Previous Post
Next Post

ShootingTheBull410’s quest for the best-performing defensive ammo for a 9mm pocket pistol rolls on with a test of the Barnes TAC-XPD. It comes in a shiny black box with a high-tech clear window, and even the bullets themselves are black…perhaps invoking the old “Black Talons”?  But the Barnes solid copper bullet has already been tested and found to be a superb performer when it was loaded in the Cor-Bon DPX round. The Cor-Bons, though, are less expensive, come in a plainer box and are loaded to a higher velocity. How will the same bullet do in Barnes’ own shiny black version? Check out the vid to find out.

Previous Post
Next Post


  1. I carry an M&P 9c. This whole series of ballistic gel testing has been very informative. Thanks.

  2. Stick with the best and most cost effective–either a 124gr +p gold dot or HST. The best IMHO that you can carry.

      • Chad is giving solid advice. Gold Dot and HST offerings in just about any caliber and loading perform quite well and they are cheap enough (typically available under $30/box) to run enough rounds through your gun to determine that they function properly.

  3. I appreciate the effort and measurements done with this series, but do take exception with

    “quest for the best-performing defensive ammo for a 9mm pocket pistol”

    There is no “the best” especially in combination with generic “a…pistol.”

    Are the differences between one ammo and another “statistically significant” or are they merely academic? That’s not being tested or reported.

    So, my “problem” with this is that there will be a bunch of folks that do not understand scientific testing and data analysis crying up and down the Internet claiming “use this…it’s THE BEST.”

    First rule is “have a gun.”

    Article series like this one, though very interesting and entertaining, fuel a fire that detracts from that. ANY ammo can fail to stop a threat, and ANY ammo can be fatal.

    There is no magic bullet; 0.1% ‘better’ expansion or penetration in gel will make precisely ZERO difference in a given “test” in a real street fight.

    As a handloader for over 20 years, my recommendation for carry ammo would be the following:

    (1) Pick two or three reputable defense ammo brands and bullet weights.

    (2) If you REALLY want to know ‘best’ for you and your gun, commit to spending the money to thoroughly test those three for accuracy and functioning reliability.

    Shoot a LOT; 1 box of each won’t do it. It does not have to all be done the same day.

    (3) In all likelihood, one will bubble up as “your favorite.”

    (4) Don’t sweat it at if some other brand or type squeaked out a better result in a SINGLE test in ballistic gel out of someone else’s gun.

    There are bigger factors at play in gunfights:

    Shot placement
    Adaptability to dynamic circumstances

    and of course….


    Now that that is done….I do like the series and the tests. They make for fun discussion and provide a point of “technical” comparison. Let’s just beware of reading too much into “best.”

    • Some dudes like to get all geeky with ammo ballastics…. Let them have their fun.

      When they get attacked by the 6ft Jello man they’ll be adequately prepared.

      • The point of the gel tests is to have a common metric for ammo testing. He has repeatedly stated it will not penetrate the same and expand the same in humans. There is statistical evidence that a projectile will stop a human threat if it can penetrate 12-18″ in gel. And since other people can make gel, we can compare results. No longer do we need to shoot sheep and pigs then try to justify out ammo.

        • All true, but the testing has to be kept in perspective for it IS vs what it is NOT.

          “Just buy this and be done with it” when the difference is not significant (or at least not shown to be) is pointless.

        • Shooting sheep and pigs is actually a much more useful way of studying terminal ballistics, though. While gel is useful as far as comparing how bullets act in that medium, it’s no simulation of a human with all the bones and squishy stuff of different densities that we have.

        • But at the same time, he’s exposing ammo like the G2 RIP.

          Plus this ammo series exposed the fact that the 135 gr 9mm Critical Duty bullets don’t like short barrels.

          Additionally, he revealed which 380 loads and bullets get desired penetration out of a pocket rocket TCP/LCP, which WAS totally worth doing.

          I don’t think he’s advocating that some magic bullets will replace what you’re advocating, I think he’s saying “if you’re carrying a short barreled weapon, some ammo performs better out of it”. I don’t think he’s implying what you seem to think he is JR.

      • “Some dudes like to get all geeky with ammo ballastics…. Let them have their fun. ”

        I agree and get into myself. Guilty as charged on the geeking out.

        It’s the next step it is taken to (with evidence in this very thread) that is a peeve.

    • I’ll let him do the ballistic tests to weed out the ammo that really under-performs. Based on his results I will see what defensive ammo I can get locally, since ordering ammo in NY is a pain in the butt.

    • I well understand that ammo performance is only one factor in an overall defensive gun use. People harp on “shot placement” and “training” and “mindset” and all that. We all know that. We all get that.

      But — there is absolutely a very real difference in ammo performance, and better-performing ammo is the one factor that you definitely can control.

      Example — if you did everything absolutely right, but the ammo itself failed to do its job, wouldn’t that suck? Wouldn’t you want to avoid that? Especially knowing that it’s pretty easy to find which rounds will, and won’t, perform from your particular gun (or class of gun)?

      That’s what I’m looking for. There is no “magic bullet”, but there is a difference between chronic underpenetrators and chronic denim-pluggers, and good-performing bullets.

      I’m on a quest to find the best-performing ammo. There doesn’t have to be one single title-holder of “the best”, but there is definitely a class of best-performing, mediocre-performing, and worst-performing. I’m trying to find out how each round I can get my hands on, falls into those particular categories, and most especially — from a 3″ carry pistol!

      I’ve seen people in various forums re-post the “Self Defense Ammo FAQ” list as if it’s gospel. It’s a list of bullets tested and approved by DocGKR — and it’s a great list, BUT — all the entries on it were tested from 4″+ barrel pistols. If I was carrying a Glock 19 or other 4″ pistol, I’d gladly use any bullet on that list. But the results DocGKR got do NOT necessarily translate over to a 3″ pistol. Some do, some don’t. Doctor Roberts himself said “if you are using a pistol with a barrel less than 3.5″, you’re in uncharted territory.” Well, I’m trying to chart that territory, to find out what are great performers from the 3″ barrel, and what aren’t.

      That’s all I’m trying to do. Great shot placement with great ammo will beat great shot placement with poor ammo. Great shot placement with a bullet that expands to .75″ will beat great shot placement with a bullet that doesn’t expand. Great shot placement with a bullet that only penetrates to 7″ may fail to stop your attacker, whereas that same great shot placement with a bullet that penetrates a full 12″ may stop your attacker cold — as was discovered during the FBI’s Miami shootout.

      I will agree with JR that the ultimate would be to conduct your own testing, from your own pistol. That is, in fact, exactly what I’m doing here — but proper testing is expensive, time-consuming, and requires meticulous preparation. Not many of us can do it. So if anyone gets value from these tests, I’m happy to share them.

      Anybody who thinks they don’t have to train just as hard, practice just as much, and place their shots just as accurately, would be entirely missing the point. All those factors are crucial. But there’s one factor that we can improve dramatically — the likelihood that our ammo will perform well. That’s an easy one, and it’s such a simple step to take. It’s the least important factor, but it does still hold importance.

      • “Especially knowing that it’s pretty easy to find which rounds will, and won’t, perform from your particular gun (or class of gun)?”

        Look, I’m simply trying to make a point. “Will” is an overstatement. Anyone that has done any hunting knows this.

        Ask a question about bullet x on the hunting forums, and you will get answers like “it works every time” and “I’ve never seen it work; don’t use it.”

        For the same bullet in the same caliber on the same animal.

        No, check that. For the same HIGHLY RATED, performance designed bullet.

        There is no “it WILL work.”

        I agree with you 100% that some are better than others…absolutely no argument there.

        But the differences between high-end known performance bullets / loads are very small and gel testings will NEVER perfectly predict actual performance in a real shooting.

        My beef is not with you or your series. Please be clear on this point. My beef is with those that will take your data – interesting as it is and informative as it is used correctly and take away something far, far beyond what makes sense and worse, give others advice based on it.

        I’ve dug bullets out of bodies. Real ones. I’ve seen bullets do weird things. There is no “it WILL work” a certain way. I’ve seen single shot kills with .22 LR and I’ve seen .357 Sig take 7 shots to put a man down. I’m familiar with a case that involved three near contact head shots from a .40 with High Performance duty ammo and autopsy showed only one of the three was fatal. This was after the assailant was shot four other times by the same gun / same ammo.

        There are no absolutes; YOU are not the person that I think does not understand that. The person that says “this is THE BEST, just buy this” is.

        And that alone is, or can be, a very, very dangerous mindset to have.

        • I understand exactly what you’re saying. And I largely agree. I think the big missing element here is — I’m trying to find what bullets will perform most reliably from a particular pistol. What my beef is, is with people who think that “oh, I heard Buffalo Bore is the best because some guy tested this from some Glock 21, so I’m sure it’ll work fine in my .380.”

          That type of mentality just doesn’t work. Bullets have a velocity window within which they will (usually) work; if you run them too slow they’ll fail to open up; if you run them too fast they may actually rip themselves apart and result in a lousy wounding profile.

          I disagree that the differences will be “minor” between all the major brands, at least as far as my testing goes, because all the major brands design their ammo mainly for duty pistols, and when those rounds are tried in a short-barrel pistol, the performance can vary drastically. A round that stops reliably at 13″ from a duty pistol, might overpenetrate 32″ from a 3″-barrel pistol — and that’s the exact result I got from the 124-grain Gold Dot. Twice.

          So, you’re right, people who don’t know what they’re doing can misinterpret information and use it the wrong way; I try to be as educational as I can in my videos but I can’t possibly cover all bases or every video would be two hours long. 🙂

          All I can do is put out there the performance that I got, from the pistol that I use, and you’re right — there is no guarantee that any bullet will perform the same, gun to gun, or even lot to lot. All we can do is deal with probabilities and likelihoods. That’s why I take five shots (and, if testing both denim and bare, I take ten shots total) — to get a better realistic representation of how the ammo is “likely” to perform. But there’s no guarantees. That is true.

        • JR,

          I think you have some valid points but are making them in the wrong place. I don’t think I’ve seen STB ever suggest his data demonstrates a certain outcome when the bullet strikes a person and he goes to great lengths to qualify exactly what his tests show and do not show. It seems to me that you have a beef with the keyboard experts that show up on every forum and have the kind of certainty that demonstrates little actual experience. STB’s piece evaluates how a certain bullet performs in a consistent test medium out of a certain, fairly common class of gun (short barreled 9mm). I find this extremely valuable and STB posts on this gun blog are some of my very favorite. STB, I hope your financial situation is such that you can do this all the time because you do it well and I enjoy it.

        • “What my beef is, is with people who think that “oh, I heard Buffalo Bore is the best because some guy tested this from some Glock 21, so I’m sure it’ll work fine in my .380.”

          That type of mentality just doesn’t work. “

          Agreed 100%. {Insert Thumb’s Up Icon}

          Thanks for the discussion, and thanks (again) for the hard work and vids.

          I mentioned this once before a while back in passing, but just out of curiosity. Have you given an thought to testing at different distances (which will lower velocity)?

          Perhaps after you narrow down your “best” list to a couple or three contenders for the title, back up from the blocks and see if there is a difference in “range of acceptable performance.”

          I don’t know…maybe that’s dumb, but it seems to me (intuitively) like it might at least be interesting and may be informative.

          I wonder if one could find a velocity (which translates to distance to target in terms of an SD shooting) that defines a critical point, below which (farther from) that particular ammo becomes suspect.

          That critical point could also be a basis for ‘classifying’ terminal performance.

        • Sorry, TT…we sorta posted at the same time.

          I don’t think I’ve seen STB ever suggest his data demonstrates a certain outcome when the bullet strikes a person”

          Nor have I.

          “It seems to me that you have a beef with the keyboard experts that show up on every forum and have the kind of certainty that demonstrates little actual experience.”

          Yes, this is a fair assessment.

          “I find this extremely valuable”

          As do I. Otherwise, I would not read the posts and take the time to comment / initiate a discussion. It’s a topic I’m obviously a bit passionate about and merely wanted to make a point.

          It was never my intent to take away from the value of STB’s work. Indeed, I will say as clearly as I possibly can that I believe his tests are among the ‘straightest shooting’ ballistic gel tests going. Pun sort of intended.

      • I find this an entertaining, informative, and well spirited series. That said I think JR was alluding to the sample size of 5 rounds. At those sample sizes it could be within the margin of error that you could have received a normal distribution of duds in this box, and a normal distribution of them in the other boxes. I’ll elaborate in case that sounds strange. You test 20 rounds each of companies A through E. The normal defect rate is 1/100. It’s within the normal margin of error that only 1 of those companies would have an expansion failure or other type of failure. Let’s say that failure occurs for company D, that makes company D’s products look inferior to A, B, C and E.

        Do you feel a sample size of 5 is good enough? What’s the normal defect rate for high quality ammunition?

        • You’re totally right, a sample size of 5 isn’t really big enough to get a comprehensive picture. You’d really need 20 or more before it becomes more statistically relevant.

          But testing 20 rounds is simply impractical — at least, without some significant sponsorship.

          I test five rounds per block because that’s as many as you can fit in a standard gel block, before you start to run serious risk of overlapping the wound tracks. (unless you’re using tiny bullets like .22lr solids, of course; you can fit a ton of those in a single gel block).

          Part of the reason I do the testing the way I do, is because I was unsatisfied with how most youtubers or other forum-based testers will shoot only one round, and then make a declaration about a particular ammo type’s suitability based on that one single round. JR’s absolutely right, rounds can vary from one to another, and as this very test showed, sometimes it’ll work and sometimes it won’t. So I put as many rounds in a block as I reasonably can, and I use two blocks for a full test (assuming the ammo passes the first test, of course). So in a general ammo test I’ll fire 10 rounds, which starts to give a fairly reasonable picture of the ammo’s overall performance profile.

          It’s not perfect, but it’s as much as can be reasonably done. DocGKR shoots five rounds per block, the FBI uses five rounds per block, and so that’s what I do too, for the same reasons. I don’t run as many different tests as they do (DocGKR nearly always includes windshields, and the FBI adds plywood, drywall, and car door steel).

          In the end, JR’s correct: I cannot give a picture of what the ammo WILL do. All I can hope to do is paint a picture of what it is LIKELY to do. And hopefully, with at least five and up to 10 rounds per test, we may be actually getting a reasonable picture of that likely performance.

  4. @STB,

    Thank you again for your testing. Don’t let JR and co. get you down. Gel isn’t perfect, but it is consistent.

    I’m curious as to your opinion of the newest .40 Cal Winchester PDX 180 grain from a 4″ barrel. If you haven’t already tried it I’ll send you a box.

    [email protected]

    • “Don’t let JR and co. get you down. Gel isn’t perfect, but it is consistent. “

      Why would he? I’m not ragging on him or his work. I’ve stated numerous times that I’m digging (and fully appreciate) his work and the video series.

      Did you even read what I said or just assume that since I was making a larger point about “ballistic testing” of high performance ammo and the way people extrapolate these results WAY out of context that I was criticizing him?

  5. @JR,

    Well, I skimmed what you said, to be perfectly honest. It sounded critical, but you also made good points.

    I don’t have blind faith in gel tests, but I don’t discount them either. We can’t be shooting violent felons through various barriers in a lab. Coroners and combat survivors also have their insights just as I have experience with less lethal weapons in the field.

    In fact I have enough respect for STB and his testing process that I’ll send him some. .40 cal to test.

  6. Hey STB, I don’t think the 1 bad round out of the 5 should have excluded it from the denim test. The 4 bullets that expanded looked nasty.

    • Hi Kevin,

      A couple of reasons why. First, as I find more and more rounds that perform superbly every time, my tolerance level for failures grows lower and lower. But more importantly, for this round in particular, I tested the identical same bullet in the CorBon DPX loading, and it was a stellar performer. After discovering one fail-to-expand and one bullet that looks like it was on the fence about whether it was going to decide to expand or not, I came to a simple conclusion — the CorBon loading is a better load for the 3″ barrel. Maybe from the 4″ they’d be equal, but from the 3″ the extra 108 fps (average) that the CorBon bullets had, pushed them deeper into the realm of reliable performance.

      That’s why I ended the video by saying “if you want this bullet, get the CorBon load — or use a pistol with a 4″ or greater barrel”. The bullet is excellent, the expansion is fantastic, but the velocity it’s loaded to by Barnes isn’t enough to assure a reasonable likelihood of great performance. The extra velocity of the CorBon loading gives it what it needs to perform reliably.

      I did shoot three rounds of the CorBon through denim, and in that test you’ll see that it performed just fantastic.

  7. STB.
    Thanks so much for your ballistics test videos.

    In spite of JR’s comments, I find them very useful.
    I understand that these tests have nothing to do with training and shot placement
    (which are clearly more important).
    But whatever your level of training, why not use the rounds
    that will likely perform best in a particular size firearm.
    I don’t view the tests as holy writ, but they are far better than anything else that has been
    readily available – at least to me.

    BTW, Have you considered adding a PayPal button to your site so that those of us who
    find your work useful could donate and help defray your costs?

    Again, thanks for all your hard work.
    It’s much appreciated.
    The price we paid was the price men have always paid for achieving paradise in this life–we went soft, we lost our edge.
    – Frank Herbert Dune

    All governments suffer a recurring problem: Power attracts pathological personalities. It is not that power corrupts but that it is magnetic to the corruptible. Such people have a tendency to become drunk on violence, a condition to which they are quickly addicted.
    – Frank Herbert Chapterhouse: Dune

  8. I appreciate the videos and the work you do to create them; I consider your testing a valuable resource for any consumer in this area.

    However, I find myself wondering what are your sample sizes? We only see 5 or 10 shots in a video (typically), I wouldn’t exactly call that a valid sample size. Perhaps that could be the issue with some of the under performers you see like the one that failed to expand in this video. Maybe you got that 1 out of 10/100/1000/10000 etc that had a manufacturing defect, etc. etc. that caused it to not perform properly. For example, 1 out of 5 that fails to perform is a whole lot worse than 1 out of 20; I think that changes the odds in an important way.

    Another side effect of increasing sample size is you would get a better understanding of how often to expect failures to occur. No manufactured item can always be 100% perfect (or ever is) so there has to be a failure every now and then. Even with rounds where all 5 shots fired perform perfectly, there has to be a failure eventually. It could give you a better idea of the performance reliability of a particular ammo. You might get 5/10/20 good performers then come away with 3/5/10/etc. ones that don’t perform as well or adequately and all sorts of other scenarios like that.

    I understand there are obvious logistical problems with my suggestion, but I think we could be missing out on something important or valuable. Just a legitimate question about your testing methods, data collection, and protocols.

    Then again, if 5 or 10 perform flawlessly then maybe that is a good enough judge to extrapolate from.

    Just a side note to make clear I am not playing favorites or in a brand war, I have no particular love or hate for barnes. I carry critical defense in my revolvers because they are generally the only .357 mag defensive round available in stores near me and hydrashoks in the semi autos because of the whole ‘its what the state police use’ defense.

    • “However, I find myself wondering what are your sample sizes? We only see 5 or 10 shots in a video (typically), I wouldn’t exactly call that a valid sample size”

      With all do respect, I doubt that a statement like this would ever be made from anyone who’s ever dedicated the resources required to do these types of tests. Calibrated ballistics gelatin is relatively expensive to purchase or produce, and has a very short usable life span. Last time I checked, it was over 4X the cost of the ammo that was going to be shot into it. It can be made without dedicated equipment, but doing so takes a lot of time, is a general pain in the rear, and can make a real mess. Having equipment dedicated to the purpose minimizes these factors, but drastically increases the overhead costs. Then there’s the fact that the stuff is relatively heavy and can be a little difficult to work with, especially if you’re on time constraint where you’re trying to get the testing done before the material warms up and is out of “calibration”. Then there’s the amount of time required to actually do the testing and video editing. It wouldn’t surprise me if he has over 12 hrs of work wrapped up in each of these 4 min videos. In short, this process is a real pain in the rear, and he’s done a fantastic job of it.

      5 shots may not be statistically ideal, BUT it’s a LOT better than 1 shot. At the end of the day, as JR, STB, and others have pointed out we’re not trying to stop the stay-puff jello man anyway. So 5 rounds gives a decent sample size that demonstrates some level of consistency without requiring so much time money, and effort that nobody is willing to do it on their own

      Just say’n.

Comments are closed.