Home Forums Hunting & Shooting Ask The Gunwriters What is acceptable standard deviation when using a chronograph both with factory loads and

What is acceptable standard deviation when working up loads for a hunting rifle? I know that lower is better but how high is too high and I don't shoot past 300 yards if that helps.

For 300 yards and in, you're fine simply working up your load looking for groups. SD starts to matter for longer range shooting.

It also matters (a ton) how many shots you are using to establish SD. I see lots of references to SD for three shots which doesn't say much if anything. Most publications use a 10-round velocity average and SD standard and that's not a bad place to start. Practically speaking though, at 300 yards it would take pretty significant velocity spreads to cause meaningful vertical dispersion.

This!

It come down to task and purpose.

For my long range TGT guns (500+ yds) I want a single digit SD. I also chronograph multiple strings at varying temps to ensure I have accurate data. With the new radar chronographs it's no problem getting readings while practicing:



For my woods guns and "normal range hunting rifles (<500yds) I don't lose sleep over it. I've also found it a lot easier to achieve low SDs in longer barrels VS my carbines.

I use an Oehler 35P, have done since they became available.
I have seen wild triple digit SD's shoot wonderful very sub MOA groups and tiny single digit SD's shoot unacceptable groups.
In the end, I reshoot groups often and over time to gauge consistency and reliability rather that just following SD initial readings. The groups are always the qualifier.

This in a nutshell.

This. Although, I would suggest that a calculation of SD using 10 shots is just barely starting to become meaningful, and 30 shots give much more confidence in the SD value. Having said that, I agree that it takes a very large ES to make much difference for BG hunting out to 300 yards.

SD of 3.14159 is a piece a cake

I look more at extreme spread. I don’t usually shoot strings that are large enough to worry about standard deviation. I write it down in my load book along with average and ES just because…..

I rely more on the results on the target unless I’m going to be shooting a load at longer range. Even then, if I have a really tight ES and low SD, if they’re all over the target it doesn’t matter. On the other hand sometimes tight groups are accompanied by large ES and SD especially with low end starting loads with slower powders.

That would be PI not cake

I agree, but what I try to do is find the right bullet powder combination where I get single digit SD and wonderful groups, it usually takes many trips to the range and I can foresee using my Garmin more than my Oehler 35P in the future

What about ES? I would think that matters a good bit as well. How often does a very wide variance show up on paper as a flyer, or does that only appear at distance?

I’m just getting started with bullet clocks so don’t have much experience to go on.

As others have already posted, groups matter more than SD at "normal" ranges.

But the other factor is that most handloaders don't shoot enough rounds to get a valid SD. Three, five or even 10 ain't enough. Ammo companies usually consider 30 rounds the minimum for a meaningful number.

You caught my sarcasm... 😁

My personal goal is a SD in the teens. At close range reasonable velocity spread should have only a slight effect on group dimensions. Disparities are much more evident when one really stretches out to like 1000 yds and vertical spread and wind effects becomes more evident. I'm fine with any SD as long as one includes the number of samples. Not a whole of lot change in a T-table once one gets to about 10 rounds if he wants to derive confidence intervals.

N T
10 1.812
20 1.725
30 1.697
1000 1.646

[quote=Pappy348]What about ES? I would think that matters a good bit as well. How often does a very wide variance show up on paper as a flyer, or does that only appear at distance?

That's what I was wondering. Over the years I've gotten to just worrying about ES, and I've been chron'ing stuff since 1997. Although I never shoot beyond 300 yards. With the one exception of a pronghorn at a good 325 yards with a .25-06 years ago.

SD and ES are two different metrics to quantify the same thing, which is the shape of the underlying probably distribution describing the quantity of interest, in this case, shot speed. The more data points (shots) you plot, the more the shape of the sample distribution will match the underlying population probability distribution (if you plotted and then normalized infinite data points).

ES describes the width of the entire sample distribution, and SD describes (in the case of a Normal distribution) the window into which the most common 33% of samples fall. Both describe the likelihood of speed deviation between shots, but ES is usually easier to visualize for most people, and neither will be very predictive of future shots if they are calculated using a small sample size.

More shots per group will tell you the most information you need to know. But that means you have to shoot pretty good too. I always let the group dispersion tell me what I need to know. If something is producing a flier, in a 5 shot group, that is likely just a statistical measure of that group, had you shot a 10 or 20 shot group. A nice round 10 shot group is what you want to look for. Keep them all within an inch, even at 100 yards, and more than likely that load is going to do very well out to 600 and beyond. You don't even need a chronograph, to know a good load, if you shoot. Guys always looking to get out of shooting, and trying to find an easy way are the ones that are losing the battle.

To tell the truth the only time I chronograph my hunting loads is with a new lot of powder. ES/SD arent a consideration, what happens at the 2 and 3 mark target does.

ES is more properly called Range.

Range and standard deviation can be converted back and forth. As Jordan Smith correctly pointed out they are both measures of the same thing, dispersion.

Measures of dispersion do not like to be cornered and made to tell the truth. 3 shots is a very poor indicator of future performance and 5 is only a little better. 30 is minimal, and 50 is starting to be decent. In practice this means that you will probably never have a very precise estimate of true long term variation.

Standard deviations do not add linearly. They add in a special way that virtually ignores the smaller sources of variation. In practice, this means that if you are shooting 8" groups offhand at 100 yards with a 4 MOA rifle, and then switch to a 1/2 MOA rifle, your groups will shrink so little that you probably won't notice.

SAAMI uses the rule of thumb that MV SD is 2% of average MV. This is used to calculate the safety margin for pressure calculations. So their assumption is that a 3000 FPS load will, in the worst case, have an SD of 60 FPS.

In practice, I commonly see commercial rifle ammo with 30 FPS SDs. It's easy to get handloads down to 20 FPS, and mildly challenging to get them down to 10 FPS. In any case, the 30 FPS ammo will practically always shoot as well as the 10 FPS ammo unless you are shooting beyond 500 yards, from a bench rest, with a 1/2 MOA rifle.