Are 5 shot groups worthless statistically?

[fly108]

SHOTS: ABCDE

PAIRS: AB AC AD AE BC BD BE CD CE DE

It is a statistically significant sample set.

There are 10 data points in the pairs. The average distance between the pairs divided by two represents the average radius of the group.

If you then take the standard deviation of that sample set you can state with a certain confidence the actual deviation of that radius.

The argument the video makes is about the level of accuracy achieved with a five shot group.
He is saying it is very difficult to measure small variations with a small sample set.
Everybody already knows that. It just took him 10 minutes to say it.

i didnt read entire 7 pages, but had to response this

10 data points might be statistically valid for 5 shots, if you think it from maths point of view
but if you try it in real life, you might see that its not

here is follow up video that explains it

this goes to same category as most math that tries to explain real life, while failing miserably
like enthropy on its "gas is even in a box" hypothesis (fyi, its not, ever, it glumps, not even close to even)

i shot 10shots groups, took the results and put lots of pellets in to furthest corner in a closet
few months later i found out that my results were bad, i didnt figure why untill i saw this vid

people who shoot for competition tests hundreds of shots on a batch, as far as i know

 

[drpietrzak]

As the nihilist would say, all is futile.

 

[.20calguy]

Well, Twenty shot groups today!
I wasn't serious in the first place, and I was too lazy to change the target. It took twenty shots to see the group emerge from the previous shooting.
Lol

 

[qball]

Based on statistics some statices are statistically incorrect.

 

[Hateful McNasty]

Well 1 shot or 20 shot group it's a walk to the rack , look at the shots made and say to myself ". Wow thats pretty good " or " wow that kinda suked". In the end and overall bottom line its just another day of shooting fun.

 

[Centercut]

IMHO, shooting groups tests the gun and ammo, shooting targets tests those two AND the shooter... Additionally, shooting targets takes MUCH MORE shooter skill than shooting groups. A few will disagree, but without fail, everyone I know that has shot five 5 shot groups on an EBR practice target, then shot 5 targets five different times will tell you that shooting the individual target five times separately (competition style) is significantly more difficult.

 

[DevilsLuck]

Yeah we all saw this guy's vid... Shot for shot is all that matters in real world.

 

[drpietrzak]

While humans are often mislead by proximity error there is a really strong archival history with sighting in weapons. The process has been in use since pre-history, long before documentation. Oral tradition on how to hit something with a weapon, and do so predictably. So we have of pretty good understanding of what works. Like breathing air vs water we knew before we could document.

Why, maybe not so clear in history. It works so people use it. The rest of are just going to argue about the the why.

 

[OldSpook]

IMHO, shooting groups tests the gun and ammo, shooting targets tests those two AND the shooter... Additionally, shooting targets takes MUCH MORE shooter skill than shooting groups. A few will disagree, but without fail, everyone I know that has shot five 5 shot groups on an EBR practice target, then shot 5 targets five different times will tell you that shooting the individual target five times separately (competition style) is significantly more difficult.

This is absolutely true. They are two different problems (as you said "and the shooter"). When shooting for score you are shooting at a tiny little spot. When you are shooting for groups you are shooting "at" the group. When you compile all your shots on an EBR target into a group it gives you the same information that shooting a group with that many shots in it gives you. You can no longer extract a score from the group.

Clearly the point of shooting groups is to test the combination of shooting platform and ammunition. The point of shooting for score is to test the shooter. The performance of the platform/ammo combination is only a "contributing" factor.

The only thing I take issue with is the assertion that it takes more skill to shoot scores than it does to shoot groups. Shooting groups requires all the same skills that shooting for score requires. Shooting for score is more discriminating. By that I mean shooting scores will reveal poor shooting skills more clearly than groups. In other words small error can hide in a group it can't on a target.

 

[OldSpook]

i didnt read entire 7 pages, but had to response this

10 data points might be statistically valid for 5 shots, if you think it from maths point of view
but if you try it in real life, you might see that its not

here is follow up video that explains it

this goes to same category as most math that tries to explain real life, while failing miserably
like enthropy on its "gas is even in a box" hypothesis (fyi, its not, ever, it glumps, not even close to even)

i shot 10shots groups, took the results and put lots of pellets in to furthest corner in a closet
few months later i found out that my results were bad, i didnt figure why untill i saw this vid

people who shoot for competition tests hundreds of shots on a batch, as far as i know

Yeah, you are right. So is he. If you haven't shot ten thousand shots through your platform and plotted and measured and tested every single one of them you really don't know the exact truth about the rifle to within one millionth of an inch.

Whatever. I think I'll stick with 5 shot groups, because (ONLY GOD KNOWS WHY) they have always worked just fine for me.

The poster of the video is making the point that you can not know what your rifle is doing EXACTLY with one or two five shot groups. All the while he is USING FIVE SHOT GROUPS to prove his point. That's kind of interesting, don't you think? Maybe we are missing the real point he is trying to make? The real point? Well, it takes a lot of shots, to get highly accurate information about your rifle/ammo combination. You might shoot ten shot groups or five shot groups but it's going to take several of them if you want a high degree of accuracy.

Thats the whole point of his video.

 

[jetpopt]

This is absolutely true. They are two different problems (as you said "and the shooter"). When shooting for score you are shooting at a tiny little spot. When you are shooting for groups you are shooting "at" the group. When you compile all your shots on an EBR target into a group it gives you the same information that shooting a group with that many shots in it gives you. You can no longer extract a score from the group.

Clearly the point of shooting groups is to test the combination of shooting platform and ammunition. The point of shooting for score is to test the shooter. The performance of the platform/ammo combination is only a "contributing" factor.

The only thing I take issue with is the assertion that it takes more skill to shoot groups than it does to shoot scores. Shooting groups requires all the same skills that shooting for score requires. Shooting for score is more discriminating. By that I mean shooting scores will reveal poor shooting skills more clearly than groups. In other words small error can hide in a group it can't on a target.

The sentence I emphasized in the quote is my mantra. I like to shoot both one-per and groups and compare them.

I noticed that when shooting a page of 20 individual points, not many were right on target, but most of them were very close to it, and in different locations (high, low, left, right). When I shot groups of five, most of the groups disguised these small variations. Should note that I continued to aim at the original POA even if the first shot did not not hit it, because I wasn’t trying to see if the gun could hit the same hole (I’d already seen that enough times before).

 

[fly108]

Yeah, you are right. So is he. If you haven't shot ten thousand shots through your platform and plotted and measured and tested every single one of them you really don't know the exact truth about the rifle to within one millionth of an inch.

Whatever. I think I'll stick with 5 shot groups, because (ONLY GOD KNOWS WHY) they have always worked just fine for me.

The poster of the video is making the point that you can not know what your rifle is doing EXACTLY with one or two five shot groups. All the while he is USING FIVE SHOT GROUPS to prove his point. That's kind of interesting, don't you think? Maybe we are missing the real point he is trying to make? The real point? Well, it takes a lot of shots, to get highly accurate information about your rifle/ammo combination. You might shoot ten shot groups or five shot groups but it's going to take several of them if you want a high degree of accuracy.

Thats the whole point of his video.

im sorry if you feel upset of my comment, so much so that you went and disliked it

also im fairly sure that you either did not watch or did not understand (or possibly both) that follow up video

it clearly states the number of shots you need to determn if your handload works better than other


i see this forum has lots of people who cant take reality as is

bye

and have a nice day

 

[qball]

wow, this topic is getting hot! LOL!!!

It all depends on what your end goal is, if a 5 shot group is your end goal then it’s more than good enough. For some 5 shot group is just a start and for some no where near good enough. Since I paid for my guns and ammo I am happy with whatever groups I’m happy with, or I’ll return/sell the gun.

 

[drpietrzak]

Much of the shooting I do is in 2 veins. One is longer distance where I am putting 10 on a bull over multiple (2-3) bulls. The other tends to be closer range 1 per bull. The movements from one bull to the next on 1 per bull for 25 bulls, or whatever, amplify the error set up of the rig adds. You are moving things, and things need to resettle. This is why bench rigs are near, if not, full rail guns. The goal to to reduce the error brought on by resetting needed for each bull.

There are trajectory changes for a 1 per bull that is different over the target than you see with groups. While in pb I do not notice the trajectory changes as much for the most part, I do with my springer. I have gone to setting up rifle and target with a laser beyond 10 yards/meters out to about of 25 yards if possible as the target is literally a "." on my targets and the trajectory matters. Shooting slight inclines to slight declines, especially differences between cards, makes enough of a difference. Consistency matters, I think. I now usually try to have the rifle leveled at the center of the target if possible.

I do not have the variations of trajectory between shots on 10-shot per bull targets. These 10-shot bulls are similar to groups in that there is less movement and more "keep every thing the same" needed. The shooting for score, for me is different than groups even then though. Thouugh it depends on what one means by shooting groups.

 

[Hateful McNasty]

Not that there's anything to it

How many shots in a group? | Pyramyd Air Blog

by B.B. Pelletier SavageSam prompted this report, but many of you have indicated an interest. The number of shots that should be in a group is determined by how confident you want to be that the group represents what the gun can do. Some people will argue that if the gun shot a particular group

www.pyramydair.com

 

[OldSpook]

im sorry if you feel upset of my comment, so much so that you went and disliked it

also im fairly sure that you either did not watch or did not understand (or possibly both) that follow up video

it clearly states the number of shots you need to determn if your handload works better than other


i see this forum has lots of people who cant take reality as is

bye

and have a nice day

You may not understand his videos as well as you think you do.

Maybe you will understand my comments, perhaps you won't.
By all means make assumptions about what I did and did not do.

You might want to rethink your opinion of "lots of people".

Happy Independence Day to you too.

 

[.20calguy]

Ok then, here, Ive done it!
100 rounds at 28 yards..
Lol, yeah, that didn't help.

 

[fe7565]

All I know is that I had many nice 5 shot groups at 100 yards, but when I went to 10 shot groups during the same session, the overall groups near doubled in size.

 

[drpietrzak]

All I know is that I had many nice 5 shot groups at 100 yards, but when I went to 10 shot groups during the same session, the overall groups near doubled in size.

The increase in size is commonly 1.5 times, or so, larger for me from 5 to 10 shots. The relative center point for poi for the 2 is pretty close though.

 

[JimD]

I have had some statistic courses but it was years ago and I am not going to try and argue with the author of the subject video. But I do think his conclusion that 5 shot groups are useless statistically is a bit deliberately provacative. I also think that it would be better to say what the confidence level would be statistically for a single 5 shot group. You'd need to further define what conclusion you were trying to support, however. If the one group is 1/4 inch center to center are you trying to say all shots with this setup will result in a similar group size? That is going to pretty much support his title - the confidence would be near zero. But if you were more realistic and wanted to support a conclusion that 80% of the groups would be no more than twice the size of the one 5 shot group I think you could state a confidence. More shots is going to increase the confidence we know the results of the load we are testing.

I also continue to believe that a large single five shot group is pretty significant. If I didn't mess up. That group is not going to get smaller with more shots. If there isn't a cluster with an outlier, one 5 shot group may be enough to say that combination isn't working well. But one really little group certainly doesn't prove the next group will be equally small. It tells me that more testing is warranted if the early groups are small. My shooting is such that I use multiple groups just because I know I am the biggest variable in the group size. So realistically I am looking at multiple groups for ammo that gives me at least some small groups. If all are large, they are not worth more time. I don't really care about the statistics. If you do, that is totally OK.