Trending Topics

Defeating the blur: Rethinking traditional police training to close quarter spontaneous attacks

The reality of attacks on officers and how we train for them


Combatives training is typically conducted in a siloed, linear fashion where instruction heavily emphasizes replicating the techniques of various movements rather than rapid, adaptive decision-making and solving novel combative problems with effective movement solutions.

Getty Images

This article is reprinted with permission from Calibre Press

By Chris Butler and Dan Fraser

If you knew you were going to be suddenly ambushed and assaulted by an assailant at extremely close quarters on your next shift, how would you prepare for it? You’d probably have a lot of questions about who your opponent is and how they plan to attack you. You would want to know the pre-assaultive cues they would exhibit, whether a weapon would be involved and if so, what kind of weapon. Would it be a fisted attack? What would be the speed and angles of the attack?

Once you had some answers to your questions you would hopefully spend all the time possible getting busy preparing to meet and defeat the attack. But how would you actually prepare? It would be ridiculous to spend 90% of your training time hitting a heavy bag and only a small amount of time sparring against a real person who moves and strikes in biomechanically realistic ways at the speed of combat. So, why do we typically train our combatives and gunfighters this way?

With respect to firearms training, most officers will spend the bulk of their precious training time standing flat-footed, shooting holes in a piece of paper. It’s no wonder we have an operational shooting accuracy that averages 35% and offenders who are hitting officers with an accuracy approaching 70% with their first round. This training is pernicious because the context of this type of training environment has virtually no relevance to “the real world.”

Combatives training is typically conducted in a siloed, linear fashion where instruction heavily emphasizes replicating the techniques of various movements rather than rapid, adaptive decision-making and solving novel combative problems with effective movement solutions. The results of this type of training are certainly easy to find. Just search YouTube for five minutes and you will have an endless number of examples of combative encounters of officers exhibiting serious performance failures.

Just to be clear, the officers who are struggling to perform well in these encounters are not to blame. We are simply seeing the predictable result of the training methodology that has been used for generations and continues to be handed down with a survivorship bias that just refuses to die.

This article is a “call to action” for trainers.

Let’s start at the very beginning

In every occupation where high-consequence environmental demands are placed upon humans, we must first thoroughly understand the nature of the threat environment. Only then can we take a critical approach to designing equipment and training for the acquisition of the skills and adaptability that are necessary for our success.

The starting point for exploring the type of training that will be most needed to meet the demands faced by officers in the field is to first analyze what is actually occurring in the field. This seems like common sense. However, in a study that assessed the performance of officers during and following academy training, the authors discovered that the majority of the content of the physical skills training curriculum was not based on any evidence but upon both traditionalism (it’s what we’ve always taught) and instructor preference (it’s what I think is best). [1]

Point 1: A high percentage of attacks initiate at extremely close range

How close?

Before we conduct any training, let’s explore the threat environment – what are your officers most likely to face in the contexts in which they operate? Current data shows that the vast majority of attacks against officers (weapon and weaponless) happen at a conversational distance – inside of 10 ft. Additionally, a high percentage of these spontaneous assaults occur at contact or near-contact range. This reinforces the fact that one of the points of greatest danger is “danger close.” Often an attack is initiated at the point of first physical contact by the officer on the subject.

Point 2: Attacks will often begin and be completed in less than the blink of an eye

How fast?

The research also shows that the movement time (from initial movement to completion) for nearly all forms of attack – from a punch, to a kick, to a knife thrust, to a gun draw – all happen in under 250 ms (1/4 of a second). [2] For example, a dominant hand forward slash attack takes only 170 ms from the concealed position (hand on weapon) from the start of the movement to the time the blade makes contact.

How fast is that? Well, considering that an eye blink is typically between 100 and 300 ms, it’s no wonder that, in many cases, officers are stabbed and don’t immediately recognize it. They often think they’ve been punched. It is extremely difficult, often impossible, to be able to discern if a person has an object in their hand and if they do, what the object is. More on object recognition in a moment.

Point 3: The incoming attack will predominantly have a trajectory towards the officer’s head

Get out of my face

FBI Law Enforcement Officers Killed and Assaulted (LEOKA) data from 2009-2018 indicates that an average of 64% of all officers who were shot were shot in the head or neck. Why? Well, just think of any athletic event where eye and hand coordination is important. Eye tracking studies have determined that prior to the initiation of the critical motor movement, performers first fixate their gaze on the most important area of the scene. This is due to what Dr. Joan Vickers calls gaze-action coupling. [3] This means that the visual cortex coordinates vision with the motor cortex to control the movement. Essentially, humans are very good at naturally, and automatically, hitting what we’re looking at.

With respect to shooting situations, research has discovered that at “typical” gunfight distances, even novice shooters are highly proficient at shooting where they look and they do not use the sights of the weapon. [4] This same process is also involved with knife attacks and fisted attacks. The most common location of first contact by the offender is the officer’s head, neck, or face because that’s where they’re looking.

As an aside, if you reflect back to the times you have acted as a role player in force-on-force scenarios and have repeatedly been shot in the hand and arm with which you were holding the weapon, you have experienced firsthand the power of gaze-action coupling as the officers visually fixated on your weapon.

Point 4: You most likely will not be able to identify what is coming

Now you see it – actually, no you don’t

How the brain and eye work together to be able to accomplish object recognition is not fully understood. It is a remarkably complex process and is impacted by many factors. However, researchers do agree that the type of vison utilized to perform object recognition is our foveal vision. Foveal vision is the vision that we use to see with clarity, and to see in normal lighting conditions. The fovea of the eye is densely packed with millions of cones that enable it to perform object recognition (among other functions). However, there is a limitation. Before you read any further, please take your pen and put a dot 1/16 of an inch (1.5 mm) in diameter on a piece of paper. That is the size of your fovea at the back of your retina.

Additionally, we only see with visual clarity between one to three degrees of visual angle. As an example, stretch your arm out in front of you and put your thumb up (like the Fonze). Stare intently at your thumbnail and bring it into clear visual focus. The lens of your eye is converging the light coming off of your thumb onto that tiny fovea. Your thumb is approximately three degrees of visual angle – that is all you see with visual clarity, and you are using your fovea to do it.

So, what’s the issue? It has been discovered that in order for your brain to accurately perform object recognition, the object must remain stable on the fovea of the eye for a brief period of time. If the object is moving rapidly through your visual field, it can be extremely difficult to determine if there is an object, and if there is, what exactly the object is.

If something were to fly in from the side you may catch it with your peripheral vision, often stated as ‘catching it out of the corner of your eye’, you just won’t be able to see it clearly. Research published in “Psychological Science” in 2005 showed that object detection (knowing that ‘something’ is there) and object categorization (like, it’s a person) occur almost simultaneously. [5]

Correctly identifying what that object is takes longer. Accurate detection and categorization take at least 50 ms (of the object on the fovea of the eye). If we want just 60% accuracy in correctly identifying what that object is it must be stable on the fovea for at least 75 ms. Now, this study was done in a laboratory by flashing images very quickly on a screen. The participants were attending to the screen and weren’t distracted by a host of competing things like an officer might be when dealing with a subject. We also know through LEOKA data that officers are significantly more likely to be assaulted during hours of darkness. This will make any object recognition even more challenging. All this is to say that officers have little chance of seeing exactly what, if anything, a subject has in their hand as it comes up rapidly toward the officer’s face. What you will certainly see, is a blur.

Training implications

First, what we need to stop

Let’s take a quick summary of what we have covered so far: attacks will often occur spontaneously, at contact or near contact range, will be completed extremely quickly and you will likely only see a blur, that’s all. Deal with it.

Officers do not rise to the occasion; they default to the level of their training. Not all training is helpful. In fact, poor training can get an officer seriously injured or killed faster than no training.

Many – if not most – agencies separate their firearms and control tactics training. Officers primarily learn to shoot on a range and fight in a padded room. The two skills are siloed and are rarely blended into an effective “combat system” that addresses the threat environment of spontaneous violent attacks with handguns, weapons, or fists at very close range. Officers spend the majority of their firearms training time on a static range, firing at a paper target while keeping their feet planted and their heads still.

Hebb’s Law states that neurons that fire together, wire together. In this case, many officers have wired together a response to draw their sidearm in response to a deadly force threat that is precisely the opposite response to what they need to survive in the real world. Trainers, when will we finally stop doing this to our officers?

The Force Science Traffic Stop Study [6] demonstrated that when faced with a deadly force threat, like a gun being produced by the driver of a vehicle, officers will overwhelmingly default to their ‘training’ and immediately draw their own gun. They have been conditioned for this ‘strong but wrong’ response as a result of static range training repetitions, where the stimulus to act is a piece of paper turning in front of them or an instructor using a sound (like blowing a whistle or shouting). In the real world, this puts officers well behind the action-reaction curve, with subjects being able to fire multiple rounds – often at the officer’s head or neck – before an officer is able to fire back. Regardless of how fast an officer may be able to address a threat with their gun, they will not be able to effectively shoot their way out of this.

To put this into a startling perspective, the average time it takes an officer to draw the handgun from a level II security holster and fire their first round is 1.5 seconds (1.7 seconds for a level III holster). In that amount of time, the offender has been able to fire between 6 and 8 rounds.

Second, the way we need to go

What do we need to do to instill the correct motor response to rapid, sudden threats at close range? Since we likely won’t be able to recognize what is coming at us, but at best, it will only be detected as a “blur.” Therefore, it is clear we cannot have separate responses based on whether that is a punch, an edged weapon, a gun, or something else.

What’s the solution? First – our understanding of the threat environment must start to inform our training design (the “what” and the “why”). Secondly, as instructors, our pedagogy (the “how” we help our officers learn effective movement solutions to defeat these types of attacks) is extremely important.

Officers have to learn the importance of effective movements to get the head and neck out of the way. This automated response must be initiated from the moment of cue/threat recognition. Next, officers must learn how to “fight with firearms.” They must be competent to go hands-on with the subject or move their feet and quickly get “off the X” depending upon the context. If an officer chooses to move, they should be able to draw efficiently and quickly while moving. Of course, it may not be appropriate to draw in all circumstances, therefore rich, high-fidelity, variable contextual drills must be performed in an array of circumstances to help officers develop the game intelligence required for an appropriate response. This can be done with Non-Lethal Training Ammunition, airsoft, blanks, or even red guns.

Next, we need to get officers “left of bang” by allowing them to repeatedly recognize the subject’s biomechanical cues prior to an attack. Officers need rich contextual drills to develop cue recognition. This means using role players who use realistic movements at realistic speeds, especially once officers have some basic skills.

In a 2015 study published in Ergonomics, researchers in Amsterdam looked at the effects of a reflex-based defensive tactics program. [7] The training included all aspects of recognition of verbal, physical and armed aggression cues, reading the situation and anticipating and assertively responding to the earliest signals of potential aggression or attack. Here, student officers were trained and then evaluated on seven criteria:

  • Goal-directed action prior to the attack
  • Visual search patterns
  • Threat recognition
  • Communication and tactical control
  • Speed of initial flinch response
  • Transition from flinch into a tactical maneuver
  • Power & effectiveness.

What was the result? Officers’ performance was tested on several variables in six reality-based scenarios before and after each training intervention. Results showed significantly improved performance, across all ratings, after the reflex-based training, while there was no such effect after the regular police training.

A 2020 study involving German police recruits compared different pedagogical approaches by exposing officers to different types of knife attacks. [8] Which pedagogical approach leads to more efficient knife defense performance – linear or non-linear? A linear style of training is heavily focused on replicating a specific technique and involves both “ideal solutions,” and multiple techniques to various attacks. For example, “If the knife comes from overhead in an icepick grip, the first step is to move towards the subject and block upward with both arms in an X pattern.” In linear training, the focus is on the repetition of isolated movements and very specific direction on technique.

The non-linear training method involves providing the goals or objectives to students and then exposing them to many varied, realistic, attacks and allowing them to “self-organize” and explore functional solutions. Many of these attacks were done in a contextual environment with “chaotic” knife attacks as opposed to the lines of officers and subjects often seen in a mat room. The result? After eight weeks of training the non-linear group not only had far better skill retention but were cut only half as much – and solved attacks in the half the time – as the linear group.

Officers continue to be assaulted, seriously injured and killed by attacks at close range – this is the known threat environment. Preparing officers means taking a critical look at our current training and how it stacks up to what the research tells us is most effective. Tear down the silos and build connective tissue between the critical skill sets of fighting and shooting. Officers must be trained to recognize and respond to threat cues ahead of potential attacks. We need to understand that good performers do not repeat techniques, they repeat outcomes. We have to get away from teaching highly technique-based responses to specific attacks. Training must expose students to contextually rich, varied attacks and allow them to explore functional solutions under the guidance of effective feedback by instructors highly trained in nonlinear pedagogy.


1. O’Neill J, O’Neill D, et al. Police Academy Training Performance and Learning. Behav Anal Pract. 2019;12:353-372.

2. Fuchs L, Lindinger, Schwameder. Kinematic Analysis of Proximal to Distal and Simultaneous Motion Sequencing of Straight Punches. Sports Biomech. 2018;17(4):512-530.

3. Vickers J, Lewinski B. Performing under pressure: Gaze control, decision making and shooting performance of elite and rookie police officers. Hum Mov Sci. 2011;31:101-117.

4. Lewinski B, et al. The real risks during deadly police shootouts: Accuracy of the naïve shooter. Int J Police Sci Manag. 2015;17(2).

5. Grill-Spector K, Kanwisher N. Visual Recognition: As Soon as You Know It Is There You Know What It Is. Psychol Sci. 2005;16(2).

6. Lewinski B, et al. The Influence of Officer Positioning on Movement During a Threatening Traffic Stop Scenario. Law Enforce Exec Forum. 2015;13(1).

7. Renden P, Savelsbergh G, Oudejans R. Effects of reflex-based self-defense training on police performance in simulated high-pressure arrest situations. Ergonomics. 2017;60(5).

8. Koerner S, Staller M, Kecke A. “There must be ideal solution…" Assessing training methods of knife defense performance of police recruits. Policing: An Int J. 2020.

About the authors

Dan Fraser is a retired sergeant and tactics trainer for a large Canadian agency. He continues to train officers in instructor development and human factors in the use of force in the Methods of Instruction Course. He is the author of the book Kickass Presentations. He can be reached at

Chris Butler spent six years as a high-angle search and rescue technician in the Rocky Mountains which began his passion for exploring and understanding human performance in dynamic, high-consequence environments. He served as a law enforcement officer for 28 years with 20 of those years dedicated full-time or part-time to coaching law enforcement use of force skills. Chris is the developer of the Methods of Instruction (MOI) Course. Chris can be contacted at

For more information on the MOI course, visit or

The Calibre Column is provided by Calibre Press, Inc., one of the most recognized and respected law enforcement training organizations in the industry. Offering hundreds of courses each year on topics ranging from Street Survival®, Active Threat Engagement and Assault Detection & Response to Highway Patrol Tactics, Emotional Survival, and Leadership, Calibre Press has trained more than one million officers nationwide over its 40-year history. In addition to on-site courses, Calibre also offers an increasing collection of training-focused videos available through

Contact Calibre Press.