How many shots makes a group?

[Miseryguts]

Hi, on reading various posts on this forum, I see a lot of talk on 3 shot and 5 shot group data. Bear with a thick novice, but surely 3 shot groups do not give a fair indication of how good/bad a group is??

I have shot some amazing three shot groups, but by the time I get to 10 shots, it has gone to hades in a handcart! below some figures I have to hand shot with a 22 at 50m. ( I have never kept a 3 shot group or notes on same)

Target 1                                                          Target 2

6 shot ES 0.5moa; MR 0.29moa                 6 shot: not noted

8shot ES 0.99moa; MR 0.39moa                8 Shot: ES 0.99moa; MR 0.27moa

10 shot ES 1.48moa; MR 0.48moa            10 shot: ES 1.74moa; MR 0.39moa

 Target 3                                                          Target 4

6 shot ES 0.45moa; MR 0.15moa               6 shot ES 0.53moa; MR 0.15moa

8 shot ES 1.01moa; MR 0.33moa               8shot ES 0.86moa; MR 0.22moa

10 shot ES 1.36moa: MR 0.42moa            10 shot ES 1.52moa; MR 0.34moa

The above was just to see how bad I am at 10 shot groups, perhaps I should stick to 3?

So, what is the consensus 3? 5? 10? 20? 

M diving for the parapet in Monmouthshire

[Popsbengo]

Nice one M !   You've opened a really interesting box of ferrets I think.

I look forward to the "debate"  😉

[meles meles]

Statistically, three shots are really just a loose association, and five just about constitutes a gang.

[One on top of two]

The question has been asked many times , is a three-shot group  worthless , and is the five-shot group the only true measure of a rifle's accuracy?

This is a complex question ...

Back in the days of Harry Pope (which is to say the late 19th century and the early 20th) the ten-shot group was standard. If you used anything but ten rounds to verify a rifle's accuracy, you had something to hide. But then, I would guess around 1950, then later the  five shots became the standard,

There were sound reasons for this. As far as a big-game rifle is concerned, any number of shots beyond three is academic, because not once in a blue moon are you going to get to fling five rounds at a big-game animal. I can remember doing this only five times; there may have been other occasions, but they're very few.

There are other reasons. First, if your rifle will put three in a tight little cluster, it will almost certainly put five in a slightly larger one, and what difference does it make? Second, if you do a lot of shooting, relying on 5-shot groups can get very expensive very quickly. Also, if you shoot big guns, relying on five instead of three is more painful. If you'd like to make sure your .458 Lott will give you match-grade groups why, have at it. Never mind that it will be used on something that weighs over 1,000 pounds and is only 50 yards away--enjoy yourself.

Manufacturers who fire test targets to send along with their guns use three shots because when you're doing that with thousands of rifles a year, it really gets expensive, and going to five is just not economically feasible.

On the other hand, a five-shot group is a much truer test of a rifle's accuracy. The statistical experts among you may want to correct me on this, but it's probably twice as hard, or a bit less, to shoot a good five-shot group as it is a three, because each time you pull the trigger there is so much that can go wrong.

For some rifles, three-shot groups are of little use. This includes anything that is going to fire more than three rounds in the course of business, and includes prairie dog guns, tactical, benchrest, and any rifle that will be used in NRA-sanctioned competition, where you fire multiple 20-rounds strings.

But for big-game rifles, three is it for the foreseeable future. Doubtless we'll someday have rifles and ammunition that are so accurate all you'll need to do is fire one shot to get your zero and that will be that. I can wait; it will be a much duller world.🤔

[Popsbengo]

Statistically the more the better however I'd be concerned that 'shooter fatigue' may distort the results.

The factors being judged:  load consistency (powder, bullet, brass, primer, assembly); rifle consistency (temperature, fouling, mechanical stability); environmental consistency (wind, temperature, pressure, humidity).  And the big one:  shooter skill.    Obviously some factors are very well controlled through a short string of shots.  The wind isn't one and neither is skill

During development I shoot three shots.  Five for final confirmation when close.   Bloody expensive for big stuff!

 

[ds1]

The best explanation that I have heard is that a three shot group can show you what does not work well in your rifle but it takes a five shot group to show you what does work well in your rifle.

[One on top of two]

1 minute ago, ds1 said:

The best explanation that I have heard is that a three shot group can show you what does not work well in your rifle but it takes a five shot group to show you what does work well in your rifle.

Now that I like 👍 and makes perfect sense too me . 

[Popsbengo]

16 minutes ago, ds1 said:

The best explanation that I have heard is that a three shot group can show you what does not work well in your rifle but it takes a five shot group to show you what does work well in your rifle.

👍.  That's what i do.

 

[lapua]

If the first 3 aren’t in the same hole, sending 2 more wont make the group smaller.

[Moorlander]

Five.

[No i deer]

However many you want it to be...

The more you fire the likelihood is the group will get bigger.

I do 3 to 5 depends on how many I load of each too try.

 

[Catch-22]

I’m in the same camp as those who do 3 then 5.

When conducting a first pass on an OCW, I will load 3x of each charge/bullet combination. Very easy to spot the crap nodes.

After choosing the desired node to zone in on, or to refine a node with seating depth, I will move to 5x shot groups. By this point I’m close to my final load, so want to know what it’s like both accuracy wise and also run a final check on velocity and ES/SD values.

But as has been mentioned, you’ve got to be careful not to let shooter fatigue come into play, even with 3x groups. This can happen easily if trying to test multiple load combinations and you’re short of time. With only 3x in a group, a slight bit of shooter error can easily make one think the group is bad when actually the shooter made it so - giving a false reading. Here a 5x group might provide better representation. 

All comes down to personal preference and confidence in your ability to truly maintain shooter technique to ensure a 3x group isn’t misrepresented.

[Miseryguts]

1 hour ago, Catch-22 said:

I will move to 5x shot groups. By this point I’m close to my final load, so want to know what it’s like both accuracy wise and also run a final check on velocity and ES/SD values.

Hi, I see here everyone is coming from re 3 shot groups and larger calibres and expense, but surely even 5 shot groups do not give sufficient data points to enable a realistic evaluation of velocity ES and Sd? Staticians/mathematicians to the fore?

1 hour ago, Catch-22 said:

But as has been mentioned, you’ve got to be careful not to let shooter fatigue come into play,

Especially  true in my case! Due to my disabilities this is a big problem for me, and is very difficult to factor into the equation

Ho Hum

M watching the rain come down in Monmouthshire

[Catch-22]

I side with your point about the need for decent data to be completely confident in your ES/SD etc. I’m a researcher myself, so totally agree.

However, with so many uncontrollable variables, I think there’s little value (personally) in trying to extrapolate rock solid statistical meaning from highly contentious data anyway. So I don’t bother really. If on the face of it, I see some pattern emerging that shows promise, AND it looks good on target, then I’m happy enough. 

I primarily focus on what the target tells me more than anything. If you’ve developed a load that delivers the goods on target consistently over multiple ranges...then happy days. 

Just an opinion.

[NathanM]

As Catch-22 said, there are so many uncontrolled variables in shooting a group, a statistical analysis comparing two is pretty pointless. At best the data is relatively meaningless, but at worst it is completely misleading. What's wrong with a 4 shot group, happy medium?! Or are even numbers a bad thing haha

[OSOK]

For load development ..... shoot two at a time on the range until your happy and it repeats . If the second shot isn't touching the first then the third shot won't bring them any closer !

OSOK

[colin jalland]

10 shot group at 100 yards and I pulled 2 of them but an ES of 48 fps but a SD of 14 work needed

 

[baldie]

When you shoot enough, you can end up getting a feeling, simply from shooting the gun, that the load is "right"

I've lost count of the times I've said it to shooting buddies..." I could pull that trigger blindfolded, and they would still go in the same hole"

Sounds cobblers, but its happened to me that many times, its untrue.

Personally, I usually shoot 5 rd groups.

3 round groups are just an easy way of making the figures look good, and to the uninitiated, sells a gun etc. If you get a flyer in a 3 rounder, you can easily dismiss that group, whereas it becomes a little more obvious if the other four have gone in one hole.

I usually settle on a load, then shoot 5 x 5 groups, and average out the 25 rounds.

A recently discovered method [satterlee ] for me though, gets you to a usable load quickly. I then like to mess about fine tuning on depth etc.

[No i deer]

I agree you do know that the load is good after a few shots 👍

[Chanonry]

On 11/5/2019 at 7:08 PM, baldie said:

When you shoot enough, you can end up getting a feeling, simply from shooting the gun, that the load is "right"

I've lost count of the times I've said it to shooting buddies..." I could pull that trigger blindfolded, and they would still go in the same hole"

Sounds cobblers, but its happened to me that many times, its untrue.

Personally, I usually shoot 5 rd groups.

3 round groups are just an easy way of making the figures look good, and to the uninitiated, sells a gun etc. If you get a flyer in a 3 rounder, you can easily dismiss that group, whereas it becomes a little more obvious if the other four have gone in one hole.

I usually settle on a load, then shoot 5 x 5 groups, and average out the 25 rounds.

A recently discovered method [satterlee ] for me though, gets you to a usable load quickly. I then like to mess about fine tuning on depth etc.

Know what you mean. Sometimes I am working hard and the group is crud but then the right load just seems to group on its own. Odd but gratifying.

Satterlee hasn't been 100% reliable for me. I got a fantastic load when I rebarreled the AX in 243 in 10 rounds (bit of luck sure) but got a real dog for the semi weight 270. Looked great but then threw occasional 3" vertical at 100 on the range. It feels a bit circular- if I start with components that are likely to give me a good load then it works quickly but if the load may be a bit flaky then it doesn't seem to be robust enough to find the narrow bands.  Maybe thats the point.

That's not a view with a huge amount of evidence to back it up, more of an idle thought...