Photo By: Lance Cpl. Christopher Mendoza
Photo By: Lance Cpl. Christopher Mendoza

Gunfights are awesome to watch on TV.  The good guy wins, and bad guys go flying through the air in a spray of blood.  Unfortunately, these types of images we see on the silver screen rarely have much in common with reality.  This can lead us as armed professionals, and concerned citizens, to make some critical errors if we ever find ourselves on a two way range.  Keep reading to discover the 17 most important gunfight statistics backed by data and real world experience.  

This article will break down a wide variety of statistics by topic, in the hopes that we can clarify some persistent errors in our thinking.  The first area we are going to cover is the likelihood an armed person will even be in a gunfight.

1. Each year there are roughly 300-400 police officers that fire their weapons at a threat (source)

2. In an Officer’s career, there is a 12% chance that they will fire their weapon (source)

3. In 2012 the Violence Policy Center counted 259 justifiable homicides, where citizens defended themselves and killed their attacker. (source)

4. National Crime Victimization Survey estimates that about 67,740 times a year a crime is stopped by the presence of a firearm. (source)

5. In 2012, there were 1.2 million violent crimes, defined as murder, forcible rape, robbery and aggravated assault which would justify the use of a firearm (source)

6. When you divide stat 4 by stat 5 you see that about 5.6% of violent crimes are stopped by the presence of a weapon.

What This Means 

I like to first point out the numbers of police officers that are involved in shootings annually, which is incredibly low, considering the amount of police contact with the public.  This is also important to note because a police officer’s sworn duties put them in dangerous volatile situations, where use of force is more common than would be the norm for your average citizen.  From these stats, we can infer that a police officer is roughly twice as likely to use their firearms as the average citizen.

Next, it’s important to understand that firearms realistically stop a low amount of crimes, and are present in the commission of many more.  For example, in statistic 3 we see that there were 259 cases of a civilians defending themselves, and killing their attacker. In that same year there were over 8300 cases of homicide, where a firearm was used.

Now, I know some people might think that I’m anti gun because of these stats.  Nothing could be further from the truth.  Facts are facts.  A non criminal is much more likely to be killed by a firearm than they are to kill with one.  That does not mean that we should not be ready and well trained to use them, because as we saw in stat 4 they were key in stopping over 67,000 criminal acts.

Statistics from the Holster 

This section will cover some key data on deciding to pull a gun, including just how long that might take.

7. The average time for a police officer to mentally justify using their weapons was .21 sec for a simple scenario, .87sec for a complex one. (source)

8. The average time it takes for someone to draw from a level 1, friction retention holster, is 1.71 sec

9.  World Class shooters can draw and fire from a level 1 holster in .76 sec in this case Jerry Miculek


What This Means 

We can see that there is huge room for improvement in our ability to draw our weapon from a holster.  Obviously level 1 holsters are the quickest, and more secure holsters i.e level 3 holsters, will take longer, as will drawing from concealment.  From the stats above it looks like we can expect that a proficient shooter can draw and fire a weapon in about 1.5 seconds, including the decision to use deadly force.

This means that we need to create at least this much time and distance to employ our weapons once we are in a situations that warrants it.  Check out this video for an excellent demonstration of the utility in creating time and space.

If you like this article and you want to get more like it emailed directly to you, go ahead and put your email in the box on the right.  If you don’t, I’ll just assume your an ISIS sympathizer.  Welcome to the Tier Three Team.  

Key Distances and Times

These stats will cover what we need to know about common distances, and times of occurrence  in gunfights.

10.  The deadliest distances for police gunfights is 3-6 feet. (Source)

11.  Experts were only 10% more accurate than novices between 3 and 15 ft. (Source)

12.  About 60% of police shootings occur during hours of darkness (Source)

What This Means

Often times we assume that skill will make all the difference in deadly force encounters.  This is true, but only at extended distances, where it is truly difficult to hit your target.  If you read the source for stat 11, you will see that even shooters that have never touched a gun before can accurately point shoot a weapon at distances of less than 15 feet.  Alarmingly, they are more likely to hit the head because that is where their vision natural goes; where as trained personnel fire center of mass most of time.

Even if you look as badass as these dudes, you’re still in a shit sandwhich if you are in a 3 ft gunfight because it requires very little skill to be very deadly. So we need some method to create time and distance to enhance our survivability.

Shooting and Moving

This section will cover the stats relating to the rounds fired, movements, and other important factors.

13. If you stand still in a gunfight you have an 85% chance of being shot, and 51% chance of being shot in the torso. (Source)

14. If you move and shoot you have a 47% chance of being hit, with 11% chance of a torso shot. (Source)

15. Seeking cover and returning fire reduces your chance of being shot to 26% with a 6% torso hit rate. (Source)

16. The most common caliber to be shot with is 9mm (Source)

17. Most gunfights average 3.59 rounds per incident (Source)

What This Means 

The vast majority of our firearms practice is flat footed on the range.  While this is a necessary part of practicing fundamentals, it will surely get us killed if we fight this way.  If at all possible you need to seek cover in a gunfight, or at the very least you need to move and shoot.

These skills are very advanced, and the vast majority of folks will need to master the simpler tasks of shooting first.  I highly recommend seeking out force on force training with qualified instructors to practice these advanced concepts.


Resources You Can Use.

There are many good resources you can use to help you become a more proficient gunfighter.  If you are trying to learn the fundamentals, I recommend checking out some of Rob Leatham’s videos on Youtube,  as he very quickly breaks down what you need to focus on an what you don’t.

I also think there is a lot of value in books, imagine that, a Marine reading.  My favorite book on the subject of pistol fighting is, Stay in the Fight!!! A Warriors Guide to Combat Pistol. (affiliate link to Amazon).  The author Kyle Lamb was a former Sgt Maj in the Army’s Delta Force, and he has a rocking beard, which is all the credibility I need.

The book breaks down what exactly you need to look for in a pistol, accessories, and how to employ it from the basics, to the most advanced principles.  If you are more into long gun shooting (Amazon affiliate link), he has an excellent version for that as well.


These stats are interesting in and of themselves, but that’s not really what you should take away from this article.  You should keep in mind how these facts might affect your future safety.  Likewise, you need to be honest with yourself and ask if the type of training you are currently doing actually mirrors the conditions you are likely to encounter in the real world. If you are a 25yd bullseye master, but you’ve never shot on the move, you are in for a rude awakening.  If self defense is your goal, then you need to train realistically, and you need to work on things you suck at.









  1. This paragraph is accurate but VERY misleading.

    “Next, it’s important to understand that firearms realistically stop a very low amount of crimes, and are present in the commission of many more. For example, in statistic 3 we see that there were 259 cases of a civilians defending themselves, and killing their attacker. In that same year there were over 8300 cases of homicide, where a firearm was used.”

    The data you cite only shows that armed civilians realistically KILL a small number of attackers. It provides no basis whatsoever for GUESSING at how many crimes are “realistically stop[ped]”.

    Why? Because the police need to subdue and capture the bad guy while the civilian only needs to make the bad guy stop the attack. This is why survey data (the only way of asking this question) demonstrates that 98% of civilian defensive gun uses are successful merely by displaying the gun and determination to use it. Bad guy SEES gun, bad guy RUNS away. No shots fired, no police report, no official statistics.

    Award-winning Criminologist Gary Kleck at Florida State found that “display and run” incidents occurred about 1.5 Million times in the late 1990s. The number will be greater now.

    Civilians need guns, civilians use guns in self-defense, untold lives are saved by that defensive use.

    • If you keep reading for a few more paragraphs you will see that I agree that civilians need to be able to defend themselves. In fact there are numerous articles on this website designed to help civilians do that very thing.
      As far as display and run away; I’m sure that does occur a fair amount, but I will also tell you that many civilians that display their firearms do so illegally. I know this from professional experience. I do agree that crime victimization surveys are very important to find out about crimes that are generally not reported. Thanks for the thoughtful comment.

      • An illegal display of a firearm would include a situation in which an attacker is frightened away by the display even when the attacker’s use of force would not have justified the defender’s use of lethal force. Even though “illegal”, it still results in a successful defense by the defender, and can stop a crime before it is committed. This is a situation in which an “illegal” act can prevent an innocent person from becoming the victim of a crime. Personally, if I were an attorney defending a person from being prosecuted for such a crime, I’d argue before the jury, “Certainly, the law can’t require someone to take a beating/be robbed/be raped when the means for preventing the crime was at hand. And my client did so protect him-/herself without injury to his/her assailant.”

        • Your correct in that the law does not require you to be assaulted or injured before displaying and using a firearm. Let me give you an example of an illegal display of a firearm. A home owner and a contractor have an argument about payment for a carport. The homeowner won’t pay till it’s done. The contractor decided to knock it down with a sledge and starts beating on his house as well. Homeowner retrieves firearm and displays it to guy beating on his house. Guy stops hitting house.
          This is brandishing and not self defense, because at no time was the homeowner in fear for his life. He was just pissed about his house. It was effective in getting the guy to stop, but it was illegal and he was charged. SO lets not rely on breaking the law to keep up safe. There is plenty of leeway in the law to protect yourself and others. I really recommend legal training for concealed carry folks, because knowing how to and know when to shoot are completely different.

          • This is jet another great example why you should never talk to the police before you talk to an attorney.
            During you initial consultation a GOOD Attorney will try to determine the facts by questioning his client. For example, from what you have told me this guy was out of his mind. As a contractor he knows he was committing a fealony destroying your property. Weren’t you afraid he would use that sledge hammer on you?
            BTW: there are other issues. For example, just WHERE was the client standing? Inside the house or outside? Is the carport considered part of the house?
            These are things a GOOD attorney should think of before he advises you to make your statement to the police. Keeping in mind that the contractor has already admitted to committing a crime, destruction of private property. The only question is whether or not the damage reached the level to be considered a fealony – and one committed with a DEADLY weapon. So the proper way to present the question is, does a homeowner have the right to display a firearm to prevent a fealony by use of a deadly weapon on his property (or inside his residence depending on whether the carport is considered part of the home).

          • I agree those are definitely good issues that an attorney should bring up in court and yes they are all important on the scene as well. If I’m playing devils advocate you bring up a lot of doubt that doesn’t address any issues relevant to a brandishing case which is what an attorney should do lol. Really this case doesn’t matter it was only an example where someone might think that displaying a firearm was ok and it was not which is something I think that probably happens a lot more than gets reported to the police.
            I also generally advise against not saying anything to the police, but I admit I could be biased in this regard. If you say nothing to me then I will have to use what evidence and statements the other party is giving which is not the whole story. Thats why I recommend some legal training so folks know what things they can say and then consult with an attorney if they feel the need.

  2. The Violence Policy Center numbers are artificially low. NRA armed citizen anecdotes provide greater numbers of successful armed defense. VPC considers a successful defense only to be the private citizen killing an attacker, which ignores “subdue and capture” , “display and flee”, or wounding incidents.

    John Lott and the Crime Prevention Research Center have much better and more broad based data- available to the VPC as well but ignored by them.

    As a trainer whininstruct cops and teaches remedial handgun to peace officers, I am continually struck by a couple of things-1) most cops do not have much skill with weapons 2) in most places in the civilian world, committed citizens are better shots than most civilian cops.

    In the civilian world, the police use of weapons and police defensive engagements are significantly different from those involving private citizens.

    • Thanks for the thoughtful comment. I agree VPC is does not consider all situations for self defense which is why there were some other stats further down. Secondly I’m leery of NRA’s number’s as I think they might overestimate just as VPC is probably underestimating. I think the crime victimization survey is probably the best, but we’ll probably never know the truth.
      I also agree with your assessment of police officers. I think anyone who actively seeks information and training whether LE/MIL/ or Joe citizen will be better off than those who shoot twice a year lol. That being said, most citizens can attain a very high skill level in shooting, but knowing when you are justified and making that decision is probably harder for them.

  3. “Violence Policy Center”??? ha, ha, ha, ha, ha, ha, ha…. Tell us another funny one.

    “The true measure of the protective benefits of guns are the lives saved, the injuries prevented, the medical costs saved, and the property protected – not the burglar or rapist body count. Since only 0.1% to 0.2% of defensive gun usage involves the death of the criminal,10 any study, such as this, that counts criminal deaths as the only measure of the protective benefits of guns will expectedly underestimate the benefits of firearms by a factor of 500 to 1,000….” Suter EA.
    Guns in the medical literature–a failure of peer review.
    J Med Assoc Ga. 1994 Mar;83(3):133-48. Review.

    “As ten studies have shown, in any year, about 1 to 2.5 million Americans use guns to protect themselves and their families. and about 400,000 of those defenders believe that they would almost certainly have lost their lives if they had not had a gun for defense.[annotated in Kleck G. Point blank: guns and violence in america. New York: Aldine de Gruyter. 1991.; Kleck G and Gertz M. Armed resistance to crime: the prevalence and nature of self-defense with a gun. Journal of Criminal Law & Criminology. Summer 1995: 86., see also Lott, J] Even if only one-tenth of those defenders are correct, the lives saved by guns would still be more numerous than the lives lost to guns. The flaws in the only study to suggest otherwise, the outlier data of the National Crime Victimization Survey (NCVS), have been discussed elsewhere. [Kleck G and Gertz M. Armed resistance to crime: the prevalence and nature of self-defense with a gun. Journal of Criminal Law & Criminology. Summer 1995: 86.] Briefly, the NCVS is a study of victimization, not defense, and, by its design, undercounts the most numerous types of defensive gun use (e.g. women protecting against domestic attacks). As additional sources of undercount error, the NCVS is the only such survey conducted by law enforcement and the only study in which the respondents are denied anonymity. When any statistic, such as the NCVS count of defensive gun use, is at odds with every other measurement, it is discarded.
    Nonetheless, even those US Bureau of Justice Statistics samples show that defense with a gun results in fewer injuries to the defender (17.4%) than resisting with less powerful means (knives, 40.3%; other weapon, 22%; physical force, 50.8%; evasion, 34.9%; etc.) and in fewer injuries than not resisting at all (24.7%).11 Guns are the safest and most effective means of self defense. This is particularly important to women, the elderly, the physically challenged, those who are most vulnerable to vicious and bigger male predators….” Suter EA, Waters WC 4th, Murray GB, Hopkins CB, Asiaf J, Moore JB, Fackler M, Cowan DN, Eckenhoff RG, Singer TR, et al.
    Violence in America. Effective solutions.
    J Med Assoc Ga. 1995 Jun;84(6):253-63. Review.

    Add the multivariate analyses conducted by Lott et al. to what we knew even in the 1990’s and your underestimate of defensive gun uses withers even further.

    • I think this is probably one of the most insightful comments on the site. So thanks for the thought put into it. I agree that VPC is probably not 100 percent accurate as they have an agenda as do many other sources. I simply include it as a count of how many folks are actually killed by citizens. Honestly I don’t think it’s truly possible to know the extent of weapons used in self defense as most are likely unreported and I would imagine some of those “self defense” uses are actually brandishings. I don’t have data on that one just personal experience as an LEO. Did you have any other thoughts on the rest of the stats?

    • I agree with Jake. One of the most insightful comments on the site, or any site. However I would like to add a couple of observations. Keep in mind that there are Lies, Damn lies, and statistics. I am more aware of this than most since I spend the last 20+ years of my career working as a programmer at CDC generating statics to support researchers conclusions which were reached before any analysis of the data began. [BTW: I had my training in Statistics as part of my MBA program, where data is used to find the FACTS since the information was to be used to allocate the COMPANY’s money – not OTHER People’s MONEY as the case with Government programs. Thus the emphasis was in spotting “smoke” and bias, which made it easy for me to spot the built in bias inherent in most, if not all, CDC research.

      And why I RESPECTFULLY submit that the data in this report and in Dr. Suter’s supplemental statements are garbage, not because of any fault on their part, but because the data they rely on is GOVERNMENT data and is as fraudulent and useless as the data I spend 20+ years fudging at CDC. It is all about how you pick the “Denominator” and “Numerator” data and how you classify the data to “spin it”.

      Consider for example:

      “The U.S. rate of firearm-related homicide is higher than that of any other industrialized country: 19.5 times higher than the rates in other high-income countries.” However, as noted by the Las Vegas Guardian Express, if figures are excluded from such anti-gun bastions as Illinois, California, New Jersey and Washington, D.C., “The homicide rate in the United States would be in line with any other country.”

      Likewise, the data in the article (unless I mis read it – and to be honest I just skimmed it because I quickly reached (jumped?) to the conclusion that it was meaningless) does not break out the data based on the number of gang related shooting which are of course closely tied to Liberal jurisdictions with strict gun control and of course race.

      Also, the data is “Cherry Picked” in that it relies on actual times a firearm was used in one way or another, ignoring what can be shown STATISTICALLY as being the most important affect, the number of crimes deterred by the FEAR that the potential victim – or someone who might come to their aid.

      The best way to do that is to compare similar jurisdictions with strict gun control laws vs the opposite. But that is virtually impossible because of the extraordinarily high correlation between Race and ethnicity, violence and gun control.

      In short, libtards are trying to impose their views on White, middle class American’s using data collected from Black and Hispanic inner city slums.

  4. Re: Gunfight Statistic #6
    Were they using Obama’s Common Core math here or what?

    That said, even though 17.7% is 3x higher than the 5% the article claims, I’m still far more inclined to buy Dr. Suter’s April 19, 2017 argument.

  5. In most states “defensive display” in which the firearm is not discharged (not USE of deadly force) but merely drawn to low ready with a statement that it will be fired IF necessary is classified as “reasonable force”. The LESS restrictive rules for reasonable force then apply. Public policy favors scaring the assailant away over waiting until shooting is unavoidable and thus “necessary”.

  6. RE: Gunfight statistic # 12.
    The time period in which 36% of the felonious deaths and of assaults with injury occur is between 20:01 and 02:00 hours with the period between 04:01 and 06:00 having the least with only 4% of the felonious deaths or the assaults with injury.
    Lighting conditions during in which 20.6% of the felonious deaths and 45.3% of assaults with injury occur is good. This means that 65.9% of all incidents occur during favorable lighting conditions with only 34.1% of the situations occurring under reduced lighting. While this may seem counter-intuitive, with a large number of incidences occurring between 20:01 and 02:00 hours, 27.3% of all incidences occur under artificial light. LEOKA data analysis by Dr. Terry N. Wollert

  7. The statement that “experts are only 10% more accurate than novice shooters” is extremely misleading. If you look at the study that concludes that statement you can read their definition of what an expert, intermediate, and novice shooters are. Paraphrasing, an expert shooter has had police academy or military training. For the majority both the police academy and military don’t produce expert shooters. This is not to say that anyone who goes to the police academy or trains in the military is getting sub-par training. Typically speaking most police officers and military personnel who are “qualified” on their service pistol are still not at a S.W.A.T or special forces performance level of shooting. They are simply qualified to meet basic standards set forth by a police agency or the branch of service they are in. Next, this study says that an intermediate level shooting is quantified by anyone who hunts every year or has recreational shooting experience. Coming from a person who works at a shooting range in a state where many of the population enjoys the sport of shooting, most people who are recreational shooters don’t have any sort of knowledge on proper technique when it comes to pistol shooting. Let alone any weapon system. Finally, a novice shooter is classified as someone who has fired 0-2 times in their life. These novice shooters, obviously, without prior research or classroom experience will have no knowledge on proper shooting technique. I would lump probably most of what this study’s author would consider expert and intermediate shooters into the novice category. This study is flawed and made to believe that “expert” shooters are somehow indifferent in shooting ability from a “novice” shooter.

    • I do agree that most police folks and military members are not “experts.” That is simply the terminology the study used. The point is that for the most common shooting distances the skill advantage of training is not as beneficial as it is at further distance. I do appreciate your thoughtful comment. It’s actually pretty rare that someone gives the article that much thought.