encourage officers to assess
situations more accurately. Officers should know that as they close distance with a suspect, they are
giving up the reaction time advantage that distance can create. We also hope that this knowledge
will help officers and trainers to develop more techniques and tactics that officers can use when near
suspects in order to increase both officer and suspect safety.

Disclosure statement
No potential conflict of interest was reported by the authors.

Notes on contributors
William L. Sandel – Dr. Sandel is an Assistant Professor of Criminology and Criminal Justice at Missouri State
University. Dr. Sandel started his career as a Research Specialist at the ALERRT Center. His research interests include
police and citizen perceptions of use-of-force, police tactics, active shooter events, and hostage negotiations.
Dr. Sandel also conducts research in the area of conservation criminology where he uses his background in
Marine Biology to examine crimes against wildlife.
M. Hunter Martaindale – Dr. Martaindale is the Director of Research at the Advanced Law Enforcement Rapid
Response Training (ALERRT) Center at Texas State University. Dr. Martaindale is responsible for the development
and implementation of ALERRT’s research agenda. His research interests include active shooter events, law
enforcement decision making, and the impact of stress on law enforcement performance.
J. Pete Blair – Dr. Blair is the Executive Director for the ALERRT Center at Texas State University. Dr. Blair’s
research with ALERRT has focused on active shooter events and policing tactics.

ORCID
William L. Sandel http://orcid.org/0000-0002-0582-0724
M. Hunter Martaindale http://orcid.org/0000-0002-8100-7698

References
Adams, R. J., McTernan, T. M., & Remsberg, C. (2009). Street survival: Tactics for armed encounters (27th ed.).
Calibre Press.
Associated Press (2008). 5 die in knife rampage at Chinese police station. NBC News. Retrieved April 8, 2020, from
http://www.nbcnews.com/id/25475269/ns/world_news-asia_pacific/t/die-knife-rampage-chinese-police-station
/#.Xo4EbdNKjYU
Aveni, T. (2003). Following standard procedure: A long-term analysis of gunfights and their effects on policy and
training. Law and Order, 51(8), 78–87. https://www.ncjrs.gov/app/abstractdb/AbstractDBDetails.aspx?id=202511
Blair, J. P., & Martaindale, M. H. (2014). Evaluating police tactics: An empirical assessment of room entry techniques.
Routledge.
Blair, J. P., & Martaindale, M. H. (2017). Throwing a chair could save officers’ lives during room entries. International
Journal of Police Science & Management, 19(2), 110–119. https://doi.org/10.1177/1461355717711452
Blair, J. P., Pollock, J., Montague, D., Nichols, T., Curnutt, J., & Burns, D. (2011). Reasonableness and reaction time.
Police Quarterly, 14(4), 323–343. https://doi.org/10.1177/1098611111423737
Blake, D. (2016, January 13). What officers really need to know about the 21-foot kill zone. Police One. https://www.
policeone.com/police-products/firearms/articles/63695006-What-officers-really-need-to-know-about-the-21-foot
-kill-zone/
Brebner, J. T., & Welford, A. T. (1980). Introduction: A historical background sketch. In A. T. Welford (Ed.), Reaction
times (pp. 1–23). Academic Press.
Burroughs, W. A. (1984). Visual simulation training of baseball batters. International Journal of Sport Psychology, 15
(2), 117–126. https://psycnet.apa.org/record/1985-16394-001
CA AB-392. 2019. Peace officers: Deadly force 2019 (Regular Session)(U.S.A.)
Campbell, J. H. (1992). A comparative analysis of the effects of post-shooting trauma on special agents of the Federal
Bureau of Investigation [Unpublished Ph.D. Dissertation]. Department of Educational Administration, Michigan
State University.
 POLICE PRACTICE AND RESEARCH 15

Christina, R. W., Barresi, J. V., & Shaffner, P. (1990). The development of response selection accuracy in a football
linebacker using video training. The Sport Psychologist, 4(1), 1. https://doi.org/10.1123/tsp.4.1.11
Copay, A. G., & Charles, M. T. (2001). Handgun shooting accuracy in low light conditions: The impact of night
sights. Policing. An International Journal of Police Strategies and Management, 24(4), 595–604. https://doi.org/10.
1108/EUM0000000006499
Dysterheft, J. L., Lewinski, W. J., Seefeldt, D. A., & Pettitt, R. W. (2013). The influence of start position, initial step
type, and usage of a focal point on sprinting performance. International Journal of Exercise Science, 6(4), 320–327.
https://digitalcommons.wku.edu/ijes/vol6/iss4/7/
Eckner, J. T., Kutcher, J. S., & Richardson, J. K. (2010). Pilot evaluation of a novel clinical test of reaction time in
National collegiate athletic association division I football players. Journal of Athletic Training, 45(4), 327–332.
https://doi.org/10.4085/1062-6050-45.4.327
Eith, C., & Durose, M. R. (2011). Contacts between police and the public, 2008. Bureau of Justice Statistics.
Federal Bureau of Investigation. (2017). Expanded homicide data [Data file]. https://ucr.fbi.gov/crime-in-the-u.s/
2016/crime-in-the-u.s.-2016/topic-pages/expanded-homicide
Follette, V. M., Polusny, M. M., & Milbeck, K. (1994). Mental health and law enforcement professionals: Trauma
history, psychological symptoms, and impact of providing services to child sexual abuse survivors. Professional
Psychology, Research and Practice, 25(3), 275–282. https://doi.org/10.1037/0735-7028.25.3.275
Force Science Research Center. (2005a, May 23). Edged weapon defense: Is or was the 21-foot rule valid? (Part 1).
https://www.policeone.com/edged-weapons/articles/102828-Edged-Weapon-Defense-Is-or-was-the-21-foot-rule
-valid-Part-1/
Force Science Research Center. (2005b, June 13). Edged weapon defense: Is or was the 21-foot rule valid? (Part 2).
https://www.policeone.com/close-quarters-combat/articles/113907-Edged-Weapon-Defense-Is-or-was-the-21-
foot-rule-valid-Part-2/
Geller, W., & Scott, M. S. (1992). Deadly force: What we know. Police Executive Research Forum.
Graham v. Connor. (1989). 490 U.S. 386.
Grossman, D., & Christensen, L. W. (2007). On combat: The psychology and physiology of deadly conflict in war and
peace. Human Factor Research Group Incorporated.
Haley, S. S., & Davis, E. (2019). Local police departments, 2016: Personnel. Bureau of Justice Statistics. https://www.bjs.
gov/index.cfm?ty=pbdetail&iid=6706
Illinois v. Gates. (1983). 462 U.S. 213.
Johnson v. Glick. (1973). 481 F.2d 1028 (2d Cir.
Keinan, G. (1987). Decision making under stress: Scanning of alternatives under controllable and uncontrollable
threats. Journal of Personality and Social Psychology, 52(3), 639. https://doi.org/10.1037/0022-3514.52.3.639
Klinger, D. A., & Brunson, R. K. (2009). Police officers’ perceptual distortions during lethal force situations:
Informing the reasonableness standard. Criminology & Public Policy, 8(1), 117–140. https://doi.org/10.1111/j.
1745-9133.2009.00537.x
KOKO. (2018, December 17). Officer shoots, kills man carrying knife and screwdriver at west OKC gas station. KOKO
News. https://www.koco.com/article/police-man-carrying-knife-screwdriver-shot-and-killed-by-officer/25602170
Law Enforcement Officers Killed and Assaulted. (2018). Law enforcement officers assaulted [Data file]. https://ucr.fbi.
gov/leoka/2016/officers-assaulted/assaults_topic_page_-2016
Lewinski, W., & Hudson, B. (2003a). September/October). Time to start shooting? Time to stop shooting? The Tempe
Study. Police Marksmen, 7(5), 26–29.
Lewinski, W., & Hudson, B. (2003b, November/December). The impact of visual complexity, decision making and
anticipation. Police Marksmen, 28(6), 24–27.
Lewinski, W. J., Avery, R., Dysterheft, J., Dicks, N. D., & Bushey, J. (2015). The real risks during deadly police
shootouts: Accuracy of the naive shooter. International Journal of Police Science & Management, 17(2), 117–127.
https://doi.org/10.1177/1461355715582975
Lewinski, W. J., Dysterheft, J. L., Dicks, N. D., & Pettitt, R. W. (2015). The influence of officer equipment and
protection on short sprinting performance. Applied Ergonomics, 47(1), 65–71. https://doi.org/10.1016/j.apergo.
2014.08.017
Luce, R. D. (1986). Response times: Their role in inferring elementary mental organization. Oxford University Press.
Martaindale, M. H., Sandel, W. L., & Blair, P. J. (2017). Active-shooter events in the workplace: Findings and policy
implications. Journal of Business Continuity & Emergency Planning, 11(1), 6–20. https://www.ingentaconnect.
com/content/hsp/jbcep/2017/00000011/00000001/art00002
Martinelli, R. (2014, September 18). Revisiting the “21-foot rule” Police: The Law Enforcement Magazine. http://www.
policemag.com/channel/weapons/articles/2014/09/revisiting-the-21-foot-rule.aspx
Martinelli, R. (2015, March 2). The 21 foot rule – Training. file:///C:/Users/wls43/Desktop/21%20Foot%20Rule/The%
2021%20Foot%20Rule%20-%20Training%20-%20LawOfficer.com%20_%20Alpha%20Krav%20Maga%20 CT_
files/The%2021%20Foot%20Rule%20-%20Training%20-%20LawOfficer.com%20_%20Alpha%20Krav%20Maga%
20CT.html
16 W. L. SANDEL ET AL.

Matulia, K. J. (1985). A balance of forces: Model deadly force policy and procedure (2nd rev ed.). International
Association of Chiefs of Police.
National Law Enforcement Officers Memorial Fund. (2017). Preliminary 2017 law enforcement officer fatalities report.
http://www.nleomf.org/facts/research-bulletins/?mfc_popup=t
National Law Enforcement Officers Memorial Fund. (2018). Causes of law enforcement deaths [Data file]. http://www.
nleomf.org/facts/officer-fatalities-data/causes.html?mfc_popup=t
Nieuwenhuys, A., Weber, J., van der Hoeve, R., & Oudejans, R. R. (2017). Sitting duck or scaredy-cat? Effects of shot
execution strategy on anxiety and police officers’ shooting performance under high threat. Legal and
Criminological Psychology, 22(2), 274–287. https://doi.org/10.1111/lcrp.12099
Police Executive Research Forum. (2015). Re-engineering training on police use of force. http://www.policeforum.org/
assets/reengineeringtraining1.pdf
Police Executive Research Forum. (2016). Guiding principles of use of force. http://www.policeforum.org/assets/30%
20guiding%20principles.pdf
Rostker, B. D., Hanser, L. M., Hix, W. M., Jensen, C., & Morral, A. R. (2008). Evaluation of the New York City Police
Department firearm training and firearm-discharge review process. Rand Corporation.
Sharrar v. Felsing. (1997). 128 F.3d 810 (3d Cir.
Shin, A. (2017, August 3). Recalling the protests, riots after fatal police shooting of Michael Brown. The Washington
Post. https://www.washingtonpost.com/lifestyle/magazine/recalling-the-protests-riots-after-fatal-police-shooting
-of-michaelbrown/2017/08/01/9992f044-5a8d-11e7-a9f6-7c3296387341_story.html?noredirect=on&utm_term=.
92515d383ac5
Simon, D., Hamasaki, S., & Park, M. (2018, March 23). Protestors block NBA arena over Sacramento police shooting.
CNN. https://www.cnn.com/2018/03/22/us/sacramento-police-shooting-protests/index.html
Singer, R. N., Cauraugh, J. H., Chen, D., Steinberg, G., Frehlich, S. G., & Wang, L. (1994). Training mental quickness
in beginning/intermediate tennis players. Sport Psychologist, 8(3), 305–318. https://doi.org/10.1123/tsp.8.3.305
Smith, J. (2018, May 17). Is the 21-foot rule really a thing? [blog]. Retrieved from http://gunbelts.com/blog/is-the-21-
foot-rule-really-a-thing/
Sobol, R., DiNuzzo, N., Leone, H., ., & Gorner, J. (2018, July 15). Violent clash between officers and crowd after fatal
police shooting in South Shore. Chicago Tribune. https://www.chicagotribune.com/news/local/breaking/ct-met-
police-shooting-fatal-20180714-story.html
Solomon, R. M., & Horn, J. H. (1986). Post-shooting traumatic reactions: A pilot study. In J. T. Reese &
H. A. Goldstein (Eds.), Psychological services for law enforcement officers (pp.383–393). U.S. Government
Printing Office.
Tennessee v. Garner. (1985). 471 U.S. 1.
Terrill, W. (2009). The elusive nature of reasonableness. Criminology & Public Policy, 8(1), 163–172. https://doi.org/
10.1111/j.1745-9133.2009.00540.x
Tueller, D. (1983, March). How close is too close? SWAT Magazine.
Welchman, A. E., Stanley, J., Schomers, M. R., Miall, R. C., & Bülthoff, H. H. (2010). The quick and the dead: When
reaction beats intention. Proceedings of the Royal Society of London B: Biological Sciences, 277(1688), 1667–1674.
https://doi.org/10.1098/rspb.2009.2123
White, M. D. (2006). Hitting the target (or not): Comparing characteristics of fatal, injurious, and noninjurious police
shootings. Police Quarterly, 9(3), 303–330. https://doi.org/10.1177/1098611105277199
Williams, A. M., & Burwitz, L. (1993). Advance cue utilization in soccer. Science and Football II, 2, 239–244.