? on ES / SD

[piutemike]

In working up loads today for my 300RUM the best one I found with the 180 E-Tip was 96.2 gr. retumbo. I shot 2 groups with that load .632 and .685. Both groups were 3 shot groups. The velocity of those 6 shots was between 3284 and 3270, a ES of 14.

My question is this, how do I figure the SD? Is there a simple way to do this? My chrono does not have this feature.

I was quite pleased with these groups since it is a factory gun, skim bedded, trigger tuned, muzzle break and wyatts box added by Shawn and I don't consider myself a good shooter.

Mike

 

[boomtube]

Statistically, computing Standard Deviation on a sampling of only three data points is meaningless.

 

[Buffalobob]

If you keep a record of your velocities once you get about 10 of them, you can enter them into Excell. Highlight them and then click on the little arrow to the right of the Sum symbol (sigma) and when it asks you if you want more functions, then click that and at the bottom is STDEV. Click that and you are done.

Std Dev as mentioned is not much use for a limited number of rounds.

 

[piutemike]

Thank you very much for your answers. I'll shoot some more tomorrow and use Excel as recommended.

Mike

 

[goodgrouper]

On a 3 shot group, the standard deviation will always be half of the extreme spread.

And I disagree with the other posters. If used in a load development ladder of .5 grain increments, a three shot group method will tell you what to re-test by showing you the nodes. If the SD stays good again on a different day, you have a winner for that temperature range 95% of the time.

In other words, a 5 shot group on one day is less meaningful than two 3 shot groups on two different days.

 

[Mikecr]

Why would you want SD anyway?
You have ES, and that's all that matters.

If I put a chrono on the market today, it would not display SD or any other monkey math. Just the raw, brutal truth..

 

[piutemike]

Just wanted to know how to figure SD. I did not know if there was a simple way of doing that.

 

[davewilson]

so just what is the standard deviation? isn't it simply a number that is exactly half of the ES? i understand the average velocity can be something that's not quite in the middle but i've always thought SD to be half of the ES and i'm like someone else that said it's kind of a "BFD" type of thing.

 

[P KUNDA]

Not always-but acceptable

On a 3 shot group, the standard deviation will always be half of the extreme spread.,,,,,,,,,,,,,,,,

That's Wrong, unless the "middle shot" velocity is exactly at the middle of the ES!

Peter

 

[Forester]

Why would you want SD anyway?
You have ES, and that's all that matters.

If I put a chrono on the market today, it would not display SD or any other monkey math. Just the raw, brutal truth..

+1 on that.

Low ES is the tougher standard.

 

[goodgrouper]

On a 3 shot group, the standard deviation will always be half of the extreme spread.,,,,,,,,,,,,,,,,

That's Wrong, unless the "middle shot" velocity is exactly at the middle of the ES!

Peter

No it is not really wrong. Since most chronographs round up or round down the speed to the closest number, for our purposes, the es will be basically double the sd on a 3 shot group plus or minus one foot per second which is better than the accuracy level of any chronograph can measure. Remember, this thread was started by a fellow who was asking how he could figure SD since (I assume) his chronograph wouldn't measure it on a 3 shot group like the CED M2. By taking the ES and dividing it by two, it will usually be the SD within 1 up or down (sorry about that gross non-absolute all you math nuts).

I probably should have elaborated on my original post that it is not "always exactly" half but it is always going to be basically half for the poster's needs. I should have realized that the math junkies were going to break it down to the letter of the law and would not allow a generalization!

I am sitting here looking at 25 pages of load data from the last several months with about 10 load ladders of 3 shot groups per page. In all that info, I only found two loads that the ES divided by two was more than one foot per second off the calculated SD. They were both 4 fps off. So the suggestion that you can take the ES and divide it by two to get the SD is pretty sound if you don't like running the longhand math like me! And remember, this is only for a 3 shot group. If you shoot more than 3 shots, the SD will begin to show it's gravity or importance even more.

 

[goodgrouper]

so just what is the standard deviation? isn't it simply a number that is exactly half of the ES? i understand the average velocity can be something that's not quite in the middle but i've always thought SD to be half of the ES and i'm like someone else that said it's kind of a "BFD" type of thing.

For all intensive purposes relating to figuring a load's capability, the sd will "basically" be half the extreme spread on a three shot group unless your chrono reads in half feet per second and not many do.

And if someone says that ES is everything and SD means nothing, they don't fully comprehend the fact that the two are intertwined to the point that either both are meaningful or both are meaningless.


My only gripe with holding ES as the holy grail of data is this:

Say you fire ten shots and they read
3000, 3001,3000,3003,3002,3000,3001,3005,3002, and then the last shot had a little tighter neck tension and it went 3050 fps. The ES method would show this load to have a spread of 50 fps and unless you went back and looked at each individual shot speed, you would think this load stunk. But the SD would still show this load kicked some serious butt. That is why both ES and Sd are helpful to have and therefore why both are included in the circuitry of modern chronographs.

 

[Winchester 69]

So just what is the standard deviation?

1. Average your velocities.
2. For each shot, subtract the average velocity from that particular velocity, and square the difference.
3. Get the sum of the squared figures.
4. Divide the sum by the number of shots. This number is the variance.
5. Get the square root of the variance. This number is the standard deviation.

The standard deviation is a measure of error that is useful when working with statistical inference based on the Gaussian distribution (bell curve). Other measures of error, like the extreme spread, may be more practical figures, depending on the application.

Isn't it simply a number that is exactly half of the ES?

Nope.
.

 

[piutemike]

For all intensive purposes relating to figuring a load's capability, the sd will "basically" be half the extreme spread on a three shot group unless your chrono reads in half feet per second and not many do.

And if someone says that ES is everything and SD means nothing, they don't fully comprehend the fact that the two are intertwined to the point that either both are meaningful or both are meaningless.


My only gripe with holding ES as the holy grail of data is this:

Say you fire ten shots and they read
3000, 3001,3000,3003,3002,3000,3001,3005,3002, and then the last shot had a little tighter neck tension and it went 3050 fps. The ES method would show this load to have a spread of 50 fps and unless you went back and looked at each individual shot speed, you would think this load stunk. But the SD would still show this load kicked some serious butt. That is why both ES and Sd are helpful to have and therefore why both are included in the circuitry of modern chronographs.

GG, Thanks for the explanation and I now understand how you can have a high extreme spread and still have a low SD.