Define a 1MOA rifle

[nun_hunter]

I’d use a 95% confidence interval.
Using Conh definition I don’t think any rifle could be called 1 moa as you could miss with the next shot.

That’s like saying a Ferrari can’t go 130mph because it could break down at any second? You can only go off of previous data. Obviously once the barrel starts getting burned out it will no longer be a moa rifle. My rifles are all moa rifles, that’s the accuracy I expect, but for them to do under moa all the time then the average 3 shot group will be 0.5moa less. If you are shooting on average 3 shot groups that are moa, the Minute you shoot a 20 round group you’ll be up over 1.5moa
[/QUOTE]
But then how is shooting a 20 round group any different than shooting 7 3-round groups (20 doesn't go exactly into 3 but you get what I mean) assuming you're allowing the barrel to not overheat and keeping all other variables the same?

 

[WalkedUp]

That’s fine - I’m happy to go with that definition.

But to be confident it will NEVER shoot outside 1”, you do need a substantially larger sample size than most use.

And I think there are many (possibly most) who would be happy using a looser definition.

I’d use a 95% confidence interval.
Using Conh definition I don’t think any rifle could be called 1 moa as you could miss with the next shot.
[/QUOTE]

Correct. That’s the issue. It can only be applied retrospectively to a scrapped gun.

 

[Mungo]

I’d use a 95% confidence interval.

That seems like an obvious way to go.

Trouble is that the 95% CI is sensitive to sample size. So you need to decide on what an appropriate sample size is.

Even with an n of 100, you get groups >1” often enough that a lot of target shooters would not be happy calling it a 1moa gun.

So actually, it increasingly looks like different applications lead to different definitions. Target shooters want the greatest possible certainty in the accuracy assigned to a rifle. They will demand a different standard to stalkers.

 

[Mungo]

Correct. That’s the issue. It can only be applied retrospectively to a scrapped gun.

That’s one of the most sensible responses so far!

I really like this

 

[jthyttin]

But then how is shooting a 20 round group any different than shooting 7 3-round groups (20 doesn't go exactly into 3 but you get what I mean) assuming you're allowing the barrel to not overheat and keeping all other variables the same?

Group size is only relevant in disclipines that measure group size (benchrest).

In other uses it's only relevant how far you hit from your supposed POI (that is different from POA in many cases).

When you fire 7pcs 3rd groups and measure, the POI is not the same for all groups. Thus it's lot easier to shoot 7pcs 1MOA 3rd groups, even in a row, than shoot a single 1MOA 20rd group.

Useful way to think it through, is that you shoot single 20rd (or preferably 30rd) group in single session, allowing for rifle to cool and spotting each shot. Afterwards you can choose any 3 shots in that group, and it would be a valid 3 shot group. I find very few people claim their rifle keeps 20 or 30 shot group under 1MOA...

 

[nun_hunter]

Group size is only relevant in disclipines that measure group size (benchrest).

In other uses it's only relevant how far you hit from your supposed POI (that is different from POA in many cases).

When you fire 7pcs 3rd groups and measure, the POI is not the same for all groups. Thus it's lot easier to shoot 7pcs 1MOA 3rd groups, even in a row, than shoot a single 1MOA 20rd group.

Useful way to think it through, is that you shoot single 20rd (or preferably 30rd) group in single session, allowing for rifle to cool and spotting each shot. Afterwards you can choose any 3 shots in that group, and it would be a valid 3 shot group. I find very few people claim their rifle keeps 20 or 30 shot group under 1MOA...

I get what you're saying and assume some people would do that in order to convince themselves of how small their group is.

I generally shoot/zero on the same target (fresh paper obviously) so can check each group size and POI Vs POA across all of them. In essence one large group spread over different pieces of paper with the same POI.

 

[Mungo]

Correct. That’s the issue. It can only be applied retrospectively to a scrapped gun.

The more I think about it, the more I think this is the route to the answer.

I think the logic would go something like this:

Decide on the total number of shots the gun will take in its working like. A strictly recreational stalking gun might take anything between a few dozen to a few hundred. A contractor’s gun might take 1-2 thousand, maybe less depending on conditions. A target shooter might take 3-4 thousand, depending on cartridge and discipline.

That sets the upper limit of sample size. The actual usable sample size will be smaller - it will consist of only those shots taken under ‘perfect’ test conditions.

The gun will then have a mean lifetime group size, with some standard deviation and some extreme spread. People are arguing that to be classified a 1MOA gun, it must never once shoot a group bigger than 1”. So the lifetime ES must be less than 1”.

So: if we establish the total number of test shots the gun will take in its lifetime, we can then establish what SD will give us an ES that never exceeds 1” for that sample size. This can be done by running simulations - which I will do when I’m back at my computer.

We can then decide what sample size we think is acceptable to verify a gun, say at point of sale, such that it gives us a robust estimate of the lifetime SD. I’m not certain how to do this part, but will find out.

 

[borbal]

But to be confident it will NEVER shoot outside 1”, you do need a substantially larger sample size than most use.

Statistically, you can never say never. And remember, a barrel only has a (very) finite lifetime and are you going to spend that lifetime testing it to see if it will be useful, or actually doing something useful?

 

[JTO]

Personally I would like to think a 1 MOA rifle will put a significant majority of shots (75%?) into 1 MOA

And the other 25% is down to driver error.
Other things to think about.
All shots from a cold barrel to simulate stalking requirement? This could take days/weeks!
All shots from a fouled and warm/hot barrel to simulate target shooting?

 

[Tim.243]

Statistically, you can never say never. And remember, a barrel only has a (very) finite lifetime and are you going to spend that lifetime testing it to see if it will be useful, or actually doing something useful?

Quite but for the "stats" boys a piece of paper is replaced by the next sheet from the pack.
Then who is to say that barrel is not one from a Friday afternoon when people want to go home early!
Had this conversation with several people who shoot rifles.
" People can't match the equipment used these days, it is far better than the person using it"
You can set a rifle up to shoot a group on a bench with 3 in the bull at a 100 yds, then faced with a GM Roe buck with the outfitter telling you the price a 1/2 hr before you pull the shot and miss... Yet you had put 3 shots on target with him the morning before.
The person will make the poor shot not the rifle every time.

 

[Tim.243]

That’s like saying a Ferrari can’t go 130mph because it could break down at any second? You can only go off of previous data. Obviously once the barrel starts getting burned out it will no longer be a moa rifle. My rifles are all moa rifles, that’s the accuracy I expect, but for them to do under moa all the time then the average 3 shot group will be 0.5moa less. If you are shooting on average 3 shot groups that are moa, the Minute you shoot a 20 round group you’ll be up over 1.5moa

What happens when you miss or pull the shot with your "all moa" rifles.

 

[Dalua]

I think you can gain some insights by looking at competitions, one opportunity is F-class. Their results are published - Results – Great Britain F Class Rifle Association

Their targets are of known dimensions. Look at the recent Europeans. The FTR winner shot 250 5's. 1/2 minute bull is the target. Their season has just finished. They clearly aren't content with 1 MOA rifles.

Similarly, .22 shooters and PRS shooters have their handles on what is acceptable accuracy. The recurring theme when I get involved in any of these sports is build quality. The winners are spending lots of money on the best kit and the best ammunition. Their measure is where they come in competitions.

Personally, I won't shoot a deer rifle that doesn't place 3 shots in less than 0.5 MOA at 200yds when fully rested.

Regards

JCS

AFAIK the F-Class 5 is a 1MoA-diameter ring on all targets except the long-range one, where it's 10". The V-bull is half those diamteters.

On reflection, I like the 95% confidence interval idea. I my trade were taught to believe (sort of) things apparently-demonstrated to that standard.

 

[Mungo]

Statistically, you can never say never. And remember, a barrel only has a (very) finite lifetime and are you going to spend that lifetime testing it to see if it will be useful, or actually doing something useful?

I think in the specific instance of a rifle where you track every shot across its usable lifetime, you could retrospectively say it never shot outside 1” (as @WalkedUphasbpointed out).

But you’re entirely right that practically, it’s not a usable definition. So the problem is then what sample size early in its life gives you confidence that it is likely to never shoot outside an inch.

 

[borbal]

So the problem is then what sample size early in its life gives you confidence that it is likely to never shoot outside an inch.

Define what you mean by "confidence" and you are in business!

 

[Mungo]

Define what you mean by "confidence" and you are in business!

Well exactly.

And that will vary from person to person and across contexts.

I suppose the take home from this is that a claim that a rifle is a 1MOA gun is usually no more than aspirational.

 

[Conh]

But then how is shooting a 20 round group any different than shooting 7 3-round groups (20 doesn't go exactly into 3 but you get what I mean) assuming you're allowing the barrel to not overheat and keeping all other variables the same?
[/QUOTE]
It isn’t any different as long as you plot where all of the groups land in relation to each other. If 7 3 shot groups are all 1/4” groups, they could be a 3/4” group when put together if they don’t lay over each other perfectly.

 

[Conh]

What happens when you miss or pull the shot with your "all moa" rifles.

You miss, obviously. I struggle to see your point.

 

[Conh]

The person will make the poor shot not the rifle every time.
[/QUOTE]
Again, not true. I take 50+ different people stalking per year. The amount of people out there struggling from poor kit is staggering. People who can’t hit a barn door with their kit, but then when they have a go with my rifles they magically start shooting really well.

 

[Tim.243]

You miss, obviously. I struggle to see your point.

I had asked nun hunter unless you are now his spokes person lol

 

[jthyttin]

So: if we establish the total number of test shots the gun will take in its lifetime, we can then establish what SD will give us an ES that never exceeds 1” for that sample size.

I think you will struggle to find statistical distribution that matches real world. In real world distributions tend to have "fatter tails" than some popular statistical distributions, many times.

Of course easiest way to have 1MOA rifle is to only fire one shot and scrap it. And the most honest way to evaluate accuracy/precision (people use these ambiguously) is to plot the shot according the desired POI, not to shoot groups.