Typical Pistol velocity St. Dev. & Extreme Spread

Well, you have a possible 7" of slop. The thing the ballistics programs do is assume a perfectly rigid barrel. With the actually gun, the faster loads will recoil slightly harder, but the bullet will also leave faster. With a revolver or a straight blow-back-operated pistol, I find this influence tends to exaggerate drop (hotter load impacts slightly higher, even at short ranges). With a pistol that has a barrel that locks up, the muzzle rise doesn't usually amount to much until after the bullet is gone, so it behaves a little more like the rigid barrel.

I used to shoot paper plates at 100 with the 1911. The trick was just to pick out a stone or other ground feature about a foot above the plate and then aim at that feature. Elmer Keith did it differently. He put horizontal lines on his front sights to mark how high the top edge of the rear sight blade should be at different ranges, and by keeping that blade level with the appropriate line, could hold the front sight directly on the target.

This series on stats at the Precision Rifle Blog maybe something some find useful.
 
With the advent of automatically generated numbers from chrono's, lots of people became more interested in sd and es. IMO, you do not get valid es/sd from six shots out of a revolver and the numbers with so few data points is statistically irrelevant. Am more interested in av velocity, or a wide variation of velocity from an individual round.

As an experiment once loaded some 257 wby cartridges with very tight necks and heavily crimped. The velocity variations were within 3 fps for 5 shots. The accuracy was extremely poor.

And while sd/es can be important considerations in some shooting, am still picking the combination with best accuracy.
 
zeke said:
With the advent of automatically generated numbers from chrono's, lots of people became more interested in sd and es. IMO, you do not get valid es/sd from six shots out of a revolver and the numbers with so few data points is statistically irrelevant. Am more interested in av velocity, or a wide variation of velocity from an individual round.

As an experiment once loaded some 257 wby cartridges with very tight necks and heavily crimped. The velocity variations were within 3 fps for 5 shots. The accuracy was extremely poor.

And while sd/es can be important considerations in some shooting, am still picking the combination with best accuracy.
Click to expand...
While I agree, sd is mighty suspect with only 6rnds, es should be reasonably reliable.

For accuracy I don't care what the charge weight is, as long as it has a reasonable avg and low sd/es. I get my accuracy from tweaking the seating depth, not the charge weight.
 
Zeke,

You probably changed the barrel time with the tight crimp. You may also have distorted the bullet with it. Either one can throw accuracy off. That's not the fault of the ES.

Small samples, assuming the vertical and horizontal distributions are normal, have predictable confidence levels. For 95% confidence, you can expect the means point of impact to vary. The SD of the variation in the average is a statistic called the standard error. You get it by dividing the population standard deviation by the square root of the sample size. In most instances, we don't know the population standard deviation, so we use the SD, which is an estimate of the population standard deviation.

Error to 95% confidence in group diameter:

Code: 
Shots      Upper   Lower  Standard Error
 2	+346.63% -77.61% SD × 0.71
 3	+146.49% -59.43% SD × 0.58
 4	+77.22%	 -43.57% SD × 0.50
 5	+53.31%	 -34.77% SD × 0.45
 6	+41.44%	 -29.30% SD × 0.41
 7	+34.21%	 -25.49% SD × 0.38
 8	+29.36%	 -22.70% SD × 0.35
 9	+25.89%	 -20.57% SD × 0.33
10	+23.23%	 -18.85% SD × 0.32
15	+15.96%	 -13.76% SD × 0.26
20	+12.53%	 -11.13% SD × 0.22
30	 +9.12%	  -8.36% SD × 0.18

In other words, repeatability and the accuracy with which you can find the center of the mean increases with shots per group.

You can also combine smaller groups. The SD's add as the square root of the sum of their squares. SE is just that number times the square root of the total shot count for all the targets. ES you can find by overlapping the targets.
 
SD or ES without MV has little meaning. SD of 50fps for 1000fps and 3000fps are two rather different concepts. It is ratio SD/MV that matters. I'd like to keep this below 1% for my loads.

It is easier to achieve in rifle than in pistol loads. 0.1gr uncertainty produces more error in a 10gr load than a 50gr load. It is unreasonable to expect a rifle accuracy from a pistol. Even at short pistol target range, a change in MV produces POI shift more than ballistics calculator predicts. Different MV leads to different muzzle rise.

Good group may happen with a load with higher SD. It is because the MV hits the sweet spot. The group will improve further if the SD is improved.

-TL

Sent from my SM-N960U using Tapatalk
 
tangolima said:
SD or ES without MV has little meaning...I'd like to keep this below 1% for my loads.
Click to expand...

Well, keep in mind there are SDs for pressure and group size variation, too. But you are correct that SD and ES have to be in the context of the mean value. The technical term for the SD as a percentage of the mean is the Coefficient of Variation (CV). All of SAAMI's pressure testing numbers, for example, have their limits based on a CV of 4% for centerfire rifle and rimfire ammunition, 5% for centerfire handgun cartridges, and 7.5% for shotgun shells. I don't know why chronographs don't give the CV as a standard feature, as it seems like an obvious thing to do. But we are left to work it out for ourselves.
 
Unclenick said:
Zeke,

You probably changed the barrel time with the tight crimp. You may also have distorted the bullet with it. Either one can throw accuracy off. That's not the fault of the ES.
Click to expand...

You betchca the bullet was distorted, but the purpose was to get as consistent velocity's as possible, not work up accurate load. For my own satisfaction. Not blaming the ES, but demonstrating consistent velocity's are not automatically associated with smaller groups. Some prefer to refer to their statistics, instead of how accurate the load is. For my purposes, consistent velocity's mean nothing unless precision accompanies the numbers.

Am not a statistician by any means, but federal guidelines stated not to use LP III unless 20 relevant data points were available. Granted those were flow analyzes, from older times.
 
zeke said:
You betchca the bullet was distorted, but the purpose was to get as consistent velocity's as possible, not work up accurate load. For my own satisfaction. Not blaming the ES, but demonstrating consistent velocity's are not automatically associated with smaller groups. Some prefer to refer to their statistics, instead of how accurate the load is. For my purposes, consistent velocity's mean nothing unless precision accompanies the numbers.

Am not a statistician by any means, but federal guidelines stated not to use LP III unless 20 relevant data points were available. Granted those were flow analyzes, from older times.
Click to expand...

in handguns its a bit harder, but its on my to do list to try. When working up a load, I shoot for the best SD and ES. What I have found is that that, by tuning the seating depth, usually in 0.003 increments, I can get the accuracy where I want it, rather than trying to do it with charge weight. Its all about timing when the bullet leaves the barrel at a consistent point in the barrels harmonics. I feel tuning seating is a much better way to do that than with the powder charge, but that's just me.

With 9mm I think I may be able to do it. with my revolver cartridges that I crimp into a groove, I am not sure how to go about doing that. I have a few irons in the fire to deal with before I get to that project though.
 
You must log in or register to reply here.
Back
Top