I’ll cut to the chase. Stopping power is largely a myth. Regardless of what the internet tells you, there aren’t any magic rounds that are guaranteed to stop someone violently attacking you. However, there certainly are calibers that perform much better than others. This article is going to look at some excellent research on over 1800 real world shootings, to determine the true relationship between caliber and stopping power. Keep reading for more.
Any long term reader of this website knows that I prefer to form an opinion based on data, rather than unsubstantiated stories. This article is going to examine some excellent data gathered by Greg Ellifritz at Active Response Training. The data that Greg collected took many hundreds of hours to gather, and over 10 years to finish. Needless to say, it’s some of the most thorough data on the subject of stopping power that I’ve seen.
Research Collection and Review on Stopping Power
Greg had many of the same questions that self defense and shooting enthusiasts often have. Which calibers reliably stop threats? Which are the most accurate? And how much more lethal are shotgun and rifle rounds compared to pistol rounds?
This caused him to examine 1800 shootings where he collected data from autopsies, forensic examinations, as well as case reporting. He then compiled the data and examined it for the characteristics he thought best answered his questions. Let’s talk about his findings.
One Shot Stopping Power by Caliber
This is one of the most controversial subjects on the internet. It is often the rationalization for carrying the largest round you can possible control for self defense. The real question is, are there any concrete differences in stopping power amongst handgun calibers, and between long guns and pistols?
If we examine this chart we see that shotguns and all center fire rifles have a much higher likelihood of stopping a subject with only one round fired. Greg defined this as an assailant immediately ceasing to shoot or fight after a shot. If they were running they must have dropped to the ground within 5 feet. It should be noted that this data comes from police and military engagements. Things like accidental shootings or suicides were not counted.
We should also be aware that this metric is derived by taking the number of incapacitations and dividing it by the number of rounds a person took. We’ll look at a few other metrics on one stop shootings a little later on as well.
You can see that generally as the size of the round decreases the probability of a one shot stop decreases as well. However, we need to be careful here as we have one large confounding variable. Handgun calibers that are used in revolvers are more likely to have higher one shot stops because you cannot fire them as quickly as semi automatic firearms. This explains why the 44 magnum has a higher one shot stop percentage than the long guns. It is just slower to shoot a big, heavy recoiling revolver than it would be to shoot an AR15, or a 9mm semi auto.
The important take away from this chart is that there is less than a 10% spread between 38 special and 45 ACP. Practically speaking this may not even be statistically significant if we controlled the amount of rounds shot in an engagement.
Average Number of Rounds Until Incapacitation
This is another key metric that will tell us how much real world damage a round does to someone. This, along with our other metrics, demonstrates the realities of defensive shootings.
We can see that smaller calibers tend to require more shots until the person is incapacitated. The 9mm round has the highest number of shots prior to incapacitation. I believe this is because it is one of the easiest rounds to shoot quickly. If we had more data, we might be able to see the probability of incapacitation given different round counts.
We can see that all popular self defense rounds require between 1.5 and 2.4 rounds before an assailant stops what they are doing. It is interesting to note that the .22 cal round is the same as the centerfire rifle. While this is interesting, I agree with Gregs assertion that this is likely more of a psychological stop than physically destroying an assailant’s ability to continue to attack.
If you like fitness, shooting, and pretending to be an operator, then click here to join the Tier Three Team. It’s totally free, and over seven thousand members get the latest articles sent directly to their inbox.
Percentage of People Actually Incapacitated by One Shot
This metic is a more straightforward look at people that were stopped by only one shot. These persons were hit with only one shot in the torso or head and immediately dropped to the ground, or ceased their violent action.
Here it is obvious, even to 1911 fans, that a shotgun and centerfire rifle stop assailants much better than any pistol caliber. I was also surprised how well .357 (both Sig, and Magnum) did. I wouldn’t put much stock in the data for .32 calibers, as there were only a total of 25 cases. This is likely not as accurate as the more common calibers. The takeaway is that there’s roughly a 10% difference in one shot stopping power amongst common handgun calibers.
We’ve seen a lot of data on how well different calibers do at stopping assailants, now let’s take a look at cases where an assailant was not stopped.
Percentage of People Not Incapacitated
This metric looks at how many people continued their aggressive action no matter how much they were shot. I bet you can guess which way this trend will go!
Surprise, surprise, smaller bullets do not stop people as well. For the mouse guns, we can see that there is roughly a 1 in 4 chance that a bad guy will not stop no matter how much you shoot them. This percentage decreases to between 10-15% for common handgun calibers. We can see that centerfire rifles are in the single digits. If you truly want to defend yourself, get a rifle!
Percentage of Fatal Hits
This metric captures how many of these shootings resulted in a fatality. Remember, not every one of these shootings resulted in someone going to the morgue.
Roughly 2 out of 3 long gun shootings resulted in someone dying. For common self defense handgun calibers, there was roughly a 1 in 4 chance of causing a fatality. Interestingly enough, no caliber was lower than 1 in 5. All of these hits were to the head or torso. Things like extremity hits were not included.
Accuracy by Caliber
This metric tracks the percentage of hits that were in the head or torso. The goal is to see if any one caliber is much more, or much less accurate than would be expected.
There really isn’t a large difference here until we get to .25 ACP. This is a very small caliber, in what is often a very small pistol, that can be hard to shoot accurately. In most shootings, 75-85% of hits were to the head or torso. It should be noted that this doesn’t include completely missing the target all together.
Discussion
We need to be careful when we start to draw concrete conclusions from this data. I think Greg did a great job of compiling a lot of great information, and it tends to indicate a few key points.
First, I think it’s incredibly important to highlight the fact that there is roughly a 10-15% chance that an assailant will not stop their attack even while taking effective fire. So much of our training stops at putting a round or rounds on target, we never consider what to do when an aggressor continues their violent actions. This is why many military and tactical units train failure to stop drills where you make center of mass hits, and then transition to the pelvic girdle or head.
Next, we need to keep in mind that many of these shootings were using ball ammo. I think there is likely to be some difference if we had data on shootings with hollow points. I don’t believe it would make a double digit difference as hollow points really only create a slightly larger wound channel, they aren’t magic. But it would be interesting to see what the difference is.
Lastly, it’s plainly obvious that any reasonable self defense caliber, between .38 and .45 will suffice. It’s more important to actually train, and shoot your pistol than it is to have a 5% higher chance of achieving a one shot stop. If you are in a position where you can use a shotgun or rifle for home defense then that is always going to be more effective.
Recommendations and Final Thoughts
If you’ve read this far into the article you likely take self defense seriously. If you truly want to increase you ability to shoot and make decisions under stress then get started in USPSA, practical pistol shooting. You will learn more in your first match than you will in years of flat range shooting.
For obvious reasons pistols are great for personal defense, but they are likely to be less effective compared to a long gun. If you want an amazing self defense rifle then I recommend checking out my review of the POF Renegade AR15 pistol. It is extremely well made and very reliable. Just remember there is no amount of gear or special ammo that will trump solid practice and training. No get out there and get to work!
Like Han Solo said…
Shoot first.
Shoot often…
And try and get some shooting in between the shooting. That shit helps.
He wasn’t wrong!
hans solo was a homo
I guess no one here has ever heard of a 10mm . As a Deputy Sheriff, I carried a 9mm, until I encountered a quadruple homicide out in the “willy-wax.” The only survivor was shot FOUR TIMES IN THE FACE with a 9mm. The next day when the gun shop opened, I went to a .45 cal. 1911, OK, only 8 rounds, but it guaranteed stopping power on first hit. Now, I live in a State where we have many potential animal predators, i.e., Mountain lions, bears, wolves, coy-dogs, rabid foxes and raccoons, etc. in a mostly forested (The Great North Woods) rural state, I carry a 10mm, which nobody seems to compare in ballistic charts. The .40 caliber in German is called the 10mm kurtz (short). In my view, two legged predators are less likely than four legged ones, but I want something that will be effective for any class of predator. My experience has shown me the truth about the 9mm which in my mind is slightly better than throwing knives.
Ellifritz study was the first definitive study since the 1970s and really turned things around …ten years ago in 2011. Glad you found it Jake in 2021. Look at the FBI’s very similar findings a few years later.
I was hoping for some 10mm findings. No one has done that study.
Yeah I think it’s just to obscure of a caliber to get enough data at this point.
Todd,
I agree with you but I understand that there just has not been enough shootings where . 38 Super or 10 mill was used.
Super is too rare these days, and 10 mm is too much caliber for the majority of people to handle efficiently.
I love my Super and I load them hot for open class and even hotter for defense.
I would have to blow the dust off my storage boxes to get the exact info, but I’m pretty sure that my spicey 125 gr loads are generating about 800 lb/ft of energy at the muzzle.
I run 1911 with fully supported chamber, have no pressure issues, accuracy even out to 100 yards (just talking accuracy for this statement) is deadly and I can shoot the hell outta that combo.
I do know that I use small rifle magnum primers, and I have been running this set up for the last 30 years
My problem is that I lost too much weight and body mass from 20 years of Lyme’s disease, so an all steel 1911, nickel tungsten guide rod, bull barrel, etc just weighs too frign much for a guy that got hit hard by the Lyme’s.
I am considering picking up the new S&W M&P 2.0 10mm that just came out. I like what I see there.
My final qualifier for my above statements is; I am using the above for a carry gun in the bush for sneaky – hungry fuzzy bad guys. Yup, I have used this caliber and load on big game successfully, and shot placement and follow up is everything. Nope, I have not shot any 2 wheel drive assailants, so this is all I can contribute, and I hope I actually am contributing (?)
I tell people – that ask – that I would “only” use a steel gun – ie; 1911 if I was competing all day “only” (so, shooting a few hundred rounds a day as opossed to just trying to keep the weight down for all day / night carry)
If the primary scenario is all day carry, and not much shooting, I’ll take a polymer gun (or a light weight 1911, I am a 1911 guy “with an open mind”. That single stage trigger is simply the best for fast and follow up accuracy.
My apologies for the long winded answer, but this is my humble 2 cents worth.
Elmer Keith would happily make me eat every word with his .41 magnum, but the question you posed didn’t mention either of those two legends. I am a .41 magnum die hard as well. Another fantastic caliber, but before I dive from one rabbit hole to another….
There are a number of police departments, West of the Mississippi, that carry 10mm’s. Stats would be measurable. Considerations not able to be considered- amount of training received by shooter. Since most LE entities are only allowed by management to receive 8-10 minutes of actual trigger time training, per year. (Course of fire time, multiplied by how many times a year folks qualify, to include trigger time tactical training (plates, dueling trees, poppers, etc.) we would need to give LE, ten times their current level of training to approach the minimalist hobby shooter.
Some of my takeaways from Greg’s dataset don’t seem to get mentioned out loud:
• .380ACP is a real caliber — seems to perform fine;
• the “lead hose” seems ultimately effective as a tactic if you have capacity (I see you 9×19);
• users with low capacity (pocket guns & revolvers) cannot hose, but can make stops if they can hit.
I wonder how numbers would change with 20-round .380s, .38s, and .357s — but who would carry that?
More realistic will be to see whether the proliferation of low capacity micro 9s will improve the apparent performance of that caliber when the “lead hose” is not a sustainable tactic for a larger number of shooters. Or will “big stick” mags and team shootings continue to dominate the appearance of 9mm performance?
The problems we face with stats on shooting are…
1) Did the shooter NEED to shoot the attacker that many times? Smaller caliber rounds tend to let the shooter fire faster… and usually they do while harder kicking rounds tend to slow the shooter down. This leads to ‘machinegunning’ the attacker. For all we know the first hits may have incapacitated them yet the higher round count is factored in as necessary to stop them.
2) Did the shooter make good hits or were there many hits in non-vital areas? This would skew the results as there are many documented shootings where the one shot received 10 – 15 hits.. but only a few, if at all, were in vital areas. Same here. The higher round count (yet ineffective) is factored in as necessary to stop them.
3) Did the rounds used perform as ‘advertised’. Did the JHPs open up to desired diameters? Go to the advertised depths? Did it matter?
4) Was the attacker doped up? Enraged? That also affects the stats as those on drugs or drunk or adrenaline can soak up lots of hits.
Stopping power is not a myth… 12 gauge tends to stop ’em close to 100 percent! But stopping power is not absolute. No one can say x round stops 80% always… just does not work that way.
What we do know is… bigger rounds TEND to do better… faster rounds TEND to do better… rounds with better shapes (JHPs & SWCs) TEND to do better. The answer? The same old advice Col. Jeff Cooper said, “Carry the most powerful weapon you can shoot well.” I’ll add.. and conceal to!
From the results of the latest tests I’ve seen, I will use only one round for self defense, and that’s Federal’s HST. In 9mm, the 124 grain HST JHP. In 380, the 99 grain HST JHP.
https://www.ammunitiontogo.com/product_info.php/pName/federal-premium-law-enforcement-9mm-124-grain-hst-jhp-1000-rounds#gel-test
https://www.ammunitiontogo.com/product_info.php/pName/20-rounds-380-auto-federal-99-grain-hst-jhp-ammo#gel-test
Left out 38Super +P or 9mm Magnum, as well as 10mm, essentially a 40 S&W magnum.
Data wasn’t available for those specifics. It was not a choice.
So, the best choice would be a .380 ACP! Maybe a glock 25 or a Beretta 84FS because of its large capacity.
Its definitely not as bad as people think. I tend to prefer 9mm due to better developed rounds. But whatever floats your boat.
I’m glad to see old school still stands pretty good. 12 gauge and .45 acp. First rounds and speed do count. Hesitation gets people killed. Let’s pray we never have to find out for ourselves but, it’s better to be prepared. It’s also better to be tried by 12 than carried by six.
One criticism I have of Mr Ellifritz & your analysis is that it treats all shootings as the same. There is a very big difference between a civilian convincing the casual street robber that they made a bad choice in victim selection and say a police officer attempting to arrest a well armed drug dealer who has no intention of returning to the gray bar hotel or someone intent on suicide by cop.
The first scenario probably explains why the. 22 LR is inordinately effective despite its diminutive size.
Another issue is that the data doesn’t differentiate bullets. All 9mm shootings are treated the same. 9mm round nose bullets are not very effective. The same cannot be said for expanding bullets. I am reasonably confident that a Hornady Critical Defense bullet is much more effective than a round nose bullet. This probably explains why the 357 magnum is more effective than 9mm. It’s due to more effective bullets.
The final issue is accuracy. There are no small number of 9mm handguns that are frankly worn out and struggle to hit the proverbial broad side of the barn. Yet these are treated the same as new handguns which can easily shoot as well as 22 target pistols.
This is an interesting analysis so far as it goes but raises more questions than provides definitive answers.
You raise a lot of good points, but you have no data to support it. I’m not saying you’re right or wrong but you just have theories. For example does a bigger permanent wound channel from pistol hollow points make that big of a difference. I would guess not, but there is no data either way.
I was surprised to see how effective a .22 can be. Also surprised the .38 special outperformed the 9mm as often as it did.
So many variables to take into account. Too many to be able to infer anything at all really from the data. About all I see is, it’s better to have a gun than not have a gun…
After watching a few LE dash cam and body cam videos recently, I have a couple of observations and ideas:
1) Most police officers seem to react in much the same way – that is, point in the general direction of the bad guy, and let loose as many rounds as possible.
2) If officers had .44 Magnums instead of 9mm semi-autos, they might be inclined to be more careful with their shots, which in turn might make them more effective.
3) More shooting experience would be better – not just range time, but kill/shoot house drills, with AARs, then follow-on drills following the AARs.
Single action is always MORE accurate than DOUBLE action and smaller calibers have less RECOIL so are more ACCURATE with multiple shots but LARGER calibers make BIGGER holes and therefore cause more TRAUMA & SHOCK for STOPPING power!!!
I don’t think it quite works that way.
What is YOUR take?
There isn’t a big difference between any caliber .38 and above in terms of ballistic damage. Otherwise I tend to agree with you other thoughts.
Ultimately, you need to create massive blood loss- bad guy passes out (heart shot or major artery shot) – or hit a central nervous system structure – (spine and brain.) to paralyze him.
Shot placement is probably 90% of the equation. Yes the bigger rounds will get the penetration , but, hole in the abdomen from.a big bullet will probably have the same effect as a hole in the abdomen from a small bullet
There’s a good article by a trauma surgeon on this issue
Marshall & Sanow’s work pooled a much greater number of shootings (anecdotes, to some– hundreds and hundreds of em) and did a better job of looking at individual caliber effects by focusing on “single hit” shootings. That focus seems to confuse some people, who read into it some philosophical significance of an approach to application of deadly force. It’s not, it’s just an attempt to distill out any significant difference in the terminal effects of a given caliber.
Fackler appears to be one of “those guys” that takes everything personally, and M&S didn’t support his “deeper is better, penetration is everything” view (see how well that worked out with the short-lived LE use of 9mm 147 gr JHP in 80’s-90’s). Lab analysis is worthwhile, but when actual outcomes in life conflict with theory, some humility on part of researcher is in order.
Ellison’s numbers are apparently too small to avoid weird artifacts (like .32 ACP outperforming most of the larger centerfire auto rounds, in some aspects.
I know this is older post but; the apparent discrepancy between the ‘lesser’ calibers superior performance (22, 32 and even 380) and larger calibers lower performance MAY have something to do with different use; your smaller calibers are more likely concealed carry guns; and likely shot at very close range, i.e, someone walks up and tries to mug you; you pull your 380 and shoot them. You are much less likely to miss at 4 feet and more likely to hit them in vital area; additionally your average mugger may be less inclined to continue to push the attack; contrast this with your larger calibers which are typically ‘service’ guns, used by police; may be shooting at longer distances and less likely to hit vital area (not all ‘torso’ shots are equally incapacitating) and under circumstances, the shootee may be more motivated to continue the attack or resistance. Nonetheless, this is interesting study. There is no “magic” handgun caliber that works all the time, and while differences do exist between calibers there may not be as much difference as people imagine.
Great comment RJ. Very well thought out.