This always just made sense to me, and I don't understand why others have a problem with understanding. A three round "group" is enough to tell me that I am a foot or more left/right/low/high. Five rounds is enough to start walking to a zero. Ten rounds gives me a pretty good zero. Twenty rounds confirms it. When I think I've got a rifle with a good zero I start shooting 30,40,50,60 rounds at a target before I go down range and ask myself...Is my average POI a good match for my POA? Or do I give it one click...?
Great. If you shoot a hunting rifle and normal bought hunting ammo, you start new box after 20. A new box will have slightly different properties and that the end of meaningful statistics. But if you hunt you don't care about MOA. You need to place a shot i a bierdeckel from 0-200m. Setting of the sights makes greater uncertainty
@@eugenohanka "Slight different properties" is called the error. And its easily accounted for in statistics or probability since you can assume that factory ammo will have a +-15 SD and seating depth along with BC's that vary by .001-003. Kind of a moot to bring it up. But fundamentally, you're mistakenly assuming that each box has a different error when in reality each bullet from every box will defer depending on manufacturing conditions and quality control. That's why you keep shooting until you get a significant sample size.
@@eugenohanka”setting sights leads to greater uncertainity” not at all, your sights have to be set somewhere so adjusting them can’t make it worse. To account for box/batch variation you just take rounds from multiple boxes and use that when you’re doing your groupings. You can either just buy a load of boxes at once or every time you get a new box put a round or two to one side for testing (the latter is better bc it stops you buying boxes from the same batch, but you have to wait a while before you’ve got enough rounds aces up to get a proper zero)
I read an article somewhere (probably TFB) about the US Army's ammunition batch testing. They buy lots of 1 million or more and randomly pull 1,000 rounds. They fire those rounds and require half the shots to land inside a circle (like 2 MoA, I believe). The writer went into the statistics of the test, showing that a 3 round group was about 66% representative of any ammo/gun combo, a 5 round group was around 90%, and 10 round group somewhere around 95% and the 1,000 round group was extremely representative.
Are those tests done with a fixed gun in a mechanical test rig or are they done with an actual human and a precision rifle? Because only the former would produce scientifically accurate results for just the ammo.
SonsOfLorgar Mechanical test rig. Not sure if it is a rifle in a test rig or a .5.56 chambered barrel in some sort of fixture, but it was not some guy holding a rifle and burning up 1,000 rounds, lol.
Sunshine_Shooter Well, I never assume thinking when actual officers are involved in the process ;) At least in this case they wouldn't have a map and a Compass!
Once I saw the normal distribution, I know what it is right away. Sample size is too small. you have to see where the mean, mode, standard deviation and median.
In Germany we call it the Mexican sniper. With enough shots fired, you will eventually have some very close to each other. If you only show off that one group of close shots, without pointing out that you shot 100 times to get those, then you committed the fallacy of selection bias. The result is insignificant unless you can repeat it consistently. PS: 4:50 "fewer" ;)
I don't think that your, "Fewer" comment was appreciated--at all. In fact, your message had--at least--one error, but I don't think you want me to point it out.
No, English traditionally uses less; try reading historical documents and you will find out which is correct. One of these stupid incorrect pedantries, like octopi.
A principle which seems to be prevalent even in "professional", well-funded trials - because everyone tends to be biased - is, "once you have the results you want, stop immediately and call it a day." So if, after three rounds, you have a good looking group, why continue and risk spoiling the results? Lies, damned lies and statistics indeed. I subbed after this video.
I have just read through the comments on this video and your many responses and I have to congratulate you on staying calm and polite in the face of extreme and repeated provocation. Well done, sir! Good video, too.
People talk about three round groups because it makes them feel good. "My new rifle shoots three round groups that can be covered by a dime. Those other shots are fliers, cause by cosmic rays hitting the bullets mid flight." We also do it to keep new shooters from getting discouraged. "Hey! You have some nice three round groups here. The rest of those shots are what we call fliers." I had assumed that you were in the UK. If that's true, how do you manage to have that handgun, an evil killing machine just waiting for a chance to run amok?
for a hunting rifle as well as others, depending on purpose and use, shooting one shell from a cold barrel , then repeating the next day, etc. gives you a real world picture of what the gun will do when shooting from a cold barrel...a first shoot capability.. Good job on taking on this explanation !
Fuck everyone's MOA and such nonsense. Consistency, confidence, and experience in what your rifle can do with what you feed it is always best. Also hits are what count.
If the rifle is one built to a military spec, all I ask is minute of man. I.E., I should be able to reliably hit a torso sized target at 100 yards with iron sights and whatever cheapo ammo I feel like throwing down range. Hunting rifles should be a bit more accurate.
MOA is an important measurement for long-distance shooting, but people tend to use it as a way to brag or lie to themselves. The trick isn't to hit the target perfectly. It's to be able to hit it repeatedly while it's moving, camoflauged, poorly lit, in the wind, rain, etc. and to have fast follow-up shots because you WILL miss often under those real life conditions. I use reinforced camo targabots or I hunt. I particularly like hunting in the rain, since it tests my gear and makes it harder to track.
great simple explanation. Recently I put together a 1000yrd rifle, savage 10 fcp-sr in 6.5 creedmoor- very accurate rifle. It will consistently print .3-.4" 5 shot groups@ 200 yrds. yesterday went to the range, using my most consistent hand load and shot a 10 round group and it measured just under an MOA, so a significant change in group size on a larger sampling.
More factors can come into play with increased group size: barrel heat, wind changes, eye fatigue, etc, etc. It’s really not possible to measure inherent accuracy by shooting groups of any size, only relative accuracy.
Lovely little assistant you have there, This is important family know weapons, not be afraid of them and know how to use one. There are a lot of 303 British in the USA.
First, let me say that I am impressed with your pistol shooting skills! Being a former instructor in shooting and balistics in the Swiss army (it was dubbed "shooting theory" then, and the school was in Walenstadt), I have the following comments: 1. The dispersion of any group of shots depends a) on the weapon/ammunition system's technical dispersion, and b) on the shooter's induced dispersion. Assuming a good shooter, the technical dispersion is predominent in, say assault rifles. With my Stgw 90 (Sig 550), which dispersion is given at 0.5 0/00, (actually a
My other way, is to throw a handfull of live ammo on the fire, stick a target above, and scientifically record results.After running rather fast away from said fire.
My hunting loads have always been fired hundreds of times to express accuracy. If a rifle of mine won't shoot 5 round groups of 3/4 MOA or less, it isn't going hunting with me. Yes, I am that eccentric about my accuracy, these are all shot from a bench and if I miss an animal in the field, it is my fault not the rifle or ammunition. I like to have other people shoot my rifles to see how they fair with the loads I build, most people are amazed at the accuracy and want to know how to do it. They are told it is time consuming and tedious, I do everything the same every time. All steps are carried out with the greatest accuracy I can put into them. Really like your videos and the range is is superb. VERY CUTE LITTLE GIRL.
The rifles that I have shot tend to be less accurate after cleaning with greater than 5 shots. Shots 6-10 would leave a less impressive group than shots 1-5.
Nothing compared to my Glock's accuracy. One shot groups are such, that every bullet goes trought the same hole!!! Thats 0.00000 moa up to the maximum range. PERFECTION.
I know when i was shooting full bore, (300-1600 mtrs) we used to weigh the projectiles as we loaded them because the heat "could" effect performance, especially at the longer ranges, and the weight of the projectiles and precise manufactured diameter ( measure each bullet) would be calculated into the load and into the firing sequence, gets into a density/friction( heat influence )/velocity/ballistic coefficient thing i never fully understood. all that's moot though, IMO, i've always done, with long arm or handgun, 3 - 5 ten shot groups, rested at appropriate range for caliber, to get a feel of the avg POI, then shoot the bloody thing for fun.
I guess, being a mathematician, I'm a bit puzzled by this. I think it would be better to say that it's a group that may or may not be indicative of the firearm's accuracy. The thing is you could also approach a bit down the line. Let's say I fired fifty rounds to produce a group, thinking that I have a definitive result. It could be that the probability of a round falling outside of the maximum diameter of that fifty-round group is quite low, but it's still a possibility. If I were to get picky -- and mathematicians have to be picky -- I would, by your reasoning, claim that a fifty-round group is not a group at all, because there is still a possibility of rounds falling outside of that group. No, Sir, I insist that you shout an infinite number of rounds to confirm the accuracy of your firearm.
Depending on the rifle I like to drop a full magazine (Or clip, rifle dependant) down range before I say "THIS is my group size" simply because of the probabilities. As low of a chance as it is that your shots land in the same hole 3x in a row, I like it to be that much LESS chance, and thus get a more accurate idea of what the rifle is capable of. I recently shot a .22lr at 100 yards and it gave me something like 60 inch groups, granted it was from WWII but still I was unimpressed especially since it was off of a rest, 12 shots per group.
If you are shooting 20 round mags then you should be looking at 20 round mag performance. The 5 shot group thing from benchrest, worked because the hunting problems were generally single shots spaced quite a bit in time, unless on ground squirrels or something.
One other reason why three round groups can be useful is adjusting the sights. If you fire at a grid (and not the target circle that you shot at), a three round group can tell you if you need to adjust your sights, and how (up or down, vs left or right) and by how much. Once you have your rifle (or pistol) adjusted to your sights, then you are that much more likely to hit a target at some distance. Another thing to consider is the distance to the target. Given that you were only, what, 25 yards away, chances are very likely that you'd miss almost completely at say 250 yards away. For that reason, soldiers are taught to make their shot groups as small as possible.
Best tip I've ever gotten to find out what a particular gun / shooter can do is to put up a fresh target backer, and then keep putting targets up with the POA centered at the same place each time. At the end of the session, the size of hole in the target backer is what that gun / shooter is actually capable of on any given shot.
Gotta somewhat agree, in the army (of Finland) we only used 3 rounds to get the sights right. Good enough, i guess, if you really focus on aiming, but more the better.
My understanding of a 3 round group was using a rifle on a bechrest-sighted in to "bullseye"-using the same brand/type of ammo, and measuring the spread. Shoot 3 more rounds at a different bullseye and measure and so on. This would indicate the rifles potential with minimal shooters skill. From there you can also test different types of ammo that the gun "likes" the best. Just my humble view! Great show you guys produce! thank you.
No, that's not how that works, in order for small sample sizes to be representative of the true accuracy. You need to fire dozens if not hundreds of 3-shot groups to estimate the true precision. Hopefully, you became smarter in the last 6 years, because this is bad form.
Not British and like I tell my 3 sons “Brit’s, well they just don’t tal right”. Love your channel. Great information. I’m smarte or having watched. That left side of the road thing, lie I told them. “1776 baby!,,,,
I'm a little confused. I must be thinking of a different three round group. the one I know of is only used while zeroing the weapon and you only use the last 3 or 5 consecutive shots to make adjustments not "any 3 out of the last 20"
Your do 3 shots from a cold bore and judge 3 holes in clean paper to zero in a rifle. Anyone can shot a thousand rounds at a piece of paper and say you clipped the same hole at least 3 times.
No, he's saying that a 3 round group doesn't provide enough data to show the precision/accuracy of a particular gun or ammo. You need to fire more shots in order to determine the potential accuracy for a particular gun or ammo. 5-10 shots is a better measure of how a gun or ammo performs.
Aren't there any apps for this? In comp archery there was an app that allowed one to log every shot on target and produce stats from that. Of course that was target shooting where every individual shot counts, not just a group. When I had my .222, I logged every group, including groups I had "fliers" with, because that is what actually happened. If you had a shot to save the world, you would want to know what the real capabilities of the combo were (off a rest in this case). I think I was at something like 1.25", or maybe .75 inches (about 30 years ago). I met a guy at a store who had the same gun, and we got talking. He asked what my gun was doing, and I gave him probably a flier less number, since that was what he was after, he was shocked, he said his gun shot 3/16" at 100 what was wrong with mine. (I was neck turning, using a turret press, etc, etc,) I remember because at the time he was claiming an average that had not yet been shot in comp in light varmint benchrest. LOL. Until the PPC came along and other stuff, it was hanging in .2 range for a while.
8th- MOA. I agree. Firing for scores must be 5 rnds in a grouping. My tightest is 0.120 Remington Sandaro 5R 100m. Most know by now my group was dumped by a 5Gp of, 0.102 .27 cal
Ok I totally get it. I was a sniper in the 82nd Airborne...But there are exceptions to every rule. If my rifle shoots the first three rounds into an inch, and then the next time I go out, it shoots the same, and the next time it does the same again then for my hunting purposes it is a one MOA rifle. If I need more than three rounds to kill a deer or an elk, then I really need more practice. Even more important to me is where the first round hits from a cold and clean bore. If that round hits the same place virtually every time, then the rifle is zeroed and I can be pretty much assured that I am good to go. Standing and shooting a .22 one handed in a target fashion is of course useful. But I don't do that in the field. When jackrabbit hunting with a .22 handgun for example, I try to take a rest, or at least use two hands. This is NOT meant to contradict your point here, but it is meant to say that firearms are used for different purposes. I no longer compete at target shooting. I'm certainly no longer a sniper. I'm a plinker and a hunter out in the desert. So for my purposes, I have several 1 MOA rifles (and a couple of quite sub-MOA ones) and several 2 MOA handguns...PS Your daughter is adorable. I bet she's all grown up by now!
I was taught in the army a 3 round group would be 3 rounds per target, if you shot the same target and didn't identify each shot individually between rounds it didn't count as a 3 round group. After you have a collection of 10 to 20 targets you have you're data. so if I was told 3 rounds stats and data about a weapon or ammo, I would assume that it was done on separate targets every group. never occurred to me that it would be mass amount of shots at once because it would be silly for said reason.
[Probably faulty post, ignore for purposes other than personal amusement] To put it like that: The confidence interval for a sample of three is +/- 56.6% (assuming 95% confidence level, ie. 19 out of 20 times, statistics shouldn't fail you). That means - to my amateurish interpretation - that you can assume that shots will land within a circle which is at worst 57% larger than the group you have shot (or make that "probably 57% larger", since one will usually pick a 'good' group for those calculations, and not just any random sample). A circle of 1.57 times the original area has a radius that is 1.25 times larger. Meaning, one-inch 3-shot groups are strictly speaking only indicative of 1.25-MOA accuracy. Or to turn it around: A sub MOA rifle should regularly be able to produce 3-shot groups where the bullets land within 0.8 inch of each other.
Ok, here I found someone with a similar approach to mine, even though he gives a different correction figure for three shot groups; unfortunately, he's very vague on the actual maths behind other than "I fixed me up some matlab and here's what Iit said", but maybe it gives you some ideas on the matter as well: www.the-long-family.com/Group%20size%20statistical%20analysis.pdf
The thing is that you misunderstand what a confidence interval is. A confidence interval tells how data points fall inside the metric. We are also measuring extreme spread, so I am not sure how to geometrically resolve that. However, central tendency is easy to see. It looks like 18/50 broke the 10 ring on the target and only 9 were outside the 9 ring. The mean radius (average distance from the center or average of the group) looks like 3 inches (middle of the 9 ring). The distribution clearly isn't Gaussian because the area of even probability is pretty big. It's kinda flat-topped. The concept holds, but the statistical math won't unless more and more rounds are fired and we find that this group was somehow mystically not representative. However, at 50 rounds that chance is quite slim.
Agreed, I think the big issue I noticed when using shooting results in proper statistics is to represent the data in a way that allows for the assumption of normal distribution - and I guess using rings or distance from center of group isn't automatically so (both being hard-limited scales, with rings possibly of arbitrary metrics). One might go for two variables (horizontal and vertical deviation), but that further complicates the maths and also the results. Might still be necessary for accurate representations - I suppose there is indeed often a noticable degree of independence between the two dimensions - i.e. vertical stringing, "this rifle has an accuracy of 1.75 MOA horizontal, 2.50 MOA vertical". I suppose the old "this thing shoots groups the size of a beercan at 100 yards all day long" is probably less problematic than any attempt to represent shooting performance in mathematical/statistical terms :)
Love your videos. Just discovered you channel today. What is the purpose of defining a "3-Round Group"? To test the accuracy of the gun? To test your performance at simple static marksmanship? The real test, the one that rally counts for 3-Round groups is not just static marksmanship, but when your target is s actually shooting back at you. It's like in America what they call the "Ultimate Fighting Championship". The REAL ultimate fighting championship is two guys fighting to the death. The last man standing is the winner.
It's the minimum number of shots you can fire and call it a group, and a lot of shooters will use it as a benchmark of the accuracy that the gun they're using is capable of. It's not truly representative, but it is common practice.
Thanks for the response. Have a question. Does testing accuracy of a gun, any gun require the use of human holding it? I don't think so. Humans move, ever so slightly. Lock the gun down on a bench and have an anchored machine manipulate the moving parts. Fire the gun. That is a truer test of the accuracy of a gun. That's what some gun manufactures actually do. That's what some Test Engineers actually do to their products. I understand that 3-shot groups are just an arbitrary standard. But get a machine to do it for 100 to 500 rounds is a better sample test.
***** It depends on whether you're testing for mechanical or practical accuracy. If it's scientific curiosity as to the mechanical interface of the gun's parts and consistent shot grouping, then locking the weapon into a weighted, stabilized rest would eliminate a variable. However, most shooters are looking to test practical or field accuracy (i.e. "How accurately can _I_ shoot this gun under normal field conditions?"), so they replicate the shooting position that they would normally use. Usually that's either supported from a bench or prone supported. This method really also tests how well the particular gun fits that shooter, which will affect how accurate the shooter can be with that gun. Not as scientific as from a locked rest certainly, but practical if you're just trying to get a measure for field or range use.
There are advantages to distinguishing between precision and accuracy. Precision is the size of a shot group, independent of how well centered the group is on the intended target. Accuracy is how well a shot group is centered on the intended target, independent of the size of the group. One can exist without the other, and errors in each have different remedies. Of course, the ideal situation is having both accuracy and precision.
That's one particular set of definitions, borrowed wholesale from metrology (where they were arbitrarily defined) but it's by far from universal in the literature (even the technical literature). It contradicts the Textbook of Small Arms 1929 definition of "accuracy", and also plenty of US army technical literature regarding permissible dispersion of ammunition types (see Hatcher's Notebook for examples). If you're an absolutist on it, there's no such thing as "accurate ammunition", only "precise ammunition". Or an "accurate rifle" for that matter, since it simply reduces "accuracy" to "correctly-zeroed". "My shot-out Mosin is really accurate - it shoots 12" groups at 100 yds but they're perfectly centred so it's perfectly accurate" said no-one ever... Nor did anyone ever write a book about "accurising" a rifle that was basically "here's how you set your sights so the point of aim and mean point of impact coincide"...
Hey Bloke, nice channel you have. There is one somewhat big problem though. The gunshots in your intro are very poorly synchronised. And since you show it every video(it's an intro after all), it might be a good idea to correct that.
Ah there's that word. . "capability"... I measure the distance between individual POI's and then work out mean, average, mode and SD. The sample size growing after every shoot.I then estimate what I believe the capability of my chosen firearm to be, in mil. From this I derive a USL and LSL and then derive a CpK to give me an idea of how consistently I make bullets fly through the air. My "combined accuracy" in one number: CpK. The real reason I do this is to avoid watching talent shows on TV with her indoors. In truth, I assess my capability by counting up the score, and then classifying it on a sliding scale from "shit, that gun is useless" to "Not bad, I think the brass is tired and neck tension is loose" to "Get in. Look at that group. Drinks on me". The ultimate target of having fun is always hit on my local range, thanks to the great people I shoot with.
Interesting indeed. I have never really group tested my M14 except to see how well I could shoot with irons. The answer is that I couldn't. However, I did shoot up some sporting clays (about 4.5 inches in diameter) with some PMC 147gr FMJ-BTs at 300 yards. A medium barrel, E2 stock, and a cheap 4-16 let me hit them all but a few times despite me being quite rusty and informally taught. I'd say I hit 17/20, with all the misses being slightly downwind (I didn't compensate for the rightward breeze). 4.5in/300 yards = 1.5 MOA. With that, I'm calling it 1.5 MOA with good FMJ, and that's plenty for me.
Just wanted to say Im surprised and impressed there is a a brit shooting channel mostly because of my understanding of UK gun laws. Could you do a video explaining gun laws where you live?
Nice Video. As you said. A Bad group answers questions even if its only two shots, a good group hardly does as long as it is less than say 5-7 shots. Which is why in 5 shot groups from reloaders there is always at least one flyer or"i jerked the trigger, felt that" shot haha.
5 rounds is the zeroing minimum I used when I was in the army during the 80s and one of the last times I fired full bore it was with an SA80 5 rounds. My first 2 shots were keyhole my 3rd I snatched and 4 & 5 were keyhole onto my first 2 shots. Which meant I had 4 shots in less than an inch BUT it was a 2 1/2in grouping because of my snatched 3rd shot. The fact I was good enough to pin my last 2 shots onto my 1st 2 is immaterial it's the 5 shots that count!!
in the US Army when we zero our rifles we usually shoot 2 magazines of 3 or 1 magazine of 5 before you start making adjustments and then after that we shoot three round groups and make the necessary adjustments. This is not saying the whole of the Army does it this way just the units I have been in usually do it in this sort of way
It's Zero the sights, fire two three round groups before making a change. Then 3 more and another three that are withing a certain diameter. So the least you will fire is 12 rounds and the most 18. (that is where you go back for retraining) So for a starting reference you have a 6 round group and confirming your sight adjustments another 6 round group. I had one rifle that never needed to be adjusted in 3 years with a battle sight of 0 right or left and 0 up or down. Now the National guard had some adapters to fire 22 rim fire out of the rifles and it was not too far off.
The point of a three-round burst isn't even to *be* sub MOA. When they teach you 3-round bursts on an AK, you're supposed to put the 3 rounds into a *reasonable* (30-ish cm) circle at 100m into the centre mass of a man-size target.
A bit off topic, but at 2:18 that's an interesting semi-old school one-handed stance -- neither the Weaver (or a variant) nor isosceles. That reminds me that I still can't help chuckling when watching old military and police handgun training films where they were taught to stand nearly perpendicular to the target with the off hand on the hip. Those trainees looked like they were in an 18th C. duelist's stance -- or they should have been holding a fencing foil instead of a handgun.
Great video. As a person who regularly deals with a variety of statistical distributions, it drives me crazy to see the tortured use of small sample sizes. I have a hard rule at work, that a measure of probability variation is useless without a percentile. EVERY firearm is a 1 MOA firearm. The question is, what percentage of shots will fall within that 1 MOA group? 5%? 50%? 90%? In my mind's eye, I picture a 3-d gaussian distribution, like having a mexican sombrero stapled to the target.
But what if you add a bench rest into the equation? Does that make the 3 round group more reliable? I understand that it still would not tell as much as a 5 shot or 10 shot group but….
no, you cant create a group of unconnected shots, you take 3 random shots from 20 - and thats not GROUP - that PLAIN AND SIMPLE LYING, FAKING THE RESULTS. What are you talking about is a non discussionable matter - 3 nonfollowing shots are not a group and group is a group no matter how many shots are fired as long as they are connected by no interruptions in aiming conditions, like chaning wind, rain etc. It makes me wonder, why you discuss a thing that has no logical or intelectual value.
I reload most of my ammo or with flintlocks weight the bullets, religiously measure the powder and never use half the flints that I purchase. Just to make my groups smaller. I can see the reasoning on why a 3 round group does not show much, but at what point of trying to reduce the size of the probability of impact does a 3 round group become valid?
the math only scales as far as you intend to shoot your rifle. If I were to only shoot my ak or m14 once every time I used it the only group that could represent my application of these rifles would be a one round group. Also remember the physical characteristics of your rifle change the longer you shoot it. Having said that I do agree with the core message this video tries to bring across.
When using the pistol like this no amount of shots will be representative of a group because it is not used in a controlled enviroment as the conditions when firing the pistol is always different from shot to shot i.e. : the pistol is never fired from the exact same position, the gun heats up and the shooter is experiencing an increasing amount of difficulty keeping the pistol on target due to the muscles tiring out so this doesn't prove anything other than the effect of shooter fatigue, heat buildup and other user errors over time on the accuracy and precision.
This is why I like Russian military doctrine. why teach a soldier about accuracy,statistics , bullet drop, windage, and parallax only to have the poor sod die a week later, when you can teach a soldier to spray and pray and replace him the moment he's down
And we all know from their history what value the Russians place on human life. Expendable civilians, soldiers, cosmonauts, etc. Its just one city of civilians when there are many more. Why train a few soldiers to complete a mission when you can expend an ignorant few thousand soldiers and complete it? Plenty of cosmonauts to expend until they get the job done. The Russians are not exactly a shining example of consumer product safety either as a result of their expendable human life doctrine. Life is cheap. A callous people.
Ok.... this wasn't even close. No one who seriously claims a certain accuracy fires 50 rounds at a target and then picks and chooses which three to look at. They fire sets of 3 and mark them. It's three consecutive rounds. Which, if several groups show the same grouping, it's accepted that the firearm is shooting with the described accuracy. And it drove me nuts your setting up a "take-down" on accuracy the way you were shooting. Single arm extended, on a windy day, with a swinging target.
An interesting point about firearm accuracy, though unrelated to the importance of a three shot group, would be to show the factory test target of that SP20. So many people discount their own lack of skill as a parameter to bad accuracy. Not a jab at you of course, it was obvious you were not going for your best performance in the video. The thought just came to me as I have some experience with ISSF pistol shooting.
Well, you may be having some problems with American military veteran lingo. When we say groups we literally mean groups of shots. When zeroing a rifle in our military we are handed a few 5 round magazine and shoot for three round groups. This literally means getting up after every five rounds and marking you target or replacing it. Idk if anyone else has pointed this out.
"Well, you may be having some problems with American military veteran lingo. When we say groups we literally mean groups of shots." So does everybody...
The UK legal straight pull rifles may actually be quite capable of consistent sub MOA, there's little reason they're not capable of this after all most use newly machined barrel's without the gas tube hole, they're the same barrel's you'll find on a good hunting or target rifle. MOA is measured in at 100 yards because sub MOA means it's capable of consistently (human error is taken out of the equation) firing a group of 3 or even 20+ shots with less than a 1 inch spread between the hits. Firing point blank and getting a group under 1" does not determine MOA, 1 minute of angle spreads out to 1" at 100 yards, 2" at 200 yards and so on. You can buy rifles that garauntee 3/8th" MOA, if it doesn't achieve this, they're obligated to send you a new one, but if you get something like a GAprecision rifle, most of which garauntee either 1/2 MOA or 3/8 MOA, or even a Tikka, which garauntee sub MOA and prove them wrong, that I would very much like to see. And if you don't, then consider you can have those same barrels machined to fit an AK or a mini 14, so there's no reason they can't be capable of sub MOA. And when you test MOA, you test each group on a fresh target, you don't fire them all together, but I get your point if this is what people tend to do to determine their rifle's to be sub MOA capable, it's certainly not an accurate or reliable test, even if it is sub MOA capable, you couldn't take their word on it based only on looking at what resembles swiss cheese more than anything.
Interesting video. I stopped shooting three round groups long ago to sight in scopes. Its just a waste of ammo if you are a confident shooter. I can usually get a rifle sighted with 2-4 rounds, then shoot a group to confirm, then have the guns owner shoot it to see if he's satisfied. Yeah, I have to sight all my friends rifles.
I like the Hammerli and I think you are my new best friend. How in heck did a British firearms enthusiast who knows a standard deviation from hole in the ground end up in Switzerland?
If my shots always hit where im aiming with my target rifle, and my 30 round group is the same as a 3 round group, then my 3 round group is sufficient enough to be noted?
@@BlokeontheRange I'm a bit late, but I'd assume described "30 round group is the same as a 3 round group" might be a case of shooting the target rifle at rather close range and not being able to measure groups with enough accuracy. (And measuring groups is whole different story).
A three round group will underestimate the real standard deviation by about 20%, i.e. you will get on average a deviation of 0.81, however with a variation of 0.3. So you will have very scattered results. Assuming a base distribution of the shots with a standard deviation of 1 then: a 2 round group will yield a deviation of 0.71 variation 0.37 a 3 round group will yield a deviation of 0.81 variation 0.30 a 5 round group will yield a deviation of 0.90 variation 0.23 a 10 round group will yield a deviation of 0.95 variation 0.16 a 20 round group will yield a deviation of 0.975 variation 0.12 a 50 round group will yield a deviation of 0.99 variation 0.07 a 100 round group will yield a deviation of 0.995 variation 0.05 For the 7 round group, you get a StDev of 0.92 instead of 1 with a variation of 0.2, i.e. your results will scatter between 0.7 and 1.1. To get reliability of 10% you need to shoot 20 round groups or more.
Do you need to wear a blind like that? I know with target shooting it is better if you don't have to keep one eye closed, but why can't you shoot with both eyes uncovered?
+Bloke on the Range oh many different trips of trial and error I found what my rifles liked. the most consistent with out sacrificing cost and efficiency
If it groups tight enough then in most cases yes. If not, you might end up chasing shots (and I grew up in a competitione environment where you only get 2 sighters...)
i love your videos but theres also a correlation between how fast you pull the trigger and how close you get to ten. i can hear the faster trigger pulls and they mostly arrive farther off of ten than any
Cold barrel, one round shots, are my best "group"
This always just made sense to me, and I don't understand why others have a problem with understanding.
A three round "group" is enough to tell me that I am a foot or more left/right/low/high. Five rounds is enough to start walking to a zero. Ten rounds gives me a pretty good zero. Twenty rounds confirms it.
When I think I've got a rifle with a good zero I start shooting 30,40,50,60 rounds at a target before I go down range and ask myself...Is my average POI a good match for my POA? Or do I give it one click...?
Great. If you shoot a hunting rifle and normal bought hunting ammo, you start new box after 20. A new box will have slightly different properties and that the end of meaningful statistics. But if you hunt you don't care about MOA. You need to place a shot i a bierdeckel from 0-200m. Setting of the sights makes greater uncertainty
I think unfortunately many of us are limited to 5-10 round groups for economical reasons…
@@danielcurtis1434 You've bought a $1000 gun, whats another $40 in ammo to get a truly correct zero and fully know your real accuracy.
@@eugenohanka "Slight different properties" is called the error. And its easily accounted for in statistics or probability since you can assume that factory ammo will have a +-15 SD and seating depth along with BC's that vary by .001-003. Kind of a moot to bring it up.
But fundamentally, you're mistakenly assuming that each box has a different error when in reality each bullet from every box will defer depending on manufacturing conditions and quality control. That's why you keep shooting until you get a significant sample size.
@@eugenohanka”setting sights leads to greater uncertainity” not at all, your sights have to be set somewhere so adjusting them can’t make it worse.
To account for box/batch variation you just take rounds from multiple boxes and use that when you’re doing your groupings. You can either just buy a load of boxes at once or every time you get a new box put a round or two to one side for testing (the latter is better bc it stops you buying boxes from the same batch, but you have to wait a while before you’ve got enough rounds aces up to get a proper zero)
You have to admit, though, going blue in the face and frothing at the mouth would be visually interesting.
Some of us would still like to hear the maths. Then shooting anyway.
+Bloke on the Range I still am waiting for the John Cleese to come out.
please, bon't but us in a boncenbation bamp
Neil Wilson 3 years late but frothing and the gash is my preferred method of frothing
I'm still surprised that some people don't understand what you're saying. Great video, though! Always a pleasure to watch :)
I read an article somewhere (probably TFB) about the US Army's ammunition batch testing. They buy lots of 1 million or more and randomly pull 1,000 rounds. They fire those rounds and require half the shots to land inside a circle (like 2 MoA, I believe). The writer went into the statistics of the test, showing that a 3 round group was about 66% representative of any ammo/gun combo, a 5 round group was around 90%, and 10 round group somewhere around 95% and the 1,000 round group was extremely representative.
Are those tests done with a fixed gun in a mechanical test rig or are they done with an actual human and a precision rifle?
Because only the former would produce scientifically accurate results for just the ammo.
SonsOfLorgar Mechanical test rig. Not sure if it is a rifle in a test rig or a .5.56 chambered barrel in some sort of fixture, but it was not some guy holding a rifle and burning up 1,000 rounds, lol.
Sunshine_Shooter Well, I never assume thinking when actual officers are involved in the process ;)
At least in this case they wouldn't have a map and a Compass!
SonsOfLorgar I can only assume that actual engineers were consulted. :)
Sunshine_Shooter So in other words two separate three round groups are more than sufficient for any and all practical purposes.
Once I saw the normal distribution, I know what it is right away. Sample size is too small. you have to see where the mean, mode, standard deviation and median.
They'll be close enough as makes no significant difference if you shoot a representative sample size :)
In Germany we call it the Mexican sniper. With enough shots fired, you will eventually have some very close to each other. If you only show off that one group of close shots, without pointing out that you shot 100 times to get those, then you committed the fallacy of selection bias. The result is insignificant unless you can repeat it consistently.
PS: 4:50 "fewer" ;)
Also known as the Texas Sharpshooter myth. The Texas sharpshooter shoots into the side of his barn and then paints bullseyes around the clusters.
I don't think that your, "Fewer" comment was appreciated--at all. In fact, your message had--at least--one error, but I don't think you want me to point it out.
No, English traditionally uses less; try reading historical documents and you will find out which is correct. One of these stupid incorrect pedantries, like octopi.
A principle which seems to be prevalent even in "professional", well-funded trials - because everyone tends to be biased - is, "once you have the results you want, stop immediately and call it a day." So if, after three rounds, you have a good looking group, why continue and risk spoiling the results? Lies, damned lies and statistics indeed. I subbed after this video.
I have just read through the comments on this video and your many responses and I have to congratulate you on staying calm and polite in the face of extreme and repeated provocation. Well done, sir! Good video, too.
People talk about three round groups because it makes them feel good. "My new rifle shoots three round groups that can be covered by a dime. Those other shots are fliers, cause by cosmic rays hitting the bullets mid flight."
We also do it to keep new shooters from getting discouraged. "Hey! You have some nice three round groups here. The rest of those shots are what we call fliers."
I had assumed that you were in the UK. If that's true, how do you manage to have that handgun, an evil killing machine just waiting for a chance to run amok?
But you're a bloke? There's only blokes in the UK I thought.
My whole life is a lie.
I still go back to this video when someone I know says their gun is such and such accurate with a three round group. I love this video because of it.
for a hunting rifle as well as others, depending on purpose and use, shooting one shell from a cold barrel , then repeating the next day, etc. gives you a real world picture of what the gun will do when shooting from a cold barrel...a first shoot capability.. Good job on taking on this explanation !
Nice explanation of probability theory with no equations! Well done for 'blokesplaining' a subject which people often find really difficult.
god you look so comfy shooting that thing, i sincerely hope you were wearing house slippers.
But also yes, very good point!
If you're on a budget a 5-10 group isn't too bad to gauge your own MOA. Then again, usually the shooter has more impact on MOA than the firearm.
Fuck everyone's MOA and such nonsense.
Consistency, confidence, and experience in what your rifle can do with what you feed it is always best.
Also hits are what count.
If the rifle is one built to a military spec, all I ask is minute of man. I.E., I should be able to reliably hit a torso sized target at 100 yards with iron sights and whatever cheapo ammo I feel like throwing down range.
Hunting rifles should be a bit more accurate.
Milspec should be between 2-3 MOA. Hunting rifles certainly should be more precise.
MOA is an important measurement for long-distance shooting, but people tend to use it as a way to brag or lie to themselves.
The trick isn't to hit the target perfectly.
It's to be able to hit it repeatedly while it's moving, camoflauged, poorly lit, in the wind, rain, etc. and to have fast follow-up shots because you WILL miss often under those real life conditions.
I use reinforced camo targabots or I hunt. I particularly like hunting in the rain, since it tests my gear and makes it harder to track.
Cube210 I am confident that I can hit a 10 pin bowling ball at 100 yards with every firearm I would expect to.
great simple explanation. Recently I put together a 1000yrd rifle, savage 10 fcp-sr in 6.5 creedmoor- very accurate rifle. It will consistently print .3-.4" 5 shot groups@ 200 yrds. yesterday went to the range, using my most consistent hand load and shot a 10 round group and it measured just under an MOA, so a significant change in group size on a larger sampling.
More factors can come into play with increased group size: barrel heat, wind changes, eye fatigue, etc, etc. It’s really not possible to measure inherent accuracy by shooting groups of any size, only relative accuracy.
Lovely little assistant you have there, This is important family know weapons, not be afraid of them and know how to use one.
There are a lot of 303 British in the USA.
First, let me say that I am impressed with your pistol shooting skills!
Being a former instructor in shooting and balistics in the Swiss army (it was dubbed "shooting theory" then, and the school was in Walenstadt), I have the following comments:
1. The dispersion of any group of shots depends a) on the weapon/ammunition system's technical dispersion, and b) on the shooter's induced dispersion.
Assuming a good shooter, the technical dispersion is predominent in, say assault rifles. With my Stgw 90 (Sig 550), which dispersion is given at 0.5 0/00, (actually a
I learned more in the last ten seconds of this video than I think I have in my whole life.
My other way, is to throw a handfull of live ammo on the fire, stick a target above, and scientifically record results.After running rather fast away from said fire.
My hunting loads have always been fired hundreds of times to express accuracy. If a rifle of mine won't shoot 5 round groups of 3/4 MOA or less, it isn't going hunting with me. Yes, I am that eccentric about my accuracy, these are all shot from a bench and if I miss an animal in the field, it is my fault not the rifle or ammunition. I like to have other people shoot my rifles to see how they fair with the loads I build, most people are amazed at the accuracy and want to know how to do it. They are told it is time consuming and tedious, I do everything the same every time. All steps are carried out with the greatest accuracy I can put into them.
Really like your videos and the range is is superb. VERY CUTE LITTLE GIRL.
Quite! As one of my favourite authors from days of yore (Robert Heinlein) once observed "Little girls and butterflies don't need a reason."
This is one reason why the NRA does five 5 round groups to come out with a median.
The rifles that I have shot tend to be less accurate after cleaning with greater than 5 shots. Shots 6-10 would leave a less impressive group than shots 1-5.
Nothing compared to my Glock's accuracy. One shot groups are such, that every bullet goes trought the same hole!!! Thats 0.00000 moa up to the maximum range. PERFECTION.
DudeNumberOnePlus my 1911 doesnt group as good one shot group cuz it leave fat 45 holes and your 9mm glock leaves a smaller 1 shot moa hole 😞
Love the classic marksmanship, amd always love your presentations
I know when i was shooting full bore, (300-1600 mtrs) we used to weigh the projectiles as we loaded them because the heat "could" effect performance, especially at the longer ranges, and the weight of the projectiles and precise manufactured diameter ( measure each bullet) would be calculated into the load and into the firing sequence, gets into a density/friction( heat influence )/velocity/ballistic coefficient thing i never fully understood. all that's moot though, IMO, i've always done, with long arm or handgun, 3 - 5 ten shot groups, rested at appropriate range for caliber, to get a feel of the avg POI, then shoot the bloody thing for fun.
I guess, being a mathematician, I'm a bit puzzled by this. I think it would be better to say that it's a group that may or may not be indicative of the firearm's accuracy. The thing is you could also approach a bit down the line. Let's say I fired fifty rounds to produce a group, thinking that I have a definitive result. It could be that the probability of a round falling outside of the maximum diameter of that fifty-round group is quite low, but it's still a possibility. If I were to get picky -- and mathematicians have to be picky -- I would, by your reasoning, claim that a fifty-round group is not a group at all, because there is still a possibility of rounds falling outside of that group. No, Sir, I insist that you shout an infinite number of rounds to confirm the accuracy of your firearm.
Depending on the rifle I like to drop a full magazine (Or clip, rifle dependant) down range before I say "THIS is my group size" simply because of the probabilities. As low of a chance as it is that your shots land in the same hole 3x in a row, I like it to be that much LESS chance, and thus get a more accurate idea of what the rifle is capable of. I recently shot a .22lr at 100 yards and it gave me something like 60 inch groups, granted it was from WWII but still I was unimpressed especially since it was off of a rest, 12 shots per group.
If you are shooting 20 round mags then you should be looking at 20 round mag performance. The 5 shot group thing from benchrest, worked because the hunting problems were generally single shots spaced quite a bit in time, unless on ground squirrels or something.
Exactly why I use 5 round groups sighting in a 100m and 20 round groups to evaluate load/gun combinations
Blokette is the star, like you and Chappie too but she is precious, reminds me of my now grown daughters.
One other reason why three round groups can be useful is adjusting the sights. If you fire at a grid (and not the target circle that you shot at), a three round group can tell you if you need to adjust your sights, and how (up or down, vs left or right) and by how much. Once you have your rifle (or pistol) adjusted to your sights, then you are that much more likely to hit a target at some distance.
Another thing to consider is the distance to the target. Given that you were only, what, 25 yards away, chances are very likely that you'd miss almost completely at say 250 yards away. For that reason, soldiers are taught to make their shot groups as small as possible.
This is a fantastic explanation of something few people in the gun community understand, I might just use it as a reference.
Thanks. Lots of comments completely misunderstand it though. Have a browse...
Best tip I've ever gotten to find out what a particular gun / shooter can do is to put up a fresh target backer, and then keep putting targets up with the POA centered at the same place each time. At the end of the session, the size of hole in the target backer is what that gun / shooter is actually capable of on any given shot.
Gotta somewhat agree, in the army (of Finland) we only used 3 rounds to get the sights right. Good enough, i guess, if you really focus on aiming, but more the better.
Am I the only one who got the Monty Python reference at 1:04?
ProudToBeNoob
What is that pistol that you're using in the video?
My understanding of a 3 round group was using a rifle on a bechrest-sighted in to "bullseye"-using the same brand/type of ammo, and measuring the spread. Shoot 3 more rounds at a different bullseye and measure and so on. This would indicate the rifles potential with minimal shooters skill. From there you can also test different types of ammo that the gun "likes" the best. Just my humble view! Great show you guys produce! thank you.
No, that's not how that works, in order for small sample sizes to be representative of the true accuracy. You need to fire dozens if not hundreds of 3-shot groups to estimate the true precision. Hopefully, you became smarter in the last 6 years, because this is bad form.
I like what you do. keep it up bloke!
Not British and like I tell my 3 sons “Brit’s, well they just don’t tal right”. Love your channel. Great information. I’m smarte or having watched. That left side of the road thing, lie I told them. “1776 baby!,,,,
I'm a little confused. I must be thinking of a different three round group. the one I know of is only used while zeroing the weapon and you only use the last 3 or 5 consecutive shots to make adjustments not "any 3 out of the last 20"
Your do 3 shots from a cold bore and judge 3 holes in clean paper to zero in a rifle. Anyone can shot a thousand rounds at a piece of paper and say you clipped the same hole at least 3 times.
***** ah, I see
guy is amateur, level super pro self dellusional
Bloke on the Range so you're talking about something completely different. you doubt their ability to shoot accurately, not the accuracy of the gun.
No, he's saying that a 3 round group doesn't provide enough data to show the precision/accuracy of a particular gun or ammo. You need to fire more shots in order to determine the potential accuracy for a particular gun or ammo. 5-10 shots is a better measure of how a gun or ammo performs.
Good stuff yet again, Bloke.
Your little girl is adorable, too.
Betcha will be teaching her the finer points of marksmanship as she grows up.
Aren't there any apps for this? In comp archery there was an app that allowed one to log every shot on target and produce stats from that. Of course that was target shooting where every individual shot counts, not just a group.
When I had my .222, I logged every group, including groups I had "fliers" with, because that is what actually happened. If you had a shot to save the world, you would want to know what the real capabilities of the combo were (off a rest in this case). I think I was at something like 1.25", or maybe .75 inches (about 30 years ago).
I met a guy at a store who had the same gun, and we got talking. He asked what my gun was doing, and I gave him probably a flier less number, since that was what he was after, he was shocked, he said his gun shot 3/16" at 100 what was wrong with mine. (I was neck turning, using a turret press, etc, etc,) I remember because at the time he was claiming an average that had not yet been shot in comp in light varmint benchrest. LOL. Until the PPC came along and other stuff, it was hanging in .2 range for a while.
It’s not a symmetrical bell curve. It’s a positively skewed Bell curve with zero on the x axis representing bulls eye.
8th- MOA. I agree. Firing for scores must be 5 rnds in a grouping. My tightest is 0.120 Remington Sandaro 5R 100m. Most know by now my group was dumped by a 5Gp of, 0.102 .27 cal
Ok I totally get it. I was a sniper in the 82nd Airborne...But there are exceptions to every rule. If my rifle shoots the first three rounds into an inch, and then the next time I go out, it shoots the same, and the next time it does the same again then for my hunting purposes it is a one MOA rifle. If I need more than three rounds to kill a deer or an elk, then I really need more practice. Even more important to me is where the first round hits from a cold and clean bore. If that round hits the same place virtually every time, then the rifle is zeroed and I can be pretty much assured that I am good to go. Standing and shooting a .22 one handed in a target fashion is of course useful. But I don't do that in the field. When jackrabbit hunting with a .22 handgun for example, I try to take a rest, or at least use two hands. This is NOT meant to contradict your point here, but it is meant to say that firearms are used for different purposes. I no longer compete at target shooting. I'm certainly no longer a sniper. I'm a plinker and a hunter out in the desert. So for my purposes, I have several 1 MOA rifles (and a couple of quite sub-MOA ones) and several 2 MOA handguns...PS Your daughter is adorable. I bet she's all grown up by now!
I was taught in the army a 3 round group would be 3 rounds per target, if you shot the same target and didn't identify each shot individually between rounds it didn't count as a 3 round group. After you have a collection of 10 to 20 targets you have you're data. so if I was told 3 rounds stats and data about a weapon or ammo, I would assume that it was done on separate targets every group. never occurred to me that it would be mass amount of shots at once because it would be silly for said reason.
Three round groups are usually sufficient for hunting rifles. Five round groups are generally for target competition rifles
Not bad shooting but your form could be improved. Hold 6 inches lower and turn the firearms about 90 degrees sideways.
[Probably faulty post, ignore for purposes other than personal amusement] To put it like that: The confidence interval for a sample of three is +/- 56.6% (assuming 95% confidence level, ie. 19 out of 20 times, statistics shouldn't fail you). That means - to my amateurish interpretation - that you can assume that shots will land within a circle which is at worst 57% larger than the group you have shot (or make that "probably 57% larger", since one will usually pick a 'good' group for those calculations, and not just any random sample). A circle of 1.57 times the original area has a radius that is 1.25 times larger.
Meaning, one-inch 3-shot groups are strictly speaking only indicative of 1.25-MOA accuracy. Or to turn it around: A sub MOA rifle should regularly be able to produce 3-shot groups where the bullets land within 0.8 inch of each other.
"Rusty" also describes mine quite well, I'm surely no authority on the subject either, but I'm getting back into it right as we speak :).
Ok, here I found someone with a similar approach to mine, even though he gives a different correction figure for three shot groups; unfortunately, he's very vague on the actual maths behind other than "I fixed me up some matlab and here's what Iit said", but maybe it gives you some ideas on the matter as well: www.the-long-family.com/Group%20size%20statistical%20analysis.pdf
The thing is that you misunderstand what a confidence interval is. A confidence interval tells how data points fall inside the metric. We are also measuring extreme spread, so I am not sure how to geometrically resolve that. However, central tendency is easy to see. It looks like 18/50 broke the 10 ring on the target and only 9 were outside the 9 ring. The mean radius (average distance from the center or average of the group) looks like 3 inches (middle of the 9 ring).
The distribution clearly isn't Gaussian because the area of even probability is pretty big. It's kinda flat-topped. The concept holds, but the statistical math won't unless more and more rounds are fired and we find that this group was somehow mystically not representative. However, at 50 rounds that chance is quite slim.
Agreed, I think the big issue I noticed when using shooting results in proper statistics is to represent the data in a way that allows for the assumption of normal distribution - and I guess using rings or distance from center of group isn't automatically so (both being hard-limited scales, with rings possibly of arbitrary metrics). One might go for two variables (horizontal and vertical deviation), but that further complicates the maths and also the results. Might still be necessary for accurate representations - I suppose there is indeed often a noticable degree of independence between the two dimensions - i.e. vertical stringing, "this rifle has an accuracy of 1.75 MOA horizontal, 2.50 MOA vertical". I suppose the old "this thing shoots groups the size of a beercan at 100 yards all day long" is probably less problematic than any attempt to represent shooting performance in mathematical/statistical terms :)
I've always heard that for target shooting the 3 round group was due to barrel heat affecting accuracy
You may have heard that, but the person was telling you lies.
Sunshine_Shooter nope, barrel heat does affect subsequent shots. The hotter it gets the more spread you’ll see.
So I assume the fireplace has molten lead at the bottom
The advantage of a 3 round group is that there are no outliers. What's best: a 7 round group + 2 removed outliers or three 3 round groups?
Love your videos.
Just discovered you channel today.
What is the purpose of defining a "3-Round Group"?
To test the accuracy of the gun?
To test your performance at simple static marksmanship?
The real test, the one that rally counts for 3-Round groups
is not just static marksmanship, but
when your target is s actually shooting back at you.
It's like in America what they call the "Ultimate Fighting Championship".
The REAL ultimate fighting championship is two guys fighting to the death.
The last man standing is the winner.
It's the minimum number of shots you can fire and call it a group, and a lot of shooters will use it as a benchmark of the accuracy that the gun they're using is capable of. It's not truly representative, but it is common practice.
Thanks for the response.
Have a question.
Does testing accuracy of a gun, any gun require the use of human holding it?
I don't think so.
Humans move, ever so slightly.
Lock the gun down on a bench and have an anchored machine manipulate the moving parts.
Fire the gun.
That is a truer test of the accuracy of a gun.
That's what some gun manufactures actually do.
That's what some Test Engineers actually do to their products.
I understand that 3-shot groups are just an arbitrary standard.
But get a machine to do it for 100 to 500 rounds is a better sample test.
***** It depends on whether you're testing for mechanical or practical accuracy. If it's scientific curiosity as to the mechanical interface of the gun's parts and consistent shot grouping, then locking the weapon into a weighted, stabilized rest would eliminate a variable. However, most shooters are looking to test practical or field accuracy (i.e. "How accurately can _I_ shoot this gun under normal field conditions?"), so they replicate the shooting position that they would normally use. Usually that's either supported from a bench or prone supported. This method really also tests how well the particular gun fits that shooter, which will affect how accurate the shooter can be with that gun. Not as scientific as from a locked rest certainly, but practical if you're just trying to get a measure for field or range use.
simonferrer
Thanks.
I am an Engineer and sometimes that Math Geek mentality creeps comes out once in a while.
Vurweapon Blog does 10 round initial and then checks again after 1,000 round to see if it is repeatable.
There are advantages to distinguishing between precision and accuracy. Precision is the size of a shot group, independent of how well centered the group is on the intended target. Accuracy is how well a shot group is centered on the intended target, independent of the size of the group. One can exist without the other, and errors in each have different remedies. Of course, the ideal situation is having both accuracy and precision.
That's one particular set of definitions, borrowed wholesale from metrology (where they were arbitrarily defined) but it's by far from universal in the literature (even the technical literature). It contradicts the Textbook of Small Arms 1929 definition of "accuracy", and also plenty of US army technical literature regarding permissible dispersion of ammunition types (see Hatcher's Notebook for examples). If you're an absolutist on it, there's no such thing as "accurate ammunition", only "precise ammunition". Or an "accurate rifle" for that matter, since it simply reduces "accuracy" to "correctly-zeroed". "My shot-out Mosin is really accurate - it shoots 12" groups at 100 yds but they're perfectly centred so it's perfectly accurate" said no-one ever... Nor did anyone ever write a book about "accurising" a rifle that was basically "here's how you set your sights so the point of aim and mean point of impact coincide"...
Hey Bloke, nice channel you have. There is one somewhat big problem though. The gunshots in your intro are very poorly synchronised. And since you show it every video(it's an intro after all), it might be a good idea to correct that.
Ah there's that word. . "capability"... I measure the distance between individual POI's and then work out mean, average, mode and SD. The sample size growing after every shoot.I then estimate what I believe the capability of my chosen firearm to be, in mil. From this I derive a USL and LSL and then derive a CpK to give me an idea of how consistently I make bullets fly through the air. My "combined accuracy" in one number: CpK. The real reason I do this is to avoid watching talent shows on TV with her indoors. In truth, I assess my capability by counting up the score, and then classifying it on a sliding scale from "shit, that gun is useless" to "Not bad, I think the brass is tired and neck tension is loose" to "Get in. Look at that group. Drinks on me". The ultimate target of having fun is always hit on my local range, thanks to the great people I shoot with.
Interesting indeed.
I have never really group tested my M14 except to see how well I could shoot with irons.
The answer is that I couldn't.
However, I did shoot up some sporting clays (about 4.5 inches in diameter) with some PMC 147gr FMJ-BTs at 300 yards. A medium barrel, E2 stock, and a cheap 4-16 let me hit them all but a few times despite me being quite rusty and informally taught.
I'd say I hit 17/20, with all the misses being slightly downwind (I didn't compensate for the rightward breeze). 4.5in/300 yards = 1.5 MOA.
With that, I'm calling it 1.5 MOA with good FMJ, and that's plenty for me.
Just wanted to say Im surprised and impressed there is a a brit shooting channel mostly because of my understanding of UK gun laws. Could you do a video explaining gun laws where you live?
My eyes! the white balance is burning them
Do 3 round burst fire rifles fire a proper 3 round grouping though?
Nice Video. As you said. A Bad group answers questions even if its only two shots, a good group hardly does as long as it is less than say 5-7 shots.
Which is why in 5 shot groups from reloaders there is always at least one flyer or"i jerked the trigger, felt that" shot haha.
Thanks. You'd be amazed how many comments completely miss that quite simple point, lol...
5 rounds is the zeroing minimum I used when I was in the army during the 80s and one of the last times I fired full bore it was with an SA80 5 rounds. My first 2 shots were keyhole my 3rd I snatched and 4 & 5 were keyhole onto my first 2 shots. Which meant I had 4 shots in less than an inch BUT it was a 2 1/2in grouping because of my snatched 3rd shot. The fact I was good enough to pin my last 2 shots onto my 1st 2 is immaterial it's the 5 shots that count!!
On the strength of this video I subscribed to your channel.
in the US Army when we zero our rifles we usually shoot 2 magazines of 3 or 1 magazine of 5 before you start making adjustments and then after that we shoot three round groups and make the necessary adjustments. This is not saying the whole of the Army does it this way just the units I have been in usually do it in this sort of way
It's Zero the sights, fire two three round groups before making a change. Then 3 more and another three that are withing a certain diameter. So the least you will fire is 12 rounds and the most 18. (that is where you go back for retraining) So for a starting reference you have a 6 round group and confirming your sight adjustments another 6 round group. I had one rifle that never needed to be adjusted in 3 years with a battle sight of 0 right or left and 0 up or down. Now the National guard had some adapters to fire 22 rim fire out of the rifles and it was not too far off.
The point of a three-round burst isn't even to *be* sub MOA. When they teach you 3-round bursts on an AK, you're supposed to put the 3 rounds into a *reasonable* (30-ish cm) circle at 100m into the centre mass of a man-size target.
A bit off topic, but at 2:18 that's an interesting semi-old school one-handed stance -- neither the Weaver (or a variant) nor isosceles. That reminds me that I still can't help chuckling when watching old military and police handgun training films where they were taught to stand nearly perpendicular to the target with the off hand on the hip. Those trainees looked like they were in an 18th C. duelist's stance -- or they should have been holding a fencing foil instead of a handgun.
Great video. As a person who regularly deals with a variety of statistical distributions, it drives me crazy to see the tortured use of small sample sizes.
I have a hard rule at work, that a measure of probability variation is useless without a percentile. EVERY firearm is a 1 MOA firearm. The question is, what percentage of shots will fall within that 1 MOA group? 5%? 50%? 90%?
In my mind's eye, I picture a 3-d gaussian distribution, like having a mexican sombrero stapled to the target.
But what if you add a bench rest into the equation? Does that make the 3 round group more reliable? I understand that it still would not tell as much as a 5 shot or 10 shot group but….
no, you cant create a group of unconnected shots, you take 3 random shots from 20 - and thats not GROUP - that PLAIN AND SIMPLE LYING, FAKING THE RESULTS.
What are you talking about is a non discussionable matter - 3 nonfollowing shots are not a group and group is a group no matter how many shots are fired as long as they are connected by no interruptions in aiming conditions, like chaning wind, rain etc.
It makes me wonder, why you discuss a thing that has no logical or intelectual value.
Bloke! You've tried to teach the blind "to paint by numbers."
That has to be the most anemic sounding pistol I have ever heard.
RK62/92 - had 3 round groups everytime I wanted. I think depends alot what rifle youre shooting with.
And the skill ofcourse..
I reload most of my ammo or with flintlocks weight the bullets, religiously measure the powder and never use half the flints that I purchase. Just to make my groups smaller. I can see the reasoning on why a 3 round group does not show much, but at what point of trying to reduce the size of the probability of impact does a 3 round group become valid?
the math only scales as far as you intend to shoot your rifle. If I were to only shoot my ak or m14 once every time I used it the only group that could represent my application of these rifles would be a one round group. Also remember the physical characteristics of your rifle change the longer you shoot it. Having said that I do agree with the core message this video tries to bring across.
When using the pistol like this no amount of shots will be representative of a group because it is not used in a controlled enviroment as the conditions when firing the pistol is always different from shot to shot i.e. : the pistol is never fired from the exact same position, the gun heats up and the shooter is experiencing an increasing amount of difficulty keeping the pistol on target due to the muscles tiring out so this doesn't prove anything other than the effect of shooter fatigue, heat buildup and other user errors over time on the accuracy and precision.
This is why I like Russian military doctrine.
why teach a soldier about accuracy,statistics , bullet drop, windage, and parallax only to have the poor sod die a week later,
when you can teach a soldier to spray and pray and replace him the moment he's down
And we all know from their history what value the Russians place on human life. Expendable civilians, soldiers, cosmonauts, etc. Its just one city of civilians when there are many more. Why train a few soldiers to complete a mission when you can expend an ignorant few thousand soldiers and complete it? Plenty of cosmonauts to expend until they get the job done. The Russians are not exactly a shining example of consumer product safety either as a result of their expendable human life doctrine. Life is cheap. A callous people.
Life is cheap indeed.
This is however no longer true. Russia's demography had changed, manpower is now precious commodity.
Ok.... this wasn't even close. No one who seriously claims a certain accuracy fires 50 rounds at a target and then picks and chooses which three to look at. They fire sets of 3 and mark them. It's three consecutive rounds. Which, if several groups show the same grouping, it's accepted that the firearm is shooting with the described accuracy. And it drove me nuts your setting up a "take-down" on accuracy the way you were shooting. Single arm extended, on a windy day, with a swinging target.
An interesting point about firearm accuracy, though unrelated to the importance of a three shot group, would be to show the factory test target of that SP20. So many people discount their own lack of skill as a parameter to bad accuracy. Not a jab at you of course, it was obvious you were not going for your best performance in the video. The thought just came to me as I have some experience with ISSF pistol shooting.
you have an interesting stance when shooting that pistol, any particular reason for it?
Really enjoyed this video. This is the one that earned my subscription :p
Hey Bloke man! Just uploaded a response (more reinforcing) video to this! Let us know what you think! ua-cam.com/video/v40Oj__Ygug/v-deo.html
so why do some rifles have 3 burst mode?
Well, you may be having some problems with American military veteran lingo. When we say groups we literally mean groups of shots.
When zeroing a rifle in our military we are handed a few 5 round magazine and shoot for three round groups. This literally means getting up after every five rounds and marking you target or replacing it. Idk if anyone else has pointed this out.
"Well, you may be having some problems with American military veteran lingo. When we say groups we literally mean groups of shots."
So does everybody...
The UK legal straight pull rifles may actually be quite capable of consistent sub MOA, there's little reason they're not capable of this after all most use newly machined barrel's without the gas tube hole, they're the same barrel's you'll find on a good hunting or target rifle. MOA is measured in at 100 yards because sub MOA means it's capable of consistently (human error is taken out of the equation) firing a group of 3 or even 20+ shots with less than a 1 inch spread between the hits. Firing point blank and getting a group under 1" does not determine MOA, 1 minute of angle spreads out to 1" at 100 yards, 2" at 200 yards and so on. You can buy rifles that garauntee 3/8th" MOA, if it doesn't achieve this, they're obligated to send you a new one, but if you get something like a GAprecision rifle, most of which garauntee either 1/2 MOA or 3/8 MOA, or even a Tikka, which garauntee sub MOA and prove them wrong, that I would very much like to see. And if you don't, then consider you can have those same barrels machined to fit an AK or a mini 14, so there's no reason they can't be capable of sub MOA. And when you test MOA, you test each group on a fresh target, you don't fire them all together, but I get your point if this is what people tend to do to determine their rifle's to be sub MOA capable, it's certainly not an accurate or reliable test, even if it is sub MOA capable, you couldn't take their word on it based only on looking at what resembles swiss cheese more than anything.
Interesting video. I stopped shooting three round groups long ago to sight in scopes. Its just a waste of ammo if you are a confident shooter. I can usually get a rifle sighted with 2-4 rounds, then shoot a group to confirm, then have the guns owner shoot it to see if he's satisfied. Yeah, I have to sight all my friends rifles.
I do check conditions of a range with three shots, then do my adjustments.
Must say I am a lazy shooter so use only standard ammo for my Fass57.
Waiting another ten or so years so the CinC can take her turn as your "Annie Oakley"
I personally shoot ten round groups. It just provides a better picture than a five round group.
I do this too. It gives me an excuse to shoot more. And then I will shoot ten round groups sitting, prone, standing.
One MOA all day long.... Great video.
"Group" is such a nice and woody word, isn't it?
And still, a of three draws from a random distribution is informative ... more so than a sample of two...
I like the Hammerli and I think you are my new best friend. How in heck did a British firearms enthusiast who knows a standard deviation from hole in the ground end up in Switzerland?
If my shots always hit where im aiming with my target rifle, and my 30 round group is the same as a 3 round group, then my 3 round group is sufficient enough to be noted?
What you're suggesting is statistically impossible.
@@BlokeontheRange I'm a bit late, but I'd assume described "30 round group is the same as a 3 round group" might be a case of shooting the target rifle at rather close range and not being able to measure groups with enough accuracy. (And measuring groups is whole different story).
A three round group will underestimate the real standard deviation by about 20%, i.e. you will get on average a deviation of 0.81, however with a variation of 0.3. So you will have very scattered results. Assuming a base distribution of the shots with a standard deviation of 1 then:
a 2 round group will yield a deviation of 0.71 variation 0.37
a 3 round group will yield a deviation of 0.81 variation 0.30
a 5 round group will yield a deviation of 0.90 variation 0.23
a 10 round group will yield a deviation of 0.95 variation 0.16
a 20 round group will yield a deviation of 0.975 variation 0.12
a 50 round group will yield a deviation of 0.99 variation 0.07
a 100 round group will yield a deviation of 0.995 variation 0.05
For the 7 round group, you get a StDev of 0.92 instead of 1 with a variation of 0.2, i.e. your results will scatter between 0.7 and 1.1. To get reliability of 10% you need to shoot 20 round groups or more.
The only rifle I own that I stop at 3 shots has yet to have a 3 round group at 100 yards that were over an inch. My others get 5 rounds groups.
Do you need to wear a blind like that? I know with target shooting it is better if you don't have to keep one eye closed, but why can't you shoot with both eyes uncovered?
A box usually tells me what the ammo is doing. 300win 30.06 and 45-70 I use the same manufacturer and the same grain always I do not change them.
+Bloke on the Range oh many different trips of trial and error I found what my rifles liked. the most consistent with out sacrificing cost and efficiency
With this in mind is a 3 round group decent to use when sighting in a new rifle?
If it groups tight enough then in most cases yes. If not, you might end up chasing shots (and I grew up in a competitione environment where you only get 2 sighters...)
i love your videos but theres also a correlation between how fast you pull the trigger and how close you get to ten. i can hear the faster trigger pulls and they mostly arrive farther off of ten than any
check if a coin is fair by flipping it three times.
are you in the UK? interesting videos. I subscribed.