Here’s how to use (or not use) Standard Deviation calculations in ammo decisions, what they are, and aren’t… Keep reading…
A standard deviation plotted out is a bell curve. Chances are outstanding that a range session calculation will plot into what they call a “normal curve.” Like any normal bell curve, it can get divided into three segments and given values, and, technically, these are the “standard deviations.” It’s “a” standard deviation rather than “the” standard deviation.
Assuming a normal curve, the values are that about 68 percent of forecasted results will lie within one standard deviation of the mean, about 95 percent lie within two, and over 99 percent lie within three standard deviations. If we have an SD calculated to be 12, that means that applying one standard deviation means that about 68 percent of all “next shots” will be +/- 12 feet per second. Since, though, the curve is in threes, in effect if not in fact, that means that a scant number of the shots pose a chance for +/- 24 and some teeny chance remains for shots to go to +/- 36. That, however, is extrapolating or predicting with data and that’s not really wise and doubtlessly uncalled for. Data collection is a record of numbers and I do know that there’s 100 percent chance that the highest and lowest velocities collected for an SD calculation did, in fact, happen. That’s what matters. No matter what the shot results calculated into for an SD, those were the two that represent the highest and lowest prints on the target.
It’s mathematically not possible for an SD to be higher than the greatest single raw deviant, but I do for a fact know that an SD can easily be far lower than the worst shot. Given how it’s calculated, along with how many samples contributed, it’s plain that the nearer the majority are to themselves the less impact a bad one or more has.
I said in a very open-ended way last article that a tolerable SD is 12. Anything more than that is not good; anything less than that probably won’t perform noticeably any, if at all, better than a 12. However! It is at this number, so I say, where the often-uttered tune of “…SD doesn’t matter…” and its refrain of “…seen good accuracy with high SDs…” starts and stops. Twelve. That’s it. Now we have an SD that “doesn’t matter.” The reason this is stuck out here is that everyone has heard this chorus but hopefully figured that it couldn’t be taken universally at literal value. Well it can’t. So now you know! It’s 12. 12 should not be responsible for a points loss, even accounting for or including coincidence of any one shot hitting the edge limit of usual group size.
(Yes, 13 or 14 or 16 or even 20, which is often given as a “limit,” might well be a realistic ceiling but I drew a line to have one. Since there’s a line, now we can cross it and commence argument. I won’t use any load in competition that wouldn’t calculate to a single-digit SD. My 600-yard .223 Rem. load tested to an SD of 3.18 with a Range of 8 fps.)
So after all this has been said, I don’t give SD as much weight in my load decisions as some do. The reason for my focus on it here, as said in the first article, is because that’s the usual “standard” measure of consistency. I look at the speeds as they come up on the chronograph display and write them down. I weigh range and extreme spread more heavily, and I want to see really small variations over the number of test rounds I fire. It’s a matter of waning patience and waxing time. If I see a variance that could cost a point, that load is abandoned.
If you don’t have a chronograph or don’t want to burden a testing session with using one, watch for a correlation between the elevation dispersion and the wind dispersion of test groups. At 600 yards I always test from position (prone, “suited up”). No chronograph (muzzle-mounted chronographs now make this a non-issue). I’ll already have speed-checked the load I’m now down on the mat with. When I shoot my groups, I honestly don’t pay much attention at all to anything but measuring how level I got my perforations. Attempting shot-to-shot wind corrections when testing for ammunition accuracy throws another variable into it that might be most misleading. If I come up with a group that’s a foot wide but only three inches tall, I’m happy.
5 TIPS FOR LOWER SDs
Aside from finding the perfect and magical load combination, ha, there are a few things that do seem to help tighten shot-to-shot velocity deviations. They’ve all be talked all the way through and back again in this space in other articles, but, considered ultimately that this is the overall effect they have, here they are again:
One. Primer seating: fully seated onto a flat pocket bottom.
Two. Consistent propellant charge: weigh the charges if metering isn’t dead-on perfect.
Three. Ignition efficiency: consider that inside flash hole deburring routine…
Four. Temperature insensitivity: choose propellants that exhibit stability under extremes.
Five. Balance: strive to find a propellant that fills the case, but “loosely” (no compressed charges); even more, avoid an overage of air space. These both allow too much variance in ignition pattern.
I learn things all the time. A most knowledgeable and helpful reader pointed out a detail in SD calculation that is better adapted to calculations for ballistics, and it helps because of the usually relatively small size sample involved. We’re not going to chronograph 100s of rounds, usually 10-20. So, instead of dividing the average square of the deviations by the number of samples, but the average square of the values, less one (n-1). That helps any distortion of results toward a number that calculates too small. Keep in mind always that SD is an estimate, in one way of looking at it.