Question about SD

[personal_aviator]

So here I wonder. How do you calculate standard deviation and what exactly is it. I mean i think it is the average difference from the average am I right? Is it important and why do we need (or do we need) to know it? Or should I just focus on ES and average velocity?

 

[ATOM]

As I recall, SD is indeed a measure of how much variation there is in your sample from the average.

Let's look at one group: 1, 1.5, 2, 2.5, 3. The average would be 2.

A second group would be 1.8, 1.9, 2, 2.1, 2.2. The average for this group is also 2, but you can see the difference.

If you were a commercial firm, it might be of some importance. Not a mathematician, so somebody else will no doubt challenge me, but the proof of the pudding is a tight group, not a calculated value on paper or screen. If you're achieving that one-ragged-hole standard, SD and variance and all the other little furry critters so beloved of statisticians are IMO irrelevant.

If it really worries you, you could use the calculator here: http://easycalculation.com/statistics/standard-deviation.php

 

[personal_aviator]

Thanks for the quick reply. I too am a believer that if your shooting one hole groups then who cares what the numbers are except for the big ones like velocity and maybe ES. Get your velocity for your dope (and of course confirm your dope).

I am redoing my load testing because after getting a new scale, my loads accuracy dropped off. My old scale was weighing + or - .2gr. I re weighed some loads that were weighed with my old scale and they vary from 90.8-91.2 (my load was 91.0). My new scale is a gem pro. I think I found my load again today at 91.1gr but need to confirm since I only had 3 rounds at 91.1 BUT the velocity from those three rounds was (in order they were shot) 2806,2815,2815. Two rounds printed one hole and I most defiantly pulled the second shot.

I'm excited about the tight velocity numbers but three rounds is hardly enough for what I would consider a conclusive test. I loaded up 10 more and will see tomorrow what they print.

 

[Suther]

You have the right idea.

Standard deviation is the average of the deviation from mean. (Which is basically what you said)

The easiest way to calculate is with Microsoft excel.

You can do it by hand but its tedious. If you are interested how, but can't figure it out from the wikipedia article on it, I can explain it later (on phone now. Not easy to even type this lol)

 

[personal_aviator]

Thanks Suther, I'm using the calculator that Atom linked. I don't think it's that important a figure to quantify but I was just curious what others thought on the matter. I suppose over time you could see if your SD is opening up there by telling you that something is slowly changing but a drop in accuracy would probably coincide with that as well and a drop in accuracy you'd (hopefully) notice.

 

[back40sniper]

Unless you are shooting at "very" long ranges don't worry about it. At long ranges the difference of 20fps versus 4fps can be a lot of vertical. So, the smaller the SD the better your group will be, at long range.

 

[personal_aviator]

Unless you are shooting at "very" long ranges don't worry about it. At long ranges the difference of 20fps versus 4fps can be a lot of vertical. So, the smaller the SD the better your group will be, at long range.

This is what im after depending on your definition of "very" long range. We shoot out to 1100 yds but hope to push it further when we find a spot to do so. This is why I'm so particular with my loading.

I shot some loads the other day and ended up with a larger spread of fps than the 3 I shot last week. BUT my brass is pretty much done now (has 16 firings on it) so I am blaming that on the age of the brass. I have more once fired brass ready to go but it has yet to be fire formed to my chamber so I wont know the true potential of that load until I try it with that new brass and it has been fire formed once. In case anyone wonders it's .338LM