Standard Deviation vs Extreme Spread

Discussions of powders, bullets and loading information.

Moderators: Kirk, Lucinda

User avatar
desert deuce
Posts: 3065
Joined: Mon Apr 14, 2008 10:51 pm
Location: Rio Rico, Arizona

Standard Deviation vs Extreme Spread

Post by desert deuce » Mon Dec 23, 2019 12:33 pm

Interesting article August 2019 Issue of Handloader Magazine #321 comparing Standard Deviation to Extreme Spread by Rick Jamison titled Velocity Uniformity.

He refers to Extreme Spread as a "Hard Number" and that agrees with my findings, bias confirmation perhaps, in comparison to Standard Deviation.

But, it seems single digit ES is extremely rare in white powder rifles even for five shot groups.

In the 45-70 and 45-90 single digit ES has been obtainable for me sometimes with extended effort, sometimes relatively easy, for five shot groups.

However, when testing for long range shooting in the 45-90 the ten shot ES is a better predictor of performance at 1,000 yards and single digit ES of as low as 5 have been obtained but any single digit ES is good for ten shots.
Sometimes you get the chicken, and sometimes you get the feathers!

mdeland
Posts: 11544
Joined: Mon Nov 20, 2006 1:47 pm

Re: Standard Deviation vs Extreme Spread

Post by mdeland » Mon Dec 23, 2019 3:12 pm

I have always thought Extreme spread the most practical of the two ways to measure consistency. My Oheler 35 records both on tape. Standard Deviation is probably more accurate over a large sampling though.

User avatar
desert deuce
Posts: 3065
Joined: Mon Apr 14, 2008 10:51 pm
Location: Rio Rico, Arizona

Re: Standard Deviation vs Extreme Spread

Post by desert deuce » Mon Dec 23, 2019 7:45 pm

Got an Extreme Spread for five shots this afternoon of 4, 45-90.
Sometimes you get the chicken, and sometimes you get the feathers!

semtav
Posts: 2233
Joined: Wed Feb 03, 2010 3:21 pm
Location: Montana

Re: Standard Deviation vs Extreme Spread

Post by semtav » Mon Dec 23, 2019 8:07 pm

desert deuce wrote:
Mon Dec 23, 2019 7:45 pm
Got an Extreme Spread for five shots this afternoon of 4, 45-90.
5 more and you could have entered them in the match :roll: :lol:

rgchristensen
Posts: 125
Joined: Sat Jan 16, 2016 10:51 pm

Re: Standard Deviation vs Extreme Spread

Post by rgchristensen » Mon Dec 23, 2019 11:02 pm

It's kind of like the comparison of group size vs mean radius, with extreme spread of velocity, you are only considering two of the 5 or 10 shots, while with SD of velocity, you are considering ALL of the shots. Why throw away all that information?
CHRIS
RGChristensen

mdeland
Posts: 11544
Joined: Mon Nov 20, 2006 1:47 pm

Re: Standard Deviation vs Extreme Spread

Post by mdeland » Mon Dec 23, 2019 11:55 pm

I don't know how to make practical use of all that additional information. Especially when group size does not necessarily have any correlation to SD or ES. At this point in the learning curve I believe the longer the distance being shot the more ES becomes the predominate indicator, especially with elevation.

User avatar
desert deuce
Posts: 3065
Joined: Mon Apr 14, 2008 10:51 pm
Location: Rio Rico, Arizona

Re: Standard Deviation vs Extreme Spread

Post by desert deuce » Tue Dec 24, 2019 12:05 am

Well Chris,...…….observing ES is important at long range because of vertical dispersion. ES is a precise number in comparison to SD.
A load with 4 ES compared to a load with 25 ES at 1,000 yards, calculate it for your long range bullet for vertical and see the difference in minutes of elevation for yourself. Even better, chronograph your load, make another series of shots 25 feet per second slower. (That is about a reduction of 4 grains of powder) Shoot both at 1,000 yards on target and see the difference 25 FPS makes. It is a visual thing. Think uniformity. Do you know the ES for your long range load?
Sometimes you get the chicken, and sometimes you get the feathers!

User avatar
desert deuce
Posts: 3065
Joined: Mon Apr 14, 2008 10:51 pm
Location: Rio Rico, Arizona

Re: Standard Deviation vs Extreme Spread

Post by desert deuce » Tue Dec 24, 2019 12:09 am

I see mdeland and I were typing at the same time.
Sometimes you get the chicken, and sometimes you get the feathers!

BFD
Posts: 2789
Joined: Mon Jan 21, 2008 7:36 pm

Re: Standard Deviation vs Extreme Spread

Post by BFD » Tue Dec 24, 2019 7:02 am

mdeland wrote:
Mon Dec 23, 2019 11:55 pm
I don't know how to make practical use of all that additional information. Especially when group size does not necessarily have any correlation to SD or ES. At this point in the learning curve I believe the longer the distance being shot the more ES becomes the predominate indicator, especially with elevation.

Do you know what the SD is of your group? You can figure it out. SD is not just for velocity. It is a fundamental measure of variation in any quantitative variable. Learning how to use it is helpful. ES is NOT the best measure. It is the EASIEST measure.

User avatar
DAVE ROELLE
Posts: 663
Joined: Sun Sep 19, 2010 11:30 am
Location: CONROE TEXAS

Re: Standard Deviation vs Extreme Spread

Post by DAVE ROELLE » Tue Dec 24, 2019 7:12 am

Standard deviation is hardware controlled with shooter and environment having minor effect

Extreme Spread has of course the hardware issues AND the shooter variables AND environmental variables

I'm betting the shooter and environment have more direct effect on accuracy than hardware does
your never lost, if ya don't care where ya are

BFD
Posts: 2789
Joined: Mon Jan 21, 2008 7:36 pm

Re: Standard Deviation vs Extreme Spread

Post by BFD » Tue Dec 24, 2019 7:19 am

DAVE ROELLE wrote:
Tue Dec 24, 2019 7:12 am
Standard deviation is hardware controlled with shooter and environment having minor effect

Extreme Spread has of course the hardware issues AND the shooter variables AND environmental variables

I'm betting the shooter and environment have more direct effect on accuracy than hardware does
Since when? I have no idea where you are getting that information from, but it is wrong.

SD is the square root of the mean squared distance from the mean of a group of observations.

Aviator
Posts: 284
Joined: Mon Sep 03, 2018 8:06 pm

Re: Standard Deviation vs Extreme Spread

Post by Aviator » Tue Dec 24, 2019 8:09 am

DAVE ROELLE wrote:
Tue Dec 24, 2019 7:12 am
Standard deviation is hardware controlled with shooter and environment having minor effect

Extreme Spread has of course the hardware issues AND the shooter variables AND environmental variables

I'm betting the shooter and environment have more direct effect on accuracy than hardware does
Your assertion is quite puzzling.

Perhaps you are (mistakingly) assuming that Standard Deviation is only related to measurement of muzzle velocity, and Extreme Spread is only related to measurement of holes in the target?

User avatar
DAVE ROELLE
Posts: 663
Joined: Sun Sep 19, 2010 11:30 am
Location: CONROE TEXAS

Re: Standard Deviation vs Extreme Spread

Post by DAVE ROELLE » Tue Dec 24, 2019 10:00 am

Brent, Aviator. Your both correct

I was looking at some targets showing groups and I screwed up thinking extreme spread of bullet holes in a target group. Not extreme spread or range of the velocity data

Merry Christmas 😉
your never lost, if ya don't care where ya are

User avatar
desert deuce
Posts: 3065
Joined: Mon Apr 14, 2008 10:51 pm
Location: Rio Rico, Arizona

Re: Standard Deviation vs Extreme Spread

Post by desert deuce » Tue Dec 24, 2019 9:34 pm

Clarification of terms: ES and SD as mentioned here are both velocity measurements of number of shots fired through a chronograph.

ES is the extreme spread of measured velocity normally taken by chronograph fifteen feet from the muzzle. An ES of 4 means there was 4 feet per second difference between the number of shots fired. ES of 25 means there was 25 feet per second difference between the number of shots fired. At a mean muzzle velocity of 1,300 feet per second muzzle velocity the lower the ES the less difference in vertical impact from shot to shot at 1,000 yards and that equals consistency in ammunition. It does not take into consideration environmental conditions or the design of the bullet.

How variations in ES affects the point of impact at the target can be "seen" by firing at the target 1,000 yards distant through a chronograph and observing the vertical dispersion at the target in relation to velocity measured fifteen feet from the muzzle. This was not my discovery it was given to me by Dan Theodore.

The theory, math and formulas are wonderful, however, when you are seeking to develop ammunition that will shoot to call at 1,000 yards you have to 'see' the results at the target. Numbers on paper are great but numbers on the score sheet matter more in the realm of possibilities we are exploring.

As stated above, reduce your 1,000 yard load by 4 grains of powder and see how much you have to come up in elevation at 1,000 yards to hit center. What matters is where the bullet hits the target or doesn't hit the target as the case may be.

As I have said before, "even seeing sometimes is not believing."
Sometimes you get the chicken, and sometimes you get the feathers!

bruce m
Posts: 3351
Joined: Tue Aug 04, 2009 5:25 am
Location: australia

Re: Standard Deviation vs Extreme Spread

Post by bruce m » Wed Dec 25, 2019 4:00 pm

this subject comes around every now and again, and there are more than 1 opinions.
here is mine. talking about velocity only.
first of all, velocity is only one indicator of drop, or group elevation, with faster bullets hitting the target higher than slower bullets.
this is complicated by a factor called positive or negative compensation, which involves barrel whip as the bullet travels down the bore.
one of the above will make slower bullets shoot higher than faster ones and vice versa.
the former is preferable, as at some distance the higher shooting ones will come down into the group of the faster ones, making a load that is excellent at that range, but with vertical further out and closer.
the other is just bad news.
while you can have a load for every range, it is a pain in the azz, compared to a load that suits all ranges.
extreme spread is a measurement of 2 shots only, the fastest and slowest, and is a definite number.
standard deviation is a statistical measure of all shots fired, and is suggestive of a result.
in my experience, an s.d. from 10 shots will suggest a TRUE e.s. for all shots of 4x.
for 20 shot s.d., the e.s. will be about 3x of the s.d.
the more shots registered for s.d., the better the number.
so often we see someone quoting an s.d. for 3 or 5 shots.
this is just irrelevant.
for hunting this is just irrelevant, but the further the range becomes the more attention needs to be taken.
I have seen s.d. coming down during a shot string, but e.s. will never do this.
assuming no positive or negative compensation, what this is saying is that the fastest and lowest shots will show max elevation of the group, but as the s.d decreases, the more shots will be closer to the centre of the group.
other than barrel compensation, such things as hold, wind direction, wind strength changes, mirage (and it can be there unseen) and other things can affect vertical.
for example, a wind direction change from 2.00 to 1.00 o'clock can change elevation by 2 moa at 1000 yds.
this makes testing loads at long range very hard.
bruce.
ventum est amicus meus

Post Reply