Thread: 3 shot groups are useless!

The title of this thread really does a disservice to the ability to usefully discuss the concept

Originally Posted by Tahr

I’m not so sure the 3 shot group when it is repeatable is so flawed.

I think that's point; when it is repeatable. If a three shot group (size/position) is repeatable, it suggests you've shot enough rounds to find it so (i.e. larger sample size). If a rifle is very precise, (for example can put 10 rounds into 0.5moa), a 3 shot group is going to a be plenty good enough representation of the larger sample (possibly like the example shown). I think it's more of an issue for less consistent systems, where a 3 shot group may be less representative - again it's all 'horses for courses' in my observation.

Originally Posted by earplay

I think that's point; when it is repeatable. If a three shot group (size/position) is repeatable, it suggests you've shot enough rounds to find it so (i.e. larger sample size). If a rifle is very precise, (for example can put 10 rounds into 0.5moa), a 3 shot group is going to a be plenty good enough representation of the larger sample (possibly like the example shown). I think it's more of an issue for less consistent systems, where a 3 shot group may be less representative - again it's all 'horses for courses' in my observation.

"repeatable" is the tricky word. If you take a strict definition - then 3-shot groups are very much not repeatable inherently - there is always a significant relative variation in size and position between 3-shot groups. If you don't overlay multiple groups, this isn't visible - of course overlaying multiple groups is "using more data", which is what I recommend if you want to make accurate statements about the precision and zero of your system, and if you want to improve your performance out further. It's just not a very efficient way to do it.

If you have a rifle where 3-shot groups are always repeatable, in terms of size (and maybe position), i.e. consistently exactly the same - congratulations, your shooter error is likely the larger effect size by an order of magnitude and the noise you introduce is hiding the group variation. It would have to be a tremendously precise rifle, or a tremendously poor shooter, for this to be the case. I have never yet seen any evidence at all of a rifle system shooting consistently sized/positioned 3-shot groups when you measure precisely.

and of course you'd have to shoot quite a lot of 3-shot groups to establish that it is in fact consistent. A lot easier and more efficient to do it all at once

if we go back and read the summary post from the first thread I made on the subject, it really wraps all the conjecture up quite nicely - in the very first thread on the matter

https://www.nzhuntingandshooting.co....8/#post1564544

Originally Posted by gimp

"repeatable" is the tricky word. If you take a strict definition - then 3-shot groups are very much not repeatable inherently - there is always a significant relative variation in size and position between 3-shot groups. If you don't overlay multiple groups, this isn't visible - of course overlaying multiple groups is "using more data", which is what I recommend if you want to make accurate statements about the precision and zero of your system, and if you want to improve your performance out further. It's just not a very efficient way to do it.

If you have a rifle where 3-shot groups are always repeatable, in terms of size (and maybe position), i.e. consistently the same - congratulations, your shooter error is likely the larger effect size by an order of magnitude and the noise you introduce is hiding the group variation. It would have to be a tremendously precise rifle, or a tremendously poor shooter, for this to be the case. I have never yet seen any evidence at all of a rifle system shooting consistently sized/positioned 3-shot groups when you measure precisely.

and of course you'd have to shoot quite a lot of 3-shot groups to establish that it is in fact consistent. A lot easier and more efficient to do it all at once

I guess to some degree, the definition of repeatable is over to the individual user, who is describing it in a manor fit for their purpose. If the centre of each 3 shot group is for arguments sake, within a moa of each other 3 shot group centre, then most users would likely describe it as repeatable. Probably unless looking for it, most wouldn't pick up the difference.

Originally Posted by whanahuia

I guess to some degree, the definition of repeatable is over to the individual user, who is describing it in a manor fit for their purpose. If the centre of each 3 shot group is for arguments sake, within a moa of each other 3 shot group centre, then most users would likely describe it as repeatable. Probably unless looking for it, most wouldn't pick up the difference.

You're dead right, it's all contextual - and if choosing to use a looser definition, then one can only make looser statements

Reading all these posts it stands out that there’s a couple of strong opinions about establishing outright accuracy which is fine, but as the saying goes “there’s more than one way to skin a cat”. If you’re happy with 3-5 shot groups and that’s worked for needs for 40+ years then stick with it. If you’re after bench-rest accuracy then delve deeper by firing more shots per group. Personally, multiple 3 shot groups over a period of time have worked for my hunting rifles although I do on occasion fire up to 20 shot groups out of interest. For NZNRA shooting I do 10-12 round groups when practicing. In a past life it was routine to shoot 20+ round groups in a practice session to check/establish zero and consistency when not in the field. I do wonder how many shooters actually bothered with large statistical groups in years gone by and how many accurately measured their groups before the recent development of phone apps etc that do all the hard work for them. But each to their own. If you’ve got a system that works for you and you regularly hit your target at ranges you’re proficient at then stick with it. With most rifles these days capable of 1 Moa 3 shot groups, which most manufacturers quote, it’s usually the operator at fault when a miss occurs. Just my 2c.

Combining posts #366 and #367

What I think is important is that people objectively look at their situation and form realistic expectations. If for what ever reason, you dont feel the need, or ability, to go to the lengths exhibited here to absolutely prove your system. Then establishing an equally realistic maximum shooting distance, that is likely well inside the ultimate capabilities of your set up, is important.

Originally Posted by whanahuia

Combining posts #366 and #367

What I think is important is that people objectively look at their situation and form realistic expectations. If for what ever reason, you dont feel the need, or ability, to go to the lengths exhibited here to absolutely prove your system. Then establishing an equally realistic maximum shooting distance, that is likely well inside the ultimate capabilities of your set up, is important.

I thoroughly agree regarding expectations.

Ofcourse everybody is entitled to their beliefs and opinions regarding precision and accuracy but there are a lot of manufacturers, gunbuilders and sellers of secondhand rifles out there who make outright fraudulent claims (both actual and implied) about precision to prey on the uneducated and less knowledgeable folks.

And that's a recipe for unfulfilled expectations and disappointed buyers. We see it on here everyday (plus I get a lot of PMs about non public cases) and it's bloody wrong.

The wheels of the bus go round and round
round and round
The wheels on the buss go round and round
All day long !.

Originally Posted by Tankd

The wheels of the bus go round and round
round and round
The wheels on the buss go round and round
All day long !.

Thats how long it takes for some of us catch up, no more (being totally reliant on) three shot groups , no more "nodes" and now a proud 223 owner.

Originally Posted by Fisherman

Thats how long it takes for some of us catch up, no more (being totally reliant on) three shot groups , no more "nodes" and now a proud 223 owner.

I think becoming a 223 owner is the 'gateway' into this rabbit hole. Can finally afford to shoot bigger groups

I got back to the range last night to put another couple rounds over the chrony of my 75ELDM load for the ballistic calc. Out of interest, I shot a couple of three round groups at 100, and a five round at 200. It was interesting to see how both three round groups were approx 1/2 moa, but when combined, are closer to the 0.8 or so which is typical of this load. 5 rounds at 200 fell right into a similar group, so at least it's consistent. I would say the three-round groups were perfectly adequate on their own for my purposes (deer to 350m), so far from 'useless', but the change in POI between groups is interesting.

Originally Posted by earplay

I got back to the range last night to put another couple rounds over the chrony of my 75ELDM load for the ballistic calc. Out of interest, I shot a couple of three round groups at 100, and a five round at 200. It was interesting to see how both three round groups were approx 1/2 moa, but when combined, are closer to the 0.8 or so which is typical of this load. 5 rounds at 200 fell right into a similar group, so at least it's consistent. I would say the three-round groups were perfectly adequate on their own for my purposes (deer to 350m), so far from 'useless', but the change in POI between groups is interesting.

Attachment 288121
Attachment 288122
Attachment 288123

Yeah based on your 10rd mean radius posted earlier, your system "max dispersion/cone of fire/choose your term" is about 1.5MOA (pretty good ! ).

Based on ratios I've observed from my empirical results, you should expect your "average" 3-shot group, if you shoot enough to get an average, to be around 0.5MOA. You'd probably have to fire about 30 rounds to establish a half decent group size average

You should expect 3-shot groups to vary anywhere in size between 0.15MOA and 0.9 MOA. Generally they'll be more clustered around the middle with the outliers (very small or very large, relatively) less frequent.

You have what is, on the spectrum, a pretty precise system, so the absolute size of the variance is smaller.



Of course these are all subject to some pretty healthy error bars because we don't have enough data to be sure. But the 10 rounds in your last post gave us enough info really quickly to say "yeah sweet it shoots well enough, get a good zero & go shoot stuff efficiently"



You could refer to it as a "half MOA" rifle, in terms of the average 3-shot group size, but you'd have about a 20-30% hit probability on a half-MOA target at best. It's a good 1.5 MOA rifle system - you should expect close to 100% hit probability on a 1.5MOA target at 100m, if you account for wind etc.