Some barrels get better standard deviation than others?

[withoutremorse42]

In testing the 4.75", 5", 6" and 6.5" barrels, I've noticed one in particular tends to get better standard deviation numbers (factory ammo) than the others. Is there a possible explanation for this from the barrel side, or is it just luck of the numbers for how consistently a given ammo will perform from varying barrels?

[bangbangping]

How big is your data set?

[withoutremorse42]

How big is your data set?

not huge, usually 20 rounds

[golfindia]

Calculating standard deviation for small sample sizes really doesn't provide much useful information. 20 is a small sample size. An N of 100+ will start getting you toward something meaningful. Envision a bell curve. A small N won't ever give you a smooth, well defined curve to let you know substantially and how many of your data points are really out from your median value. It just looks like a shotgun blast. Large N will let you see this well.

In handloading, velocity extreme spread is far more informative of how things are going.

[bangbangping]

What golfindia said. I think most of us have a tendency to read too much into small samples, whether SD, ES, or group size. (for example, a .25" three shot group at 50 yards does NOT mean you can consistently hit a 1" target at 200)

[dellet]

When all the barrels have at least 100 rounds through them, check them all again with the same lot of factory ammo as the original test.

New barrels, unless you have checked the bore, will give unreliable feedback until the are broke in. Hand lapped barrels get there quicker, but you will notice velocity and group size changes for as much as couple hundred rounds.

[golfindia]

What golfindia said. I think most of us have a tendency to read too much into small samples, whether SD, ES, or group size. (for example, a .25" three shot group at 50 yards does NOT mean you can consistently hit a 1" target at 200)

But a 2 shot sample size is sufficient if your ES between those two shots is huge.

Shooting more samples of that same load will never reduce the ES.

For the amount of time I have available to load and shoot, when I'm trying to get consistency, I usually look for 2 identical velocity readings out of a 5 rd ladder sample string. Usually doesnt take too many tries. I frequently can even get 3 dupes without spending too much time. Hardcore purists I'm sure would call this a bush league effort...

I've put SD calculation formulas into spreadsheet cells for some gun/load combos that I shoot a lot of. But it takes too long to get a sample size big enough to match up with what I've already determined by watching the ES along the way. Not worth the effort to me.

[bangbangping]

What golfindia said. I think most of us have a tendency to read too much into small samples, whether SD, ES, or group size. (for example, a .25" three shot group at 50 yards does NOT mean you can consistently hit a 1" target at 200)

But a 2 shot sample size is sufficient if your ES between those two shots is huge.

Shooting more samples of that same load will never reduce the ES.

Very true. And I rule out a load with a sample size of 1 if the bullet hits the paper sideways.

[withoutremorse42]

Thank you all, good info here