Group size: Moving towards using 10 shot ATC instead of ES

[P2000]

If this has been covered already, I apologize.

To my annoyance, I hear the term MOA being thrown around constantly. Not just online, but everywhere people talk about accuracy. The way most people use the term MOA is an absolutely meaningless measure of accuracy.



For example,

"My rifle/load shoots 1/2 MOA"
This statement might really mean that the extreme spread of the best 3 shot group measured 1/2'' at 100 yards. The next group could have easily measured triple the size, or 1.5''.


With the availability of free calculators online which will quickly measure 10 shot groups using the superior ATC(Average To Center), what do you think about trying to make it the "norm". It would make it easier for us to know someone's method of determining group size, adding value to their statement. For example:


"My rifle/load shoots 0.212 10SATC"
With this statement, we know we are talking about a 10 shot group measured with a calculator tool, using ATC. A subsequent group is unlikely to measure significantly different, perhaps 0.290. Definitely not triple, or double.

What do you think? Would this ever catch on? I realize many members already measure ATC, because M4carbine is ahead of the game, but If we started using 10SATC, it may catch on.

[Cameron]

Relax mate, we know what people mean when they say 1/2 MOA.

Cameron

[montrala]

You have something like that in mind?



100m, HK MR223, Ruag 55gr FMJ bulk ammo, Aimpoint CompM4S, rifle rested on backpack.

10 shots ATC has this good side, that lone "flyer" does not kill ability to compare groups so much, without need to remove "called flyer caused by some butterfly passing near bullet" from equation.

I use only 10 shots groups and ATC as main indicator if given change to rifle or ammo resulted in gain or loss for accuracy.

For hunt of best CTC groups nothing beats 1-shot groups

[P2000]

You have something like that in mind?

Exactly. Real data with solid statistics. Nice shooting.

Also, you bring up a good point about flyers...with ATC they don't throw off your group too badly. With the other method, people tend to just shoot more 3 shot groups until they are pleased.

[QuietShootr]

Where can i read more on this? I'm ashamed to say that i work with statistics a lot, and this idea never occurred to me. Really, it makes a hell of a lot of sense- since there are a lot of reasons that a flyer can happen, and they aren't all indicative of the system's capability. Accounting for outliers but not allowing them to define the distribution is brilliant.

[sinister]

What you call "Average to Center" is the "Mean radius" -- how the Army measures match rifle and ammunition performance.

The AMU typically proofs match rifles, pistols, and carbines from machine rests, usually three ten-shot groups, measuring bullet speed at 10 feet from the muzzle and at 300 meters.

Molon has written and photographed wonderful ammunition proofing groups, both here and on the other site.

[shootist~]

Molon's posts dealing with testing various ammo usually include the mean radius of the groups, which is the same thing as ATC. I've always considered (just by eyeballing) how a group "clusters", or fails to cluster most shots near the center. Crude, but similar.

On Target software is a better way to do this that just using the Mk1 eyeball.

[Tuukka]

Well, having an accurate semi auto makes this whole thing a bit easier

[P2000]

Where can i read more on this? I'm ashamed to say that i work with statistics a lot, and this idea never occurred to me. Really, it makes a hell of a lot of sense- since there are a lot of reasons that a flyer can happen, and they aren't all indicative of the system's capability. Accounting for outliers but not allowing them to define the distribution is brilliant.

I remember some posts by Molon, which actually got me using the online calculators awhile back. I would try and find those and read up if possible.

I will try and explain why fliers are less important with Average To Center than Extreme Spread.
-With ATC, every shot is a data point. In other words, a 10 shot group has 10 data points, which are the distance of the shot to the mathematical center of the grouping. One flier only has 1/10 of the weight of the group...it isn't going to matter that much. With 10 shot groups, you can shoot group after group and they are going to have relatively similar ATC values.
-With ES, only the distance between the two farthest apart shots are counted. This means that with a 10 shot group, there is only one measurement, or one data point. So you could have a dime sized 10 shot group with one flier, and your group measures very poorly.

I would download the free copy of the older version calculator at http://www.ontargetshooting.com/download1.html

Give it a try. All you need to do is take a picture of your target, preferably with a ruler on it. Then upload the picture into the calculator software, set the scale based on the ruler, enter the bullet holes and voila...reliable statistics at one's disposal.

[QuietShootr]

I remember some posts by Molon, which actually got me using the online calculators awhile back. I would try and find those and read up if possible.

I will try and explain why fliers are less important with Average To Center than Extreme Spread.
-With ATC, every shot is a data point. In other words, a 10 shot group has 10 data points, which are the distance of the shot to the mathematical center of the grouping. One flier only has 1/10 of the weight of the group...it isn't going to matter that much. With 10 shot groups, you can shoot group after group and they are going to have relatively similar ATC values.
-With ES, only the distance between the two farthest apart shots are counted. This means that with a 10 shot group, there is only one measurement, or one data point. So you could have a dime sized 10 shot group with one flier, and your group measures very poorly.

I would download the free copy of the older version calculator at http://www.ontargetshooting.com/download1.html

Give it a try. All you need to do is take a picture of your target, preferably with a ruler on it. Then upload the picture into the calculator software, set the scale based on the ruler, enter the bullet holes and voila...reliable statistics at one's disposal.

I'm sold. I'm going to shoot my 300 early in the morning, and I'm going to process the data with the software and see what I come up with.