FBI Ammunition Test Protocol & Relevancy to Self Defense

FBI Ammunition Test Protocol

Rancorous debate often accompanies the topic of selecting a self-defense cartridge. Everyone has an opinion and cites various authorities to bolster their positions.

Perhaps the most frequently used standard in the best caliber for self defense debate is the FBI ammunition test protocol. But, is the FBI testing method good for predicting effectiveness of ammunition? If so, does this ammo test hold the same relevancy to the general shooting public that it does for the federal law enforcement agency?

What is the FBI testing protocol?

Alternatively called the FBI penetration testing, the ballistics test set can help gauge law enforcement bullet terminal performance in a standardized medium. This allows for a bullet penetration test that is, in theory, repeatable and useful for establishing a known standard of performance in controlled conditions.

A standardized test allows for the direct comparison of bullet performance. Using this test, a law enforcement agency can compare things like hollow point bullet penetration and expansion between two different cartridges before selecting which ammunition to issue its officers.

This is advantageous for anyone who will have to justify the costs of duty ammo to a budget committee or agency head. Likewise, it is a practical place to start for someone looking for the best performing duty ammunition.

Federal HST in Ballistic Testing
Federal HST has an excellent track record with law enforcement agencies and in lab testing. Are we certain it is the best choice for armed self defense?

The testing itself requires multiple rounds of the ammo load to be shot into a specific mix of ballistic gelatin. Some of the tests are conducted with intermediate barriers placed between the shooter and gel. Bullet velocity, penetration, expansion and weight retention information is collected.

What are the specific ammunition tests?

As I mentioned previously, the FBI ammo testing requires the use of a standardized medium: ballistics gelatin. This ballistic gel mix is a 10% ordnance gelatin formulation that is chilled to form blocks. This formulation is used as the gelatin is similar in consistency to human muscle tissue.

For handgun load testing, bullets are fired into a ballistic gel block from a distance of 10′ and data is collected about how each load performs. There are six (6) different tests conducted by shooting ballistic gel. They are:

  • bare gelatin, no barriers
  • heavy clothing on gelatin, no barriers
  • light clothing, steel barrier
  • light clothing, wallboard barrier
  • light clothing, plywood barrier
  • light clothing, automobile glass barrier

Each of the barriers and clothing choices are made to simulate shooting circumstances that FBI agents have encountered or could reasonably expect to encounter during a lethal force confrontation. For the testing, each of the barriers and clothing articles are described in exacting detail to enhance the repeatability of the testing.

At the time I am writing this article, the heavy clothing is defined by the FBI protocol as four layers of clothing that might be encountered during winter conditions. The layers are:

  • cotton t-shirt material (~5.25 oz/yard, 48 threads per inch)
  • cotton material (~3.5 oz/yard, 80 threads per inch)
  • Malden Mills Polartec 200 fleece
  • cotton denim (~14.4 oz/yard, 50 threads per inch)

Light clothing is defined as two layers:

  • cotton t-shirt material (~5.25 oz/yard, 48 threads per inch)
  • cotton material (~3.5 oz/yard, 80 threads per inch)

Outside of the Polartec fleece, no other nylon or synthetic materials are used for the clothing portion of the testing.

In the FBI protocol, bullets that fail to penetrate 12″ or penetrate beyond 18″ are penalized. Bullets that expand and penetrate between 12″ and 18″ of gelatin are generally considered acceptable. There is a complex formula for analyzing performance for suitability in a law enforcement context that is beyond the scope of this article.

Performance Does Not Equal Effectiveness…Or Does it?

The FBI testing protocol quantifies a bullet’s terminal performance in a consistent medium. The tests are designed to be repeatable by anyone with the resources to purchase, mix and chill the gelatin. Results from the testing can then be used to compare one bullet to another in terms of its performance in the testing.


The FBI tests do not predict bullet effectiveness.

More than a few of you will stop and reread the previous statement. This is because the FBI tests are frequently used as “evidence” that one round is better than another. Yet, this is simply not true.

The FBI tests only measure a bullet’s performance in a controlled setting and medium. You may be able to correlate performance “on the street” with performance in gelatin, but there is no direct relationship there. In other words, a bullet’s score on the FBI test does not equate to a certain level of performance in real life.

“I don’t think any test can be a predictor of performance other than ‘the bullets that seem to work well tend to meet these test protocols’,” said firearms expert Grant Cunningham. “Beyond that, it’s really generalization. For example, expanding bullets tend to be more effective than non-expanding bullets all other things being equal.”

According to Cunningham, people have misconstrued the FBI testing to support personal preferences in ammunition selection.

“This is why we get dogmatic assertions of bullet effectiveness, such as ‘If it doesn’t penetrate at least 12 inches, it’s not suitable for self defense’,” said Cunningham. “We also still hear predictions of ‘effectiveness’…based on those…measurements, which the tests were never designed to determine. The issue, then, isn’t the tests themselves…the problem is getting people to understand what the tests are and what they do (or don’t do.)”

Based on largely anecdotal data, the handgun ammunition that performs well in the FBI testing tends to perform well in violent encounters. Nevertheless, this is a correlation only and you cannot reasonably extract a predictive formula from it.

There have been a number of significant studies on real world performance of ammunition. Perhaps one of the most famous – and most controversial – was the long term look at handgun stopping power conducted by Evan Marshal and Edwin Sanow. More recently, Greg Ellifritz took a stab at compiling and analyzing data on firearm performance in the real world.

Though the details are different in all of the studies I have seen, the results tend to suggest that:

  1. there is a minimum level of cartridge power to see reliable performance from a handgun,
  2. expanding bullets tend to perform better than non-expanding bullets,
  3. long guns perform much better than handguns, and
  4. no previous study of actual shootings provides reliable data to make specific predictions on the performance of any specific cartridge.

None of the studies I’ve read are able to provide clear evidence that the FBI testing protocol is an accurate predictor of the elusive thing called “stopping power.” Unfortunately, the FBI testing protocol seems to be the best tool we have for studying potential in controlled conditions.

Is FBI penetration testing useful for self defense?

The FBI penetration test is specifically designed to measure the performance of law enforcement ammunition. It was not designed to test for the best self defense ammunition, nor have I ever seen any claims from the agency that it is a good test for measuring self defense loads.

bullet test in steel
This bullet was recovered after FBI testing though rolled steel. Is it a good indicator of performance for citizens in self defense scenarios?

Nevertheless, the FBI testing is the only widely adopted standard for measuring the performance of handgun ammunition. It is completely understandable why it is the de facto standard for self defense ammo performance even though it wasn’t designed to be it.

While the FBI ammunition test protocol is good, I don’t believe that it is ideal for self defense ammunition. Simply put, what works for law enforcement may not be good for an armed citizen protecting himself against attack.

Active Self Protection is a YouTube channel hosted by firearms instructor John Correia. Currently, the channel has more than 1,000 real-life self defense videos posted. Most of these depict armed citizens defending themselves against attacks.

Throughout these videos, several things become obvious as it relates to the differences between common self defense shootings and law enforcement shootings. While law enforcement may deal with barricaded subjects on a semi-regular basis, there are few instances where armed citizens are shooting through steel, walls and laminated auto glass.

From my own observations – both reviewing shootings caught on film and participating in law enforcement investigations – armed citizens are most frequently defending against attackers who are not shooting from cover.

If you accept that the idea that a bullet penetrating 10% ballistics gelatin to a depth of 12-18″ is more likely to stop a determined attacker than one that does not, then loads developed to perform well in all of the FBI testing will likely be adequate for self defense also.

However, since 2/3 of the current FBI test for handgun ammo is skewed toward defeating hard, intermediate barriers, and self defense shooters are not likely to need this kind of ammunition, does the FBI testing actually give us a good view of what makes the best self defense ammunition?

“[FBI tests three through eight] are unimportant in the private sector,” said Cunningham. “They may be important to people whose job is to arrest other people, which is where barrier penetration becomes an important characteristic. We don’t, and generally shouldn’t, need to worry about that.”

Do we need a better self defense ammunition test?


If the question is “do we need a better test that can reliably predict cartridge performance in the real world?” then the answer is most definitely yes. However, such a test seems to be beyond the community’s grasp. Such a test may never be developed.

However, if the question is merely “would private citizens be better served by another test instead of the FBI protocol?” the answer is “possibly.”

In my mind, the bias toward barrier penetration in the FBI protocol skews ammunition development away from bullet designs that might have the potential for better terminal performance. Several self defense experts I have spoken to hold a similar opinion.

Instead of six tests with a bias toward defeating hard barriers, I envision two tests. The basic two would look like this:

  • bare gelatin – A basic test that acts as a baseline for bullet performance.
  • clothing over gelatin – This would consist of a two or more layers of clothes such as a heavy sweatshirt cotton with a t-shirt cotton or a synthetic material similar to a Nike Dri-FIT underneath.

I would also like to see the use of a temperature stable testing medium. The FBI’s gelatin choice requires significant logistics complications which makes independent testing of ammunition difficult. If there is anything I’ve learned in the testing of flashlights, independent testing is exceptionally valuable for consumers.

Some flashlight manufacturers appear to game the specifications to create a performance impression that my testing shows isn’t true. I believe independent testing of ammunition is equally important.

Independent testing would allow for better consumer education about how well loads perform out of compact pistols. According to firearms trainer Paul Carlson:

If the average Joe is carrying a Glock 43, he needs to know that ammo performance may not replicate the test results from a duty size gun.

Clear Ballistics is a popular testing medium for many people conducting independent ammo testing. It does not need to be refrigerated and can be used in both cool and warm environments. While it is similar in performance to 10% ordnance gel used in the FBI testing, it is not the same. Testing done with Clear Ballistics should not be compared directly to FBI test results.

However, tests using Clear Ballistics can be compared to each other. A few years back, Lucky Gunner did a detailed analysis and test of popular self defense handgun loads using the Clear Ballistics testing medium. While the data is not suitable for direct comparison to any testing done in accordance to the FBI testing criteria, it serves as an excellent comparison between loads in the testing.

Perhaps just as importantly, Lucky Gunner published its testing standards. This allows other people to use the same standards with the Clear Ballistics product to replicate the testing. People testing the loads using the same criteria can help confirm the validity of the Lucky Gunner testing.

I am not advocating for Clear Ballistics, simply using it as the most obvious temperature stable alternative to the FBI’s gelatin preference.

If independent testing is not a concern, Cunningham may have the easiest approach.

“I don’t think we need new testing protocols. The problem is getting people to understand what the [FBI] tests are and what they do (or don’t do), developing a partial weighted score from the first two tests in the suite, and then getting everyone to agree to use them.”

Final Thoughts

I learned a long time ago to not accept conventional wisdom without verifying the information. Many people suggest that the FBI testing is the best way of identifying the optimal self defense ammunition. That may be true, but has anyone really developed an alternative evaluation method that could identify superior products for the armed citizen? Something that is measurable and not marketing hype?

As a starting point, the FBI testing protocol is a reasonable place to begin a search for quality self defense ammo. However, the standard was developed for users that are in different circumstances than most armed citizens who carry a gun for self defense.

A new set of standards that is designed for self defense – not law enforcement – should at least be considered. If developed, I believe the standards should use a temperature stable ballistic medium and test for the most common things an armed citizen may encounter. The tests should be designed for easy replication by all shooters with locally available materials.

What are your thoughts? Think I’ve missed the point, or hit on something worth exploring? Please leave your take in the comments section below. I know people can get passionate about calibers and stopping power – and that’s great – just keep things civil. I run a family friendly site and welcome all shooters.

Thanks to Federal, Hornady, Speer and Winchester for the use of their photos in this article.

By Richard Johnson

Richard Johnson is a gun writer, amateur historian and - most importantly - a dad. He's done a lot of silly things in his life, but quitting police work to follow his passion of writing about guns was one of the smartest things he ever did. He founded this site and continues to manage its operation.

37 replies on “FBI Ammunition Test Protocol & Relevancy to Self Defense”

First, I will agree that a private civilian test protocol needs to include clothing. I would argue that it needs a heavy clothing stage given that many of us live in cold climates. Now, the question becomes how much do you weigh the bare media results versus the clothed media results? If both light and heavy clothing are tested, then what is the weight between the two results?

It seems that the FBI included the bare gel results as a control to compare their results against tests from other researchers. From the FBI results that I have collected, I have noticed that the only time that ammunition penetrates less in clothed testing than in bare gel is when the bullet expands so quickly in bare gel that it sheds its mushroom, leaving just enough of the core remaining to achieve adequate penetration. The clothing tests of the same ammunition delays the bullet’s expansion just enough that the mushroom remains attached, reducing the penetration.

Now I will mention the hard barrier tests for the “puzzle putters” as Ed McGivern dubbed them. It is fairly safe to rule out the two front windshield auto glass tests (FBI #6 and #8) for private civilian testing. One could argue that you might need to shoot through window glass from inside your car in say a carjacking scenario, but more often not, wouldn’t that involve shooting through a side window? Side windows are not constructed the same as the front windshield. Even shooting through one’s own windshield from inside the car would present a different test configuration than the FBI’s two tests.

The car door sheet steel test (FBI #3) has always seemed a bit iffy even for law enforcement. Car door construction is so variable given the thickness of door panels and the placement of internal brackets, the lock mechanism, and the window mechanism. Plus, if someone was presenting a threat from inside a car, wouldn’t the window be down in many cases? That would add yet another layer of material for the bullet to penetrate.

The wallboard test (FBI #4) has its own issues as given the interior layout of houses and buildings. An aggressor taking cover behind a doorway likely has more than just two layers of wallboard between you and them, such as framing timbers.

The plywood test (FBI #5) might be applicable if the aggressor was taking cover directly behind an opened interior door.

Hi Daniel,

Your insight, as always, adds a lot to the discussion. Thanks for taking the time to share it.

Heavy clothing as used in the FBI testing is fine, but I wonder what effect a different composition of clothing would reveal (if anything.) The current “undershirt, dress shirt, fleece, denim” layering is understandable, but is there a better representation for what people are more likely to wear? Also, while the popular idea of ‘a bullet’s performance against thinner clothing will always be as good as against thick clothing’ is probably correct in most circumstances, has that been tested & verified? With modern man-made fibers?

Side note – do you have any documentation/links to docs on why Fackler determined 12+” in 10% gelatin was the preferred penetration depth? I’ve read several of his papers where he makes this reference, but never an explanation on how he got there.


I’ll have to dig for the exact reference, but it was based upon Fackler’s live animal testing at the Letterman Army Institute of Research (LAIR). He concluded that 10% ordnance gelatin duplicated the penetration achieved in live swine rear leg tissue within 3%. The previous DOD and NIJ testing standards used 20% ordnance gelatin.

I read the original requirements of 12″ minimum back in 2010 and to paraphrase…

The 12″ minimum was determined to be the minimum distance required to reliably penetrate to a vital organ on a human target that would cause immediate incapacitation of the target and reduce them to a state of non-violence.
In short, 12″ was the minimum distance that was needed to penetrate the heart from pretty much any angle of entry into the human torso, which is one of the two vital organs that need to be ventilated to stop an aggressor that may be on a myriad of different controlled substances. (The other organ being the brain, which is ballistically easier to incapacitate but tactically more difficult to target.)

Hi Jaster,

Thanks for taking the time to respond. My question is how was the 12″ determined to be the minimum?

Also, isn’t that skewing the performance toward the least likely shots that might be made? If, for example, research indicated that 80% of all defensive shootings occur within a 30Ëš angle of the front or back of the attacker, should we develop minimum standards on the less than 1% of shootings (or whatever the % might be) that are at the most extreme distance from the edge of the torso to the heart?

Perhaps the difference between the mean and extreme are so small that it makes sense to simply build for the extreme. But without the data, it’s tough to know.

Please understand that I’m not advocating for or against a 12″ minimum penetration standard. Rather, I’m trying to find and understand the research that led /some/ people to believe that it was needed while other contemporaries seemed less convinced by it.



Great write-up on this. My take away is that the necessary size and load of ammo may be less than those that pass the FBI tests. Number one factor in effectiveness is still hitting your target! Practice, practice, practice.

I don’t think you’ll find a claim anywhere from the FBI that this test is the perfect solution.

Their first reaction to the 1986 Miami shoot-out was the .40 S&W. Once cooler heads prevailed they came up with this test, because they recognized neither cartridge nor caliber matter as much as does projectile performance.

The greatest significance of this test is that (some) bullet manufacturers now build to a common standard, and one that has at least a minimal basis in scientific testing. Which is precisely why the 9mm is experiencing a resurgence. Because they’re making bullets better optimized to the characteristics of the cartridge, and the new testing allows for apples-to-apples comparison of differing cartridges and calibers.

So the FBI has stated a theory and begun testing in the form of real-world shootings. Whether they continue to follow some form of a “scientific method” will depend on whether they evaluate the outcomes of the testing and then adjust the initial hypothesis accordingly. I think it’s reasonable to presume that they are evaluating the real-world effectiveness of bullets that they themselves in essence designed. I know others are, particularly Dr. Gary K. Roberts (AKA DocGKR), military officer, medical professional and gun guy whose work is posted in many tactical and LEO websites.

But there’s no question this test is a step forward, and the fact that the FBI took that initial step shows they recognized a need for change. One would hope that they will tweak the hypothesis if the bullets produced by it prove less than optimally effective. As always, time will tell. But even as is, flaws and all, this test is still better than anything that came before it.


Thanks for reading my article and taking the time to post your thoughts. Believe it or not, I’ve seen a lot of people use the FBI test as the only standard by which to judge ammunition, that it doesn’t need any adjustments, and that it equates to a predictor of “stopping power.” I am skeptical of all of these claims.

If I recall correctly, the FBI transitioned first to the 10mm. Eventually, the agency moved to the .40S&W when it was developed as a less powerful version of the 10mm in the early part of the 90s. By that time, I think the gelatin testing standard was already in place – though an earlier version of it.

The FBI didn’t transition back to the 9mm until 2015/2016, well behind what many people in the law enforcement and self defense sectors were doing. The agency appears to be pretty well entrenched in the .40 suggesting that they were invested in the larger cartridge/bullet story. But the agency had hung a lot of the blame for the ’86 disaster on the 9mm Winchester Silvertip – carefully avoiding many questions about the training, tactics, preparation, and equipment used by the FBI in that incident. Who knows how much keeping the .40 was a terminal performance issue vs. a “saving face” one.

I am unaware of any tracking or research the FBI has done to specifically correlate its gelatin testing with the performance of self-defense/law enforcement ammunition performance in the field to improve the testing. While I agree that it reasonable for them to do so, it doesn’t mean the agency has. I’ve seen the FBI do many unreasonable things. So, I make no presumptions there. If someone has a link to share with us on this, I would greatly appreciate it.

While I agree that the FBI test is better than shooting wet phonebooks or clay, I’m not convinced it is the best way to measure the potential effectiveness of a cartridge for self-defense purposes.

It may be the best test that can be devised. But without questioning presumptions and probing for facts, can we really be sure?

Thanks again for reading!


“…If I recall correctly, the FBI transitioned first to the 10mm….”

I stand corrected. I left out the FBI-Lite 10mm load. The FBI never issued a full-house 10mm, it was downloaded from Norma’s original load (200gr@1181 fps, according to Norma’s literature) to 180gr@950 fps. Which yielded substantially less FPE and also left quite a lot of unused case capacity. S&W simply cut the case back and created the ballistic twin to the FBI’s lite load and rechristened it the .40S&W. It had roughly the same FPE as the 10mm Lite but could be shot in a 9mm-sized frame.

But you’re still blaming knuckleheads for making too much of the FBI’s new protocol, which is none of the test’s doing.

You are correct – I am saying that a lot of people misinterpret what the FBI tests actually measure.

I am also suggesting that the FBI tests may not be the best tool for evaluating self defense cartridges. The FBI testing protocol may be “good enough” and that’s fine. Certainly a lot of people who actually understand the tests think so. But without asking questions and challenging assumptions, how are we to improve?

Without a doubt, we are better with than without the existing testing protocol.


Why did the FBI pick 12″ as the minimum for penetration and 18″ for the maximum? Unless somebody pulled numbers out of the air, the implication is that a bullet that penetrates less than 12″ into gel is unlikely to penetrate deeply enough into a human assailant to stop him reliably. Similarly, a bullet that penetrates more than 18″ into gel is likely to exit the assailant and have no more effect on him.

Hi Kendahl,

Thanks for taking the time to read the article.

You start with the hedge of “Unless somebody pulled numbers out of the air…” Therein lies the problem. Without asking questions and examining data, how do we know the numbers were not pulled out of the air? Suggesting that the FBI wouldn’t just make up the numbers because it is the FBI is merely an appeal to authority, and not useful.

I know you didn’t make that argument, but other people have. And that’s where my skepticism comes in. I’m just looking for the data that correlates the penetration depth with…well, anything.

I don’t claim to be an expert – just someone asking honest questions. If we are going to make an assumption that the 12″-18″ penetration of 10% ballistics gelatin is “good” – let’s talk about (1) why and (2) what it can be used to predict (if anything.)


As for the FBI’s determination of the 12-18″ limits, The FBI has said in descrbing the protocols that testing shows this level of penetration in 10% gel represents an overall approximation of the distances a bullet entering a human body from different angles would have to travel to reach and damage vital organs and the CNS. The IWBA independently established a mimm of 12.5 inches which agrees with the FBI minimum closely enough to provide some degree of corroboration.

The FBI tests use 10% gel to represent a human body. But a human body is made up of different tissues of differing thicknesses and densities. Using a single uniform medium to represent multiple tissues is not going to duplicate actual performance in an acutal body. What the FBI is saying in their test criteria is that a bullet that travels 12-18″ in 10% gel is probably going to be able to travel the necessary distince through a human body to strike a vital structure. It is not saying the bullet must travel 12-18″ in a human body in order to do so.

12″-18″ was the recommendation of a paper written by Special Agent Urey W. Patrick titled, “Handgun Wounding Factors and Effectiveness,” after an exhaustive study of the failures in the 1986 Miami shoot-out (still widely available in PDF format on the Internet). This is the upshot of his conclusion:

“Kinetic energy does not wound. Temporary cavity does not wound. The much discussed “shock” of bullet impact is a fable and “knock down” power is a myth. The critical element is penetration. The bullet must pass through the large, blood bearing organs and be of sufficient diameter to promote rapid bleeding. Penetration less than 12 inches is too little, and, in the words of two of the participants in the 1987 Wound Ballistics Workshop, “too little penetration will get you killed.” Given desirable and reliable penetration, the only way to increase bullet effectiveness is to increase the severity of the wound by increasing the size of hole made by the bullet. Any bullet which will not penetrate through vital organs from less than optimal angles is not acceptable. Of those that will penetrate, the edge is always with the bigger bullet.”

Too little penetration will get you killed. Period. Full stop.

I’ve read the S.A. Patrick’s brief on the subject previously and have re-read it again now. None of it is original source information, but compilation of information he found to support his position that penetration is the single most important aspect of wounding effectiveness.

But, even Patrick’s boss, John Hall, states in the forward: “The information contained in this article is not offered as the final word on wound ballistics.” Nor should it be.

Patrick compiled the information based on data available to him in the late 1980s. Technology and science vastly improved during the past decades – new understandings of physics, better diagnostic tools, and medical advances all should contribute to an evolving understanding of wounding and incapacitation.

The words “too little penetration will get you killed” are often repeated. Yet they originate, as far as I can tell, from two of eight people at the FBI panel that debated the 9mm vs .45 ACP cartridges. Nevertheless, they have become a near religious mantra for some people.

Please understand that I’m ~not~ saying that the ideal penetration depth of a self-defense round in 10% ballistic gelatin isn’t between 12″ and 18″.

Rather, I am merely questioning that assumption to see if it has merit. I take Thomas Jefferson’s urging seriously: “Question with boldness even the existence of a God; because, if there be one, he must more approve of the homage of reason, than that of blind-folded fear.”

To date, I’ve not been presented with any evidence to suggest a penetration depth of 12″-18″ in gelatin correlates highly with a more rapid incapacitation of an attacker that a bullet that penetrates outside of that specific range. In fact, it looks like all of the source material I can find points back to a single workshop in 1987 where these numbers were supported by half of the attendees.

A transcript of that workshop would be instructive as to the context in which those numbers were presented. Alas, I do not think such a thing exists. Likewise, I’ve not found any other original research to back these specific numbers in the past 30+ years.

Returning to the forward of Handgun Wounding Factors and Effectiveness, Hall correctly described Patrick’s paper as “an important contribution to what should be an ongoing discussion of this most important of issues.”

An ongoing discussion is what we as a community should be having about the concepts of bullet effectiveness. It is in this spirit I wrote the article and have engaged you in conversation.


I’ve seen many many gel tests of various handgun ammunition. One or two videographers do excellent presentations. But as soon as a bullet only goes 11 inches. They say FAIL (insert buzzer noise here). Or the bulletin goes 19 or 20 inches (also FAIL more buzzer). The most common failure is hollow points going through clothing and failing to expand or only have partial expansion. It seems the 12-18 inch standard though reasonable seems arbitrary. All stemming from an FBI fiasco in the 1986 Miami shootout , the heroism of the Agents notwithstanding and definitely not in question, and the FBI ultimately blaming the ammunition for the tragedy. In essence those agents brought handguns to a rifle fight even knowing the perpetrators had
semi-auto rifles. I was a police officer at the time and our training staff dissected that shooting pretty well so we could learn from the incident.

Minimum 12″ penetration, I believe, came about due to the “failure” of the 9mm Silvertip to puncture the heart of one of the suspects in Miami after entering his right arm and crossing his torso to stop just short of the heart. Still, in my mind, pretty good performance for a 115 grain expanded JHP. I agree that the obsession with these tests is unrealistic. Like any ballistic testing, they are useful- to a point. So is water jug testing. Gunfights are not standardized. They will not conform therefore to test results. Nothing is a guarantee. The FBI’s big sin in my mind is ramming their opinion down the throats of every LE agency in the country despite a relative lack of gunfight experience. Why don’t we look to Border Patrol ( ICE) or DEA for additional insight into what works? Granted I’m no qualified expert.

With respect to ammunition evaluation, during my tenure there (1997 – 2012 at BRF), the FBI never “rammed their opinion down the throat of any LE Agency”. DATA and FBI selection criteria were all that were provided. Agencies were encouraged to make their own choices, based upon their circumstances. Agencies were told: “We will help you learn how to evaluate ammunition so you may make educated decisions.”
The closest we came to “recommending” a cartridge would be identifying what we used. This was only done if asked. Answering a question is factual. Many may have construed it as a recommendation but it was not.
As far as I am aware, the above continues to be the policy of the FBI BRF.

To pass the FBI protocol, a round has to pass all the tests, right? If so, it automatically passes those tests you’re saying are most relevant to the private citizen.

I don’t think there’s any real “weighting” here. If it passes the FBI protocol, then it passes the bare gel and clothing tests.

The only thing the FBI protocol would do is possibly reject rounds (for failing to pass the barrier tests), that might otherwise be suitable for private citizens not concerned with barriers. I’m willing to live with that.

If I can have a round that does BOTH (passes the non-barrier AND barrier tests), rather than simply passing the non-barrier tests, why wouldn’t I want that?

Of course terminal ballistics tests are meant to be predictive. Unfortunately, there is no way to meaningfully correlate lab penetration with real world effectiveness. In other words, lab results are predictive on the basis of reason, but supposed predictiveness cannot be demonstrated empirically. If only we could carry out executions by handgun firing squad.

I don’t think any of the barrier penetration tests are very relevant to self defense. As to which tests I think are, I go by the recommendations of the now disbanded international Wound Ballistics Association. The IWBA had a two-prong test:

– in bare, calibrated, 10% gelatin consistent expansion with an average penetration of 12.5 to 14.0 inches
– in calibrated, 10% gel covered with four layer of heavy denim consistent expansion with an average penetration of 13.0 to 16.0 inches.

Consistency of expansion is measured by the standard deviation of penetration depth — the smaller the standard deviation, the more consistent the expansion. In the IWBA’s draft assessment specifications they suggest a consistency goal of a standard deviation of 0.6 inch as being easily achievable with premium JHPs.

The two tests have equal weight — fail either one and the round in question is deemed inadequate.

I’ve never seen the FBI offer a rationale for relaxing the IWBA’s standards; thus, I’ll stick with the professional scientists and practitioners of the IWBA. After all, in this day and age the FBI has proven it can’t be trusted.

Develop a new standard for comparing handgun self-defense ammunition?
I think Lucky Gunner already *did* that. They developed a reasonable test protocol derived from the FBI methodology, then stepped up and did the damn work. While the rest of us continued to yammer about the topic; or went out to shoot one round from three brands in one caliber into a ballistic catch unlike any other test anywhere. If you’d rather have 8″ or 30″ of penetration, or you like variable instead of uniform expansion/penetration — fine; data’s there, pick yer poison. If a load performs similarly on the most similar FBI and LG tests, I’d consider it a bonus if it also does well on FBI hard barriers. But I’d agree it makes little sense to weight the hard barrier tests much for routine civilian defense use, or to blindly favor an agency round for personal use.
Might there be some better protocol or measure of relevance? Sure, why not? Cook something up; do the work; throw the results out for the world to view. That’s how science operates. But — if you want to buy a box of ammo this afternoon, or next month, or without putting a year of your life and half your bank account into it — the substantial FBI and LG sets of real data are already available as alternatives to advertisements and strongly held/weakly supported opinions.
My same reasoning accepts the work of Ellifritz as “close enough” to demonstrating that caliber barely matters. Have a reliable gun; have the gun with you; be able to run the gun under stress — that matters a lot.

A gun fight is a gun fight, at that point, whether you have a badge or not, the requirement is to stop the threat,whether barriers are involved or not.
What I’m hearing is that a test for REDUCED CAPABILITY is OK for “regular citizens” because “regular citizens” have no business needing to shoot through barriers. If a “law enforcement official” finds himself in a defensive position that requires the need to shoot through auto glass or sheet rock, that barrier blind ammunition could save his life. To say that a “regular citizen” in that same position doesn’t require, OR DESERVE, the same capability, is unacceptable to me. I’m hearing too much of that lately, and it is being MANDATED BY LAW!
To hear that NO ONE NEEDS 10 BULLETS TO KILL A DEER, and then having that person sign a bill into law restricting “regular citizens” to 7* rounds, and then exempting “law enforcement officials” because the restriction would “PUT THEIR LIFE IN DANGER”. Then being told no one needs “high capacity magazines” to defend themselves – but RETIRED “law enforcement officials” need to be exempt in order to retain THE ABILITY TO EFFECTIVELY DEFEND THEMSELVES, sure makes a gun owner CRINGE every time there is a distinction made between what is NEEDED for a “LAW ENFORCEMENT OFFICIAL” to survive a shooting situation and what is ACCEPTABLE for a “REGULAR CITIZEN” to eventually be RESTRICTED TO.
Once there is a distinction between LAW ENFORCEMENT AMMUNITION and SELF DEFENSE AMMUNITION (for “regular citizens”), THE NEXT STEP IS RESTRICTING “LAW ENFORCEMENT AMMUNITION” TO “LAW ENFORCEMENT ONLY”! (The EXACT signs on ar15’s and Glock 19’s and 17’s etc with standard capacity magazines)
Next will be that “civilian” shootings, statistically involve no more than 3 rounds fired.
Soon after, “regular run-of-the-mill non-law-enforcement citizens” will be limited to 3 round revolvers with frangible bullets.
The WHOLE INTENTION of the Second Amendment is that there is NO DISTINCTION between our rights and that of the enforcers of the state! If the “law enforcement” officials evolve to be enforcing for the next Nazi party and are loading US into box cars, there can be NO REDUCED CAPABILITY at that point!
Any suggestion of LAW ENFORCEMENT AMMUNITION being different from SELF DEFENSE AMMUNITION because “regular citizens” do not need to have the capability to effectively shoot through barriers, is opening ANOTHER can of worms that is a great DIS-SERVICE to free law abiding gun owners that are literally in a WAR to keep their Second Amendment rights. I want the same Gold Dots used by local “law enforcement” agencies in MY handgun, PERIOD.
This whole line of thinking REALLY PISSES ME OFF. Could RESTRICTING “law enforcement” officials to “SELF DEFENSE AMMUNITION” instead of “LAW ENFORCEMENT AMMUNITION” possibly PUT THEIR LIFE IN DANGER?
If so – IT COULD POSSIBLY PUT MY LIFE IN DANGER! (I am VERY sick of being mandated by LAW to PUT MY LIFE in a position of defending myself and my family with a standard of firearm, magazine capacity, and now ammunition, THAT WOULD PUT A LAW ENFORCEMENT OFFICIAL’S LIFE IN DANGER, but by law I’m told it is all I need.)
THAT is exactly where this will go.
You know, I do believe that in New Jersey, HOLLOW POINT AMMUNITION (the standard for all self defense) is RESTRICTED to LAW ENFORCEMENT ONLY!
Burn this article, it is a dangerous SEED OF THOUGHT for those that work relentlessly to incrementally disarm us.
*The Federal Court determined that Andrew Cuomo’s 7 round limit mandated upon “REGULAR CITIZENS” was arbitrary – and it was bumped up to a 10 round limit.

Absolutely nothing in my article is intended to support any of your suggestions. Rather the words mean exactly what they mean – I’m merely considering that the FBI testing protocol may not be the best method for determining the most effective ammunition for the average armed citizen engaging in self-defense.

Good article. I have no problems with testing barrier penetration for personal defense ammo. There is no reason to assume a private citizen will not or should not ever have to shoot through a barrier of some kind. But because the organic gel used by the FBI is so sensitive to ambient conditons during both preparation and testing, and is not reusable, I concur that developing a test with Clear Ballistics gel would be useful. however, there are problems with Clear Ballistics gel that came to light during the Ammoquest testing on YouTube:

a) The wound channel in Clear gel appears to remain open longer than in Cordinace gel which allows the bullet to bounce back from the end of the wound channel. Some bullets came to rest as much as 6″ back from their farthest point of penetration. Should the penetration distance be measured to the bullet’s final resting location or to the end of the wound channel? The bounce back suggest the gel has quite a bit of elasticity along the axis of travel. Is the susequently measured length of the wound channel accurate or is it distorted by the elasticity?

b) A bullet penetrating denim carries pieces of denum into the gel leaving fragments and fibers all along the wound channel. This contamination reduces the reusability of the gel.

Regarding usefulness of FBI protocol for the civilian interested in self defense, it may be helpful to observe how a certain round that has proven itself on the street performs in clothed and unclothed 10% gel. Then, if an unproven round duplicates that performance, it is reasonable to assume that the unproven round may be as effective as the proven one.
This is a simple apples to apples comparison when price or availability make the proven round unavailable.

Hi George,

I think you hit on the missing piece: any kind of correlation study between FBI testing and street results. Yes, the rounds that tend to do well in the FBI testing /seem/ to function well in the real world. Without a correlation study, we’re still guessing.


I think what is helpful is to look at the permanent cavity left in the gel by a bullet when choosing a round to try in one’s own pistol. While the temporary cavity can be quite impressive, we still have no way to know exactly how much that cavity effects the one receiving it. I do however believe that what happens to a body will have an effect on that body, it’s just that we don’t know how much of an effect it will have. If the permanent cavity of an unproven round matches that of a proven round and if penetration and expansion matches, then we have a possibly useful correlation.

As an addition to my above post, I’ve heard it said that all bullets do is “poke holes”. They let in air and let out blood. Thus the bigger the permanent hole that a bullet leaves in the gel block, the better. For clarification, the permanent cavity is the hole the bullet made through the gel after everything settles down and goes still. The instantaneous temporary cavity can be quite spectacular, but it goes away leaving the only thing you can measure that really matters; a permanent (or crush) cavity that is “so many inches in diameter and so many inches long”.
If an unproven load can meet the dimensions left in the gel by a proven load, then I think we may be on to something usable.

Excellent article on one of the most hotly debated topics in all of firearms! The FBI ballistic gelatin tests are a good starting point in evaluating ammunition performance testing. They are standardized, repeatable, and include clothing/barriers related to real-life FBI shooting situations making the results somewhat practical. However, these tests can only go so far, and can only provide a somewhat broad range of information for shooters to make ammunition selections.

First, and perhaps foremost, is the fact the FBI ballistic gelatin can only mimic bullet performance in human tissue, and cannot account for human bone – something often encountered during real-life shootings. Second, the clothing used in the FBI testing may be relevant to warmer weather shootings, but their winter clothing selection is poor. Cotton and polar fleece, are not like thicker cotton canvas found in Carhartt-style jackets or heavy insulated winter coats. These thicker materials could have a significant impact on bullet performance, particularly in hollow-point expansion and penetration (rounds most likely to be selected for self-defense).

The intended results of a typical self-defense shooting (stopping the threat) has everything to do with bullet placement, and penetration through vital organs or structure. Intermediate barriers are not a factor in most cases. With that in mind, a person looking for a good self-defense cartridge would most likely be better off knowing how the bullet performs through various layers of clothing. Knowing how the bullet performs through intermediate barriers is good, but far from a critical decision factor for self-defense.

In general, a hollow-point round that penetrates 12-18 inches in gelatin, and expands to a diameter greater than the bullet’s original diameter, will likely do well in stopping a human threat with proper shot placement.

In studying researched human dimensions, I think a reasonable inference can be made as to where the FBI came up with the 12-18 inch penetration numbers. A human dimensions chart published by Research Gate, shows the “average” human adult male (a very fit male no doubt) has a torso that is approximately 12.5 inches wide (side-to-side) in the abdomen, and is approximately 15.75 inches wide from roughly arm pit to arm pit ( Add a few more inches on either side of the upper torso measurement to include upper arm width, would arrive at around the 18 inches in width (when looking at a person from front or rear). Not perfect, but working within researched “averages.” I know many males are “thicker” than this, but even adding 30% to these numbers gives an abdomen width of 16.25 inches and an upper torso width of 20.475 inches. Another human dimension scale shows an “average” adult male has an upper torso depth (distance from front of body to rear, or anterior to posterior) of about 9 inches. Again adding 30% gives us about 12 inches in depth.

With these measurements in mind, I think the FBI gelatin tests using a 12-18 inch penetration standard is reasonable and has at least some anatomical merit (taking into consideration the gelatin’s limitations). Bullets need to be able to penetrate deep enough into a person to create significant trauma to vital organs, or skeletal structure, to reduce their ability to continue aggressive action. On the other hand, bullets that over-penetrate become legal liabilities once they exit the intended target.

Take all the research in proper perspective, then select ammunition that meets your individual needs within a reasonable expectation of performance in the intended target.

I found this article as a result of a search I did which was inspired by an ammo test video in which the guy absolutely refused to use any ammo which did not meet the 12-18″ criteria. And I find the zealous adherence to this arbitrary standard baffling. First, overpenetration is a moot point. Many of the shots fired in defensive situations are misses. Worrying about hits continuing on is silly. Second, I’m not convinced underpenetration is relevant either. In a defensive situation you aren’t as likely to be shooting at strange angles since an attacker has to be facing you, more or less, to pose a plausible threat. It’s not like a law-enforcement scenario where I might be trying to flank a bank robber. To my mind, the primary criteria for personal defense is caliber; bigger is better. The rationale is that a large projectile will damage non-vital tissue to the point of incapacitating an attacker even if the shot is poorly placed. A 9mm to the arm may not effect an attacker who is drugged or hyped enough to not notice the pain. But a 12ga slug to the arm renders that arm useless regardless of the attacker’s state of mind. I’d rather have a massive bullet that doesn’t go quite as deep than a small bullet that goes all the way through. I think the general efficacy of hollowpoints bears this out.

If, as the one cited study claimed, shock, energy, and temporary cavity don’t make a difference then how does that study account for the performance difference between .380, .38spl, .357mag, and 9mm… all of which are roughly the same diameter and weight? I don’t know of anybody who would say even the hottest .380 was comparable to a .357mag in terminal performance. But the only difference between the two is kinetic energy.

I also think we put way too much emphasis on center-of-mass shots. I realize that’s to compensate for poor marksmanship. But a headshot with even the weakest of cartridges tends to end a fight very quickly. You don’t need 12″ of penetration with a headshot.

The 12-18″ standard came about by obvious means-
they used the top rounds in M&S and declared that the standard.
If you look at the 125gr SJHP .357, the 185gr +P Rem GS .45, and 115gr +P Corbon 9mm, you’ll see ideal gel performance, despite these rounds’ having been developed without gel. They were M&S standouts. After 50 years of street rep (for .357), they just shot a bunch of these known good rounds into gel and inferred that this penetration must be ideal. “Do what those do.”
But that is simply half of the picture. People who shoot animals (in fact the entire continuum of rifles in a particular size class) know that things revolve around energy on target *assuming* sufficient penetration.
There’s a good feral hog night hunt on Youtube filmed with a FLIR. Hogs get dropped and get back up. One charges 50 yards to attack the shooter after being shot and dropped. The knockdown effect is real but is as-yet unquantified.

Hollowpoints exist to facilitate maximum energy transfer, not to protect bystanders. If energy effects weren’t real, hollowpoints would be no better than FMJ. Corbon’s 115gr +p 9mm only expands in ideal gel to a little bigger than the size of a .45ACP hardball round. Something else makes one highly effective and the other notoriously not so.

I suspect that since M&S was published in 1992 after the FBI’s 1986 testing, that M&S had little to no influence at all on the 12-18″ reference.

A question…….has anyone ever made a study of how autopsy reports compare to/with the 12″ bullet penetration requirement? I agree that there are many variables involved but if 100 deaths were the result after only 10″ of bullet penetration into the upper chest, would that not bring the 12″ criteria into question? Or is this an over simplification?

The argument for the “perfect” round will go forever. Let’s look at real world situations. Let’s say you are in an area where grizzly or polar bear are likely to be found. If you are confronted by a large grizzly, what handgun would you wish you had on you. I think the first reply would be a 44 magnum by most of us. With that in mind I would have to agree that size and power are the most important factor here.
Now for every day personal carry in a civilized situation, a large frame revolver would not be a reasonable solution. So we each have to come up with a comfortable handgun to carry concealed, with a round that would be sufficient to stop a deadly attack, but able to allow the shooter to be able to get off consecutive, accurate shots. That is probably more important than what round tests the best in a ballistic test. Another thought should be what gun has the best dependability? It never ends!

Leave a Reply

Your email address will not be published. Required fields are marked *