Measuring rifle / ammo performance

[rsilvers]

I have collected some papers on the subject:


http://home.wanadoo.nl/jhogema/skeetn/ballist/sgs/sgs.htm#_Toc96439743

http://www.the-long-family.com/group_size_analysis.htm

http://www.ar15.com/forums/topic.html?b=3&f=118&t=279218

http://www.shootersjournal.com/Features/WHICHONE.pdf (dead link)

http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA387108&Location=U2&doc=GetTRDoc.pdf

http://www.bobwheeler.com/guns/GroupStat.pdf


You can use RSI Shooting Lab or this program to do statistics:

http://www.rawilson.net/shareware/Gsc/index.html

http://home.wanadoo.nl/jhogema/skeet..._precision.htm


End result - radial standard deviation is best, and average mean radius is second best, and extreme spread is the least best... But... extreme spread is much easier to calculate from shooting at paper, and it turns out that if (and only if assuming not a huge amount of shots will be fired) you average several groups, you can get good results.

Shooting 3 shot groups over and over until you get a good one is not a valid measure of rifle performance.

It takes 20 shots to be 84% certain which is better when comparing two rifles/ammo which differ in performance by 20%. So to see the performance differences of two rifles with some degree of confidence, use at least average 4 groups of 5 shots each.

And likewise, any rifle of any quality can shoot a 0.5 MOA 3 shot group if you try enough times, so for a guarantee of precision of a rifle, the test should be at least an average of 4 groups of 5 shots each. Rifle makers may have to change their guarantee to 0.75 or 1 MOA or even 1.5 MOA though if you make them do an actual valid test.

 

[Lowlight]

Re: Measuring rifle / ammo performance

<div class="ubbcode-block"><div class="ubbcode-header">Originally Posted By: rsilvers</div><div class="ubbcode-body">
End result - radial standard deviation is best, and average mean radius is second best, and extreme spread is the least best... But... extreme spread is much easier to calculate from shooting at paper, and it turns out that if (and only if assuming not a huge amount of shots will be fired) you average several groups, you can get good results.

Shooting 3 shot groups over and over until you get a good one is not a valid measure of rifle performance.

It takes 20 shots to be 84% certain which is better when comparing two rifles/ammo which differ in performance by 20%. So to see the performance differences of two rifles with some degree of confidence, use at least average 4 groups of 5 shots each.

And likewise, any rifle of any quality can shoot a 0.5 MOA 3 shot group if you try enough times, so for a guarantee of precision of a rifle, the test should be at least an average of 4 groups of 5 shots each. Rifle makers may have to change their guarantee to 0.75 or 1 MOA or even 1.5 MOA though if you make them do an actual valid test.

</div></div>

It is why the rules on SH are "supposed" to be 5 - 5 shot groups.

Okay, I have to revisit the posting rules for Sniper's Hide specifically regarding 3 shot groups.

I know early on I said we can do 3 shot groups if they are beyond 100 yards, but I am gonna call myself out on this and change the rules.

From this day forward I am deleting any 3 shot groups period, and especially any groups that are groups within a group.

So here are the revised rules for posting groups on Sniper's Hide.


1. You must show the Entire target unedited.

2. Inside 300 yards you must post at least (5) 5 shot groups that can be identified on target or you may post (1) 10 shot group for record.

3. Anything beyond 300 yards a minimum of 5 shots is required, also please adhere to rule #1.

4. The group shot rules can be bypassed only with a target using 10 or more 1/2" dots with one individual round on each dot for a minimum of 10 shots.

5. Shots on steel can only be used for demonstration purposes and not as part of an exhibition of group shooting. These images must be in conjunction with rifles, optics, etc... they cannot be standalone.

6. Groups shots must be within 2 MOA of the aiming point, otherwise I consider the shots a miss.


This is my effort to help everyone strive to be a better shooter and to shun mediocrity. At Sniper's Hide we want to set the standard of what shooting good groups means, and we encourage those who are practicing to display their groups especially if you are working on improving your skills.

Any group currently posted on the site will be grandfathered in, however moving forward I ask the membership to be vigilant in enforcing the group shot rules and regulations. I think you'll thank me in the long run and we'll help raise the bar for everyone.

 

[jasonk]

Re: Measuring rifle / ammo performance

Interesting post and good info. I was wondering this just the other day after some email exchanges about load development. The 4 5-shot groups per load is exactly what I was thinking about for testing some 260 loads I'll be making soon.

I also agree with those that think it's a good idea to round robin with other charge weights so that all test loads get to experience different levels of fouling in the barrel. For example when testing 4 different charges around 40 grains of powder.

39.8 shoot 5, 40 - shoot 5, 40.2 shoot 5, 40.4 shoot 5.... and then repeat 3 more times. Just using that as an example, obviously you can load in .1 .2 or .3 grain increments, whatever works.

 

[THUNDERBOLT68]

Re: Measuring rifle / ammo performance

Great now I am not going to be able to post any groups.....
LL, that 5th shot gets me every time....

 

[JSTARSZ]

Re: Measuring rifle / ammo performance

I can live by those rules. I am not the best shot but nor have I ever claimed to be. I am not on the Hide to show my groups anyways as I am more than sure it will not be all that impressive to the group anyways. I am here to gain understanding and hopefully some wisdom from others on this site. I always appreciate the information on the site as I have looked for years for something like this and I applaud LL for bringing this to us. I just wish I could have thought of it first. Like any good education you learn better by being around others that share the same passions as you and are willing to share the knowledge with others.
Thanks for the post – the links were very informative.

 

[rsilvers]

Re: Measuring rifle / ammo performance

Wow - that rocks. Rules for posting groups is great. Can you plese consider extending the rules to manufactures who advertise accuracy guarantees? I have spoken to some makers and, at least the ones I talked to, know it is a farse but they are forced to guarantee accuracy based on an the best 3 shot group out of an unlimited number of 3 shot groups to compete with other makers. We need a standard in the industry just as SAE net horsepower has mostly fixed the wild and ridiculous horsepower claims of the 1960s.

I would not mind if 4 groups were allowed as 20 rounds are a box of ammo, but either way.

 

[rsilvers]

Re: Measuring rifle / ammo performance

<div class="ubbcode-block"><div class="ubbcode-header">Originally Posted By: THUNDERBOLT68</div><div class="ubbcode-body">Great now I am not going to be able to post any groups.....
LL, that 5th shot gets me every time....

</div></div>

Yes - 'we' are all lying to ourselves with three shot groups. If you take two rifles and fire a 3 shot groups of each, you cannot know which is better unless their actual difference in performance is wildly different. If one is "only" 40% more precise than the other, you cannot determine which is the better rifle using a 3-shot group. As such, an accuracy guarantee based on a 3-shot group, especially when you can try multiple times, has no significance. It is misleading.

 

[el gordo2]

Re: Measuring rifle / ammo performance

Going to miss the Tac Ops target group shot on a piece of cardboard.

 

[rsilvers]

Re: Measuring rifle / ammo performance

Round robin with powder charges is the way to go. The thing about it though is that the barrel heats up more so by the time you get back to the first target, the impact may be way off. But that is ok for comparing loads - I am just saying - don't be surprised if the groups are larger than expected because if the group has 5 shots on it, the barrel got as hot as if you fired 25 shots or whatever over the course of that 5 shot group. It may be best to limit shooting to 1 round per minute.

 

[cobaltbomb]

Re: Measuring rifle / ammo performance

good thing that you dont allow anything else than 5 shot grps.