UPDATED March 7th, 2025: We have previously put out our study on the FBI’s active shooting reports. If you watch entertainment television police shows, you would think something always seems to go wrong when guns are used defensively (including shooting bystanders to getting in the way of police to failing to protect the permit holder to continually having the gun stolen and use in a crime to it being used in an accidental shooting). During the ten years from 2014 to 2023, there were 180 active shooting cases (as defined by the FBI) where a concealed handgun permit holder stopped an active shooting attack. We decided to do a deep dive to see how many cases there were out of those 180 cases where a concealed handgun permit holder accidentally shot a bystander (one case, 0.56%), got in the way of police (zero cases, 0.0%), had the handgun taken away (one case, 0.56%), and got themselves killed (two cases, 1.1%). What was more common were cases where the permit holder was injured in saving the lives of others (44 cases, 24%). Fifty-eight of those cases were instances where a mass public shooting was likely prevented. An Excel file with the data for civilians and police is available here.
While civilians with concealed handgun permits stopped 51.5% of the active shootings in non-gun-free zones, police stopped 44.6% of the cases (124 arrested or killed by police, 32 committed suicide when police arrived = 156/350 = 44.57%). Interestingly, police officers were much more likely to lose their lives or be wounded in stopping these attacks than armed civilians. Twenty-seven officers were killed in 19 attacks (7.7% and 5.4%, respectively). That is 5.94 times the rate that permit holders were killed. One hundred officers were wounded in 48 attacks (28.6% and 13.7%, respectively). That rate is 17% higher than for civilians. In four cases, the police shot and killed the wrong person — twice they accidentally shot fellow police officers (Prince George’s County Police Department District 3 Station on March 13, 2016 and Borderline Bar and Grill on November 7, 2018) and twice they accidentally shot civilians (Galleria Mall in Hoover, Alabama on Nov. 23, 2018 and Highlands Ranch, Colorado on May 7, 2019). The bottom line is that the rate of police shooting the wrong person is very low, though it is slightly more than twice the rate that civilians shoot a bystander (1.14% versus 0.56%). The police accidentally shot other police officers at very slight higher than the rate that civilians shot bystanders.
We have also examined the data to determine the sex of the permit holders who stopped these attacks. Eighty-one percent of the permit holders were men, 6% were women, and 13% were not mentioned in the news reports on the cases. We have included a graph with the raw numbers at the bottom of this page.
| A Deep Dive into Cases Where Civilians Stopped Active Shooters 2014 to 2023 | ||
| Number of the 180 cases that had this result | Percent of cases | |
| Accidentally shot a bystander | 1 | 0.56% |
| Got in the way of police | 0 | 0.00% |
| had the gun taken away from them | 1 | 0.56% |
| Civilian stopping the attack is shot and wounded | 44 | 24.44% |
| Civilian stopping the attack is shot and killed | 2 | 1.11% |
| A Deep Dive in Cases Where Police Stopped Active Shooters 2014 to 2023 | ||
| Number of the 350 cases that had this result | Percent of cases | |
| Accidentally shot another police officer or wrong civilian | 4 | 1.14% |
| Got in way of Police | N/A | N/A |
| had the gun taken from them | 0 | 0% |
| police are shot and wounded | 100 | 28.57% |
| Police are shot and killed | 27 | 7.71% |
This is an important retrospective report on the Good Samaritan. We need to know that they accomplished their goal of saving others, but they did have a significant 24% chance of being injured themselves.
As the data shows, armed civilians are more effective at defending themselves and the public than the police in certain situations. This makes sense, as individuals who carry firearms for self-defense might undergo more specialized training than police officers.
A study could determine whether this hypothesis is correct: Are concealed carriers better trained than police officers?
They seem to be too categories of severe who concealed carry. Those who receive a considerable amount of training and those who receive none. The statistics listed above would improve if more people would receive some defensive handgun training. “You never rise to the level of your expectations, you fall to the level of your training”. Archilochus
There seems to be two categories of those who concealed carry. Those who receive a considerable amount of training and those who receive none. The statistics listed above would improve if more people would receive some defensive handgun training. “You never rise to the level of your expectations, you fall to the level of your training”. Archilochus
I’m not sure if civilians who carry possess more specialized training. What would that be, exactly? If anything, one might attribute the differences to the element of surprise. A civilian, carrying concealed, in plain clothes and not pulling up in a marked car with sirens announcing their arrival, will usually have a better opportunity to stop the perpetrator.
I don’t know if the studies would bear that out, but it seems to be common sense.
There is an essay, “Nation of Cowards” by Jeff Snyder, that points out that it makes sense civilians are less likely to shoot innocent bystanders than cops, because there’s no ambiguity as to who’s the bad guy when someone pulls out a weapon and says “you’re coming with me!”, whereas cops have to come in from outside the situation and do their best to assess what’s going on, which is much more prone to error.
I strongly encourage everyone to read that essay: it’s the essay that pushed me from “statistics say it’s best to be armed” (thanks, John Lott, and “More Guns, Less Crime”) to “damn the statistics, I have a philosophical and moral duty to carry a gun, and be trained in its use!”.
I would surmise that an information disparity between bystander, who will often have witnessed the incident from the beginning and is thus in a better position to know who are the “good guys” and who are the “bad guys”, and later arriving law enforcement who encounter a “fog of war” scenario, might be at the root of the matter.
The reason almost certainly is not that the non-police civilian (bystander) has better training or experience, but that they are there when it happens and are therefore inside the attacker’s OODA loop from the start. Civilian police show up with a great deal of fanfare that gives the attacker plenty of time to prepare for their arrival. In fact, the attacker has almost certainly had police arrival and intervention worked into their thinking since they began planning their attack days, weeks, or months prior. ‘Speed, surprise and violence of action’ are what win gun fights, and the police are short on all three when responding. The the bystander has all three.
I suspect is more circumstances. Police are going to be responding to an incident with incomplete information. A bystander is likely to have witnessed the start of the incident and have a clearer picture of what is going on.
That and a bystander will not present as a threat to the shooter the way a cop will.
If I had to guess, I’d attribute much of the relative success of civilians to the element of surprise. Shooters will likely not anticipate the presence of an armed civilian, giving the civilian a significant advantage. Further, a shooter would likely prioritize a uniformed officer as the greater immediate threat.
“Are concealed carriers better trained than police officers?”
Maybe. I think civilians are more likely than police to face consequences for screwing up.
I suspect it’s more likely the case that the concealed carriers who intervene are more highly trained. We have no way of knowing (at least from this data) how many total concealed carriers were actually present at these incidents. We only have a count of how many were sufficiently confident in their skills/experience to engage the shooter. There may well have been concealed carriers who elected to either retreat or take a defensive position instead.
It also seems possible that police are wearing a big blue flag that says “shoot me”, whereas an armed civilian has the advantage of surprise and “hiding in plain sight”. There may also be a B-17 fallacy going on here. Police likely end up disproportionately stopping the most severe events where shooters are better prepared, better covered, and better armed.
LEO are duty bound go towards threat ccw holders have the option to let the threat come to them.
Most likely, no, it’s not a matter of the civilians being better trained. I’d be willing to bet that the cops pushing forward or having to search for the shooter when the civilian doesn’t have to is more likely the reason.
While your hypothesis might be true, it occurs to me that there are other possible factors.
We are looking at two groups of events (shootings stopped by civilians vs shootings stopped by police) and assuming that the only difference is who stopped them, otherwise they must be homogenous, but that is not necessarily the case.
It is plausible that at least some of the events stopped by civilians were significantly different from those stopped by police from the get go. If, as an example, some situations are initially beyond the scope of what a civilian could stop then those situations would never be part of the first category and become part of the second category.
Or, another possibility could be that situations tend to change over time. One would guess that in general situations resolved by civilians are more quickly resolved since the civilians are there from the beginning while the police generally have to be summoned and then arrive. Things could become more dangerous or chaotic the longer a situation lasts.
Related to the last point, in general I would expect a civilian who had been present at, or near, the beginning of an incident to have better knowledge of what is going on, who and where the bad guy is etc. than an officer who just arrives on the scene.
Lastly, I would guess that an armed civilian is far more likely to have the advantage of surprise than a police officer.
Those are just a few items I can think of that might cause a difference in the data sets without showing that better training is related at all. Once again, not saying you are wrong, just pointing out that there are a whole lot of possible explanations and figuring out the true ones can be difficult.
Really interesting statistics.
Thank you!
Could be. Or it could be the civilian is already embedded in the situation and hence less noticeable until he/she starts shooting while the LEO, coming in later and in uniform, is a better target.
Interesting data.
Question: There is pretty clearly an increase in the fatality rate when police are involved. That might be simply due to the fact that the police are involved, or it might be because the civilian permit holder recognized that the situation is too dangerous for him to act, thus leaving the police to deal with inherently riskier circumstances?
Is it possible to control for that?
I applaud the research and agree with the conclusions. I would note a few reasons for the anomalies. These mainly stem from the facts that the police officers generally arrive on scene after the shooting incident is well underway (someone calls it in, they rush there), and they are instantly identifiable as police officers.
A civilian on scene generally sees the incident start, has at least some time to size it up, and already knows who the target is. The civilian also has the advantage of surprise, both because the shooter will not have identified him in advance as a threat; and also because it is often very difficult to tell where unexpected fire is coming from, doubly so when the shooter is distracted because he is busy identifying victims he wants to shoot. And often the civilian is close by making an ambush shot possible and increasing its accuracy.
Police, on the other hand, show up later, so they have the initial task of figuring out who is who and what is happening and whether the threat is still active. They are generally further away from the shooter and have to close the distance in order to get an accurate shot. Conversely, the shooter will have more time to identify the incoming defenders and will be able to identify his new targets instantly.
As a result, it’s no surprise that cops are more frequently shot and make more mistakes. They have inherent disadvantages when it comes to mass shooting interventions. We should accordingly encourage more civilian defenders, not discourage them.
I wonder about the effect on these stats of police wearing uniforms. A shooter with limited ammunition might prioritize shooting at the guys in blue while an armed civilian is more likely to have the advantage of surprise.
Officers getting killed or wounded twice as often as civilians is most likely due to them being a more obvious or expected target, generally being in uniform.
Interesting, with the small number of incidents the numbers for professional responders and civilians are statistically identical (none of the differences having any hope of being outside he margin of error) with the likely exception of the Police being killed by the perpetrators.
Since most of the responding police were likely wearing body armor and the civilians almost certainly were not, this is even more interesting. I can see two possible explanations (acknowledging that I may be missing better explanations).
1) Police since they are responding after a significant response time that allows the perp to have a better position to get the drop on the Police while the civilian responder is a surprise disruption of the perp’s plan.
2) Police try harder to resolve the situation without killing the perp putting themselves at greater risk.
I disagree with your 2nd conclusion. We are seeing a quantum leap in police aggression and disregard for deescallation in police encounters in this country, they are the hammer and everyone they encounter are nails.
I do believe concealed carry holders are better trained than the run of the mill police officers ona beat.
I’m a firearms instructor and the people that come to my classes are planning on /or have taken more classes. They tend to do dry fire training, invest a lot of time and money from their own pockets. And many of them come from families where they had or were shooting guns from a very young age.
My students also tell me they are taking class so they can not only survive themselves, but feel led to protect core and extended family and friends. They prepare emotionally, mentally, physically, financially and strategically daily.
I’ve encountered so many police officers , especially in small Departments, where training is both economically not feasible and the Officers themselves are not prepared.
It could be that one reason that police officers are more likely to be killed or injured stopping active shooters is that the nature of police officers’ duties make them more likely to attempt to stop a shooter even at risk to themselves. Armed civilians may be more likely to attempt the stop only if they perceive less risk to themselves. No one is going to shame a civilian who did not attempt to stop an armed shooter, where the same is not true for police officers. I’m not saying that this is, in fact, a contributor to the disparity, but it may be.
Police are much more likely to have armor which matters a lot if the shooters weapon is a handgun. On the other hand, police training for active shooters is to move to the sound of the guns putting them more at risk.
The fact that police were more likely to be killed stopping a mass shooting event seems to indicate that they are more likely to take risks than civilians. They are also more conspicuous as threats to the shooter, obviously. Do give the police some credit.
I disagree they are more likely to take risks. In the Uvalde school shooting in Texas, several civilians attempted to engage the shooter while almost 100 plus officers held them back, and those same 100 plus officers refused to engage the shooter themselves, even though they had him out numbered 100 plus to 1. Officers also tend to do “mag dumps” when they do engage.
Officers should be better trained on rules and strategies of engagement, have body armor, know how to find and use cover and concealment. That is their job. So the meme that they underperform on calls to active shooters doesn’t hold water. They know in advance, they are likely going to a gun fight and will have backup, are as, or better armed than the perps. The concealed carry holder has none of those things. They are thrown into a gunfight “on demanded” behind the curve. All they have to rely on is their mental, emotional, and tactical training, and a lot of luck.
I disagree they are more likely to take risks. In the Uvalde school shooting in Texas, several civilians attempted to engage the shooter while almost 100 plus officers held them back, and those same 100 plus officers refused to engage the shooter themselves, even though they had him out numbered 100 plus to 1. Officers also tend to do “mag dumps” when they do engage.
Officers should be better trained on rules and strategies of engagement, have body armor, know how to find and use cover and concealment. That is their job. So the meme that they underperform on calls to active shooters doesn’t hold water. They know in advance, they are likely going to a gun fight and will have backup, are as, or better armed than the perps. The concealed carry holder has none of those things. They are thrown into a gunfight “on demanded” behind the curve. All they have to rely on is their mental, emotional, and tactical training, and a lot of luck.
I disagree they are more likely to take risks. In the Uvalde school shooting in Texas, several civilians attempted to engage the shooter while almost 100 plus officers held them back, and those same 100 plus officers refused to engage the shooter themselves, even though they had him out numbered 100 plus to 1. Officers also tend to do “mag dumps” when they do engage.
Officers should be better trained on rules and strategies of engagement, have body armor, know how to find and use cover and concealment. That is their job. So the meme that they underperform on calls to active shooters doesn’t hold water. They know in advance, they are likely going to a gun fight and will have backup, are as, or better armed than the perps. The concealed carry holder has none of those things. They are thrown into a gunfight “on demanded” behind the curve.
Missing the most important element for comparison: time of interaction.
Cops are a late reply to a situation where as a CCer is most likely imbedded in the initiation of the situation.
Not to mention the element of surprise. A CCer doesn’t roll into a situation with lights flashing and sirens blaring.
Thanks for this research, John. It confirms what we already knew, for roughly the last several thousand years: that an armed population is the best deterrent and defense against criminal acts.
I love the study. Now if the ANTIS could read
Ironically, NBC’s The Black List, which is usually anti-gun and has entries in the media bias section, had an episode with a successful defensive gun use. An assassin tried to kill a young woman in her house, and she drove him off with a shotgun.
In fairness, some factors not mentioned have a strong impact in relation to police officers having more negative statistics:
Armed civilians can choose whether or not to intervene. Police officers are obligated to intervene, whether to their advantage or not.
An armed civilian is often not an obvious threat to the shooter and may take the shooter by surprise. Police officers are procedurally trained and obligated to identify themselves. Their uniforms alone make them more of a target.
Thus an armed civilian may have significant advantages over a responding police officer.
One of the articles from Greg Ellifritz’s “Weekend Knowledge Dump” from Feb 28th was about the greater likelihood of police being shot OUTSIDE the event (on arrival) as compared to “in the fight” and the need for better training for the response and approach. This suggests that many of the earlier comments are correct: that the CCer has surprise vs the Police arriving so conspicuously.
Definitely believe that which you said. Your favorite reason appeared to be on the net the simplest thing to be aware of. I say to you, I certainly get irked while people consider worries that they just don’t know about. You managed to hit the nail upon the top and also defined out the whole thing without having side effect , people could take a signal. Will likely be back to get more. Thanks