Here to stir pot on seating depth

[Ledzep]

This data is from 100 rounds of what I would consider "average" factory match ammo. Not the best in the world, not the worst. I broke the 100 shots, of which I have MV, X and Y for each shot as recorded on an acoustic target and Oehler 85 optical chronograph system, into 33x 3 shot groups, 20x 5-shot groups, 10x 10 shot groups, 5x 20 shot groups, and 3x 33 shot groups. The groups were taken in sequential order, no mixing and matching.

Group size averages, then the variation range behind in parenthesis:
100 shot group: 1.309 MOA
33 shot groups: 1.24 MOA (1.20-1.309)
20 shot groups: 1.13 MOA (.96-1.309)
10 shot groups: .79 MOA (.45-1.309)
5 shot groups: .56 MOA (.20-1.309)
3 shot groups: .46 MOA (.20-1.10)

So the exact same ammo produced anywhere from .20 up to the full 1.309 MOA results with 5-shot and 3-shot groups, and .45-1.309 MOA in 10-shot groups. This is the EXACT SAME AMMO on all tests. Not testing changes in seating depth, powder charge, etc.. So how can you say that you're conducting a test with seating depth where you shoot 3 or 5 shot groups-- even if you repeat it several times-- and have confidence that one flavor is "better" than another? When you have test-to-test variation with no variables that could be more than 2x the average... Especially if you're not correlating PoA between groups.

ETA: My point in all of this is that if you do the work up front, and use single large sample tests of variables at the beginning, you can make judgement calls to pick a load and be DONE with it. I find an acceptable load, I load that same load until the barrel dies. 150-200 rounds up front and DONE fucking with things on the reloading bench, on to getting better at shooting.

 

[357Max]

@Ledzep - is that from a machine rest?
If so, Bi pod & bag would add a lot more noise into the results.

So what are your thoughts on the Large jump variation test Berger recommends for VLD’s? 10 round groups enough to get you pointed in the right direction.

I’ve started shooting load dev at 200. It seems to be a little easier to see changes, but those changes I’m sure are still in the noise with 5 shot groups. So what is enough? 10, 15?

Sometimes 5 is all it takes to know that 1 1/2” + group isn’t a combo the gun likes.

 

[Franko]

With the sheer amount of noise in our testing processes (as demonstrated by @Ledzep ) maybe a more productive discussion is along the lines of "what is a reliable indicator of a load (or seating depth) being bad".

 

[timintx]

With the sheer amount of noise in our testing processes (as demonstrated by @Ledzep ) maybe a more productive discussion is along the lines of "what is a reliable indicator of a load (or seating depth) being bad

Large variance on velocities. Not good for long and extreme range . Good groups and level velocities usually go together but not always .

timintx

 

[Ledzep]

@Ledzep - is that from a machine rest?
If so, Bi pod & bag would add a lot more noise into the results.

So what are your thoughts on the Large jump variation test Berger recommends for VLD’s? 10 round groups enough to get you pointed in the right direction.

I’ve started shooting load dev at 200. It seems to be a little easier to see changes, but those changes I’m sure are still in the noise with 5 shot groups. So what is enough? 10, 15?

Sometimes 5 is all it takes to know that 1 1/2” + group isn’t a combo the gun likes.

Accuracy fixture bolted to a pedestal, 1.25" straight barrels. I'm not familiar with what Berger recommends, I can look into it tomorrow maybe. 10-shot groups are good "feelers" to get an idea if something is going to play well or not, but honestly I think even 20 shots is skimpy when you're dealing with 2 good loads or 2 good seating depths to say which is "better" or "best". Obviously if one 10-shot group is 1.5 MOA and the next is 0.4 MOA, you can probably rule out the 1.5 MOA one, but if it's .6 MOA and .4 MOA... Not decisive.

If you want tests that you can have good confidence in quantifying and saying "A is better than B" (Assuming neither of them is complete trash), the floor is 25-35 shots per variable change, and recognize that there's still an "error window" (albeit MUCH smaller than 3,5,10 shot groups) that those results are not entirely absolute. If you can find some way to get X and Y coordinates for each shot so you can put them into Excel and get mean radius, all the better-- in that case, shooting 7x 5-shot groups or even 35 shots worth of one shot dot drills and using the same POA/POI reference is essentially the same thing as a single 35. Getting the same data.

With the sheer amount of noise in our testing processes (as demonstrated by @Ledzep ) maybe a more productive discussion is along the lines of "what is a reliable indicator of a load (or seating depth) being bad".

Best results I've ever seen were in the realm of 0.4-0.5 MOA for 20 shots. Best ES/SD data I've ever seen is in the realm of 30-35fps ES and 5-6fps SD for 20 shots.

Personally, for PRS and "practical" long range shooting, I think anything under ~0.85 MOA for 20 shots is good enough, and I'll feel confident with SD's of about 13fps or better. That 13-15fps range for SD on 20 shots starts getting a little uncomfortable, and anything over 15FPS for SD is probably something that needs to be adjusted (probably need a different powder). 8-12fps SD is pretty easily achievable with decent powders for a given cartridge+bullet.

Small sample tests have the ability to show you a load is "bad", and IMO that's if the first 5-10 shots are over 1 MOA and as you keep pumping rounds into it there's no indication of things trying to cluster up. Mean radius is a great tool here because if you have 17 clustered together and 3 shots that make it 1.25 MOA, that's not the same as a 1.25 MOA donut or shotgun pattern. One of those three situations may be salvageable.

 

[MarkyMark007]

Obviously if one 10-shot group is 1.5 MOA and the next is 0.4 MOA, you can probably rule out the 1.5 MOA one, but if it's .6 MOA and .4 MOA...

probably this is the best way to do it. if you are smart, know statistics and read this thread.
but all other 99% people will continue with their ladder tests and 3 shot groups...

rule out what is just rubbish, and keep what is OK. for powder charge, powder flavor, primer, seating depth... and try to make as good ammo as you can to minimize velocity SD for long range. this is in my opinion the best way.

but i think most of precision is driven by the barrel. and the best barrels get those shooters, who on youtube do 3-shot ladder tests on 100y and tell us which load is the best by the shape of the hole...

 

[Near miss]

I have no problem saying I have issues getting all of the shots perfect. But if I see absolutely no clustering (low mean radius) at all, it is much more telling than otherwise good/acceptable extreme spread.

And it works backwards as well. If I shoot my match load and get no clustering, I know there is something wrong with my shooting or setup.

When creating a load, the lower the shot count is, the tighter they all have to be clumped up and that puts much more strain to the shooter not to mess one shot. And after doing a 3 shot test, I look at the single groups vs aggregate of all the shots and see if there are patterns of me or the ammo doing things.

Usually one shot that looks like an outlier in one group might actually fit really good into the big picture.