I'm a bit new, but here's what I do - reload, shoot, analyze deeply...

[Ron_R]

Hey guys,
I've only posted a couple of things here, so I am a new guy. Before I get into the details of what I do, here's a bit about me.

I'm a retired electrical engineer and I love crunching and analyzing numbers. I now shoot almost every day and I shoot typically 30 rounds per session. I mostly test things in large sample sizes, i.e. 30 samples per whatever. For example, right now I'm loading 223/556 ammo with a light load, 22.0 gr of Winchester StaBall Match. I will load 30 rounds as identical as I can get them and go shoot them. The next day I may use the same components but increase the charge to 22.3 gr,

It's quite a task to search for any and/or all threads that would be similar to this, so if I'm duplicating, please forgive. I figure you guys that have been on the forum a long time will know about this kind of stuff.

I chrono (Garmin Xero) just about every reload I shoot and record the group sizes. I've been shooting ten groups of 3 shots per group but as of late I've started shooting 30 shots into one group. Using ShotMarker, I can feed all the hit coordinates into a spreadsheet and plot their location, as seen here.


In addition, because I have all the shot positions, I can also plot how the 10, 3-shot groups would look. The 3-shot groups are simply the shots, in order shot, broken down into groups of three. And I plot them along with the 30 shot group. That looks like this.


I won't get into all the info available in this, but it does enable one to see (top pic) how the mean radius looks compared to the classical extreme spread method. And it also shows (lower pic) what 3-shot groups might look like as opposed to one 30 shot group.

 

[charlie b]

Looks like you have the tools you need for the job. A Shotmarker is something I'd love to have but cannot justify the expense. I hand plot all my groups as I shoot them. Then analyze later. I'll plot and overlap different groups of shots as well to better see what trends there are from changes in powder charge, seat depth, etc.

When I shoot a ladder I do all the loads the same session. That way the range conditions are the same for all those shots. I have been known to repeat a ladder on another day, including changing the sequence of the shots, to see if anything changes. Most of the time results are similar.

The number of shots per charge (or seat depth setting) is interesting. 30 rounds is high. If you shoot 30 shots of each powder charge, eg, 10 powder charges, then that is 300 rounds. Then seat depth, another 300 rounds. Now the barrel has changed a bit. Do you repeat? Another 300 and you are looking at very little barrel life left for some cartridges.

This is why some choose 3 round groups to start, then 5 or 10 round to verify or narrow down the final load selections. More than 10 and you are looking at the far 'wings' of the distribution curve.

Then there is the biggest problem for me, my ability to shoot consistently

 

[Ron_R]

This is why some choose 3 round groups to start, then 5 or 10 round to verify or narrow down the final load selections. More than 10 and you are looking at the far 'wings' of the distribution curve.

Then there is the biggest problem for me, my ability to shoot consistently

Actually, 30+ shots is where the most reliable results are. Three to ten shot are considered a small sample size in statistics and are not expected to be reliable or repeatable. Some even say 50-100 samples are required for reliable, repeatable, predictive indicators.

Hornady did an extensive test, shooting thousands of rounds and carefully measuring everything. Ond of the things they found is that even shooting a precision rifle, mounted in a pedestal, in an indoor range, when they shot many 3 shot groups, like 30+, the group sizes varied by as much as 70%! That's a lot!

For example, if the tightest group was 0.5 MOA, a 70% variation would put the largest group at 1.67 MOA. When they shot large sample sizes, i.e. 30-50 shots per group, the group size variation went waaay down.

With the large sample sizes, again, 30+ per charge, they found that all the powder charges shot essentially the same group size. I find the same thing with my 30 shot groups.

It's a long watch, but well worth the time.

 

[Dufas]

Circle of Probability.

 

[Ron_R]

Circle of Probability.

Not following??? Help

 

[shoot4fun]

Hey guys,
I've only posted a couple of things here, so I am a new guy. Before I get into the details of what I do, here's a bit about me.

I'm a retired electrical engineer and I love crunching and analyzing numbers. I now shoot almost every day and I shoot typically 30 rounds per session. I mostly test things in large sample sizes, i.e. 30 samples per whatever. For example, right now I'm loading 223/556 ammo with a light load, 22.0 gr of Winchester StaBall Match. I will load 30 rounds as identical as I can get them and go shoot them. The next day I may use the same components but increase the charge to 22.3 gr,

It's quite a task to search for any and/or all threads that would be similar to this, so if I'm duplicating, please forgive. I figure you guys that have been on the forum a long time will know about this kind of stuff.

I chrono (Garmin Xero) just about every reload I shoot and record the group sizes. I've been shooting ten groups of 3 shots per group but as of late I've started shooting 30 shots into one group. Using ShotMarker, I can feed all the hit coordinates into a spreadsheet and plot their location, as seen here.
View attachment 1736044

In addition, because I have all the shot positions, I can also plot how the 10, 3-shot groups would look. The 3-shot groups are simply the shots, in order shot, broken down into groups of three. And I plot them along with the 30 shot group. That looks like this.

View attachment 1736043
I won't get into all the info available in this, but it does enable one to see (top pic) how the mean radius looks compared to the classical extreme spread method. And it also shows (lower pic) what 3-shot groups might look like as opposed to one 30 shot group.

In the end, what will you really know?
I'm talking about how much shooter error plays into the results and that is something that, at least I think, cannot be calculated into the equation.
Also I would want to the the absolute maximum accuracy potential of the rifle being used.
You have three three-shot groups under .75MOA and six of the other seven well above 1 MOA.
Concentrate on getting ten 3-shot groups under 1 MOA instead of thinking in terms of statistics.
Please don't misunderstand my point here. I am not being critical of what you want to do. It's just that the point is consistently small counts and these statistics don't provide that.
I would also like to see the numbers from the Garmin here as well.
Lastly, I am assuming this data was gathered at 100 yards?

 

[Steve Donlon]

I wonder what ammo Hornady used to get their results.

 

[Ron_R]

In the end, what will you really know?
I'm talking about how much shooter error plays into the results and that is something that, at least I think, cannot be calculated into the equation.
Also I would want to the the absolute maximum accuracy potential of the rifle being used.
You have three three-shot groups under .75MOA and six of the other seven well above 1 MOA.
Concentrate on getting ten 3-shot groups under 1 MOA instead of thinking in terms of statistics.
Please don't misunderstand my point here. I am not being critical of what you want to do. It's just that the point is consistently small counts and these statistics don't provide that.
I would also like to see the numbers from the Garmin here as well.
Lastly, I am assuming this data was gathered at 100 yards?

"In the end, what will you really know?"

Excellent question!
From 10 3-shot groups, I can calculate the 3-shot group size I have a 95% probability of hitting. In this case it's a dreadful 1.86 MOA. The 90% group size would be 1.7 MOA.

From the actual 30 shot group, the ES group size is 1.89 MOA. Notice how close the overall (ES) group size is to the 95% probability group of 1.86! Not surprising - it's all the same data, just two different ways of looking at it.

"I'm talking about how much shooter error plays into the results and that is something that, at least I think, cannot be calculated into the equation."

Another excellent point!

The shooter is 100% captured in the data. The results are the combination of the shooter, gun, ammo, brace, environment, etc. It's all in the data.

The only way to separate the shooter from the data is to put the gun in a vise such that the shooter is no longer a factor. That's the way Hornady did it.

"You have three three-shot groups under .75MOA and six of the other seven well above 1 MOA.
Concentrate on getting ten 3-shot groups under 1 MOA instead of thinking in terms of statistics."

That's what I do, but there will be group variation. This set was shot in an AR with a 12" CF barrel. I was disappointed with it's performance.

With a precision rifle in a pedestal mount, Hornady shot many 3-shot groups. They found that 3-shot group sizes vary between 60-70% - even with the gun in a pedestal. Based on my 10 groups, my group size variation is 69%.

The problem of NOT using statistics is which of the 10 groups best represent what we can expect from the gun? Given the 10, 3-shot groups I shot, sad though they may be, which one bests represents what I can expect from the gun? Using statistics, I can calculate what can be expected based on the actual group sizes. In this case, I have a 95% probability putting 3 shots in a 1.89 MOA group. Not a very good expectation.

It's just that the point is consistently small counts and these statistics don't provide that.

Not following, "consistently small counts"??? Don't all the groups count equally for evaluation of performance? The small group sizes, i.e. 3-shot groups, could not be obtained from the target. I had to extract them from the 30 shot group by the coordinates of each shot.

Bryan Litz talks about all this in an interview on the Ultimate Reloader channel. Bryan talks bout how difficult it is to boil the accuracy of a rifle down to one number. Again, his point is about group size variation. He even goes so far as to say that, and I have to paraphrase, "After shooting so many groups, I've come to realize those outliers are just a part of the statistical possibility."

 

[jelenko]

0.5 MOA, a 70% variation would put the largest group at 1.67 MOA.

How do you mean? .5 + .7[.5] = .85.

 

[Ron_R]

I wonder what ammo Hornady used to get their results.

I don't recall if they mentioned the type of ammo in their podcast. Probably did, and I don't remember.

 

[Ron_R]

How do you mean? .5 + .7[.5] = .85.

Hornady calculates the % variation by,
(max-min)/max * 100%

I've been using the above formula for a lot of groups and I keep getting in that 60-70% range.

At first I thought they would use (max-min)/min to see how much bigger (%) the large group was over the small group, but that gives numbers like 150%. Which is correct for using the min as the reference.

I think they should be using

(max-min)/(max+min)/2

So in my case I would get

(1.71-0.53)/(1.71+0.53)/2 * 100% = 105%

That's the difference formula.

 

[jelenko]

Not following, "consistently small counts"??? Don't all the groups count equally for evaluation of performance?

I think the question is if short range benchrest shooters can score consistently over many, many shot very small groups, what is preventing you from doing the same?

 

[Ron_R]

I think the question is if short range benchrest shooters can score consistently over many, many shot very small groups, what is preventing you from doing the same?

Equipment and technique. Not to mention triggers with breaks measured in ounces. You're thinking my AR with a 12" barrel would shoot just as good as highly customized BR guns?

 

[Webster]

Hey guys,
I've only posted a couple of things here, so I am a new guy. Before I get into the details of what I do, here's a bit about me.

I'm a retired electrical engineer and I love crunching and analyzing numbers. I now shoot almost every day and I shoot typically 30 rounds per session. I mostly test things in large sample sizes, i.e. 30 samples per whatever. For example, right now I'm loading 223/556 ammo with a light load, 22.0 gr of Winchester StaBall Match. I will load 30 rounds as identical as I can get them and go shoot them. The next day I may use the same components but increase the charge to 22.3 gr,

It's quite a task to search for any and/or all threads that would be similar to this, so if I'm duplicating, please forgive. I figure you guys that have been on the forum a long time will know about this kind of stuff.

I chrono (Garmin Xero) just about every reload I shoot and record the group sizes. I've been shooting ten groups of 3 shots per group but as of late I've started shooting 30 shots into one group. Using ShotMarker, I can feed all the hit coordinates into a spreadsheet and plot their location, as seen here.
View attachment 1736044

In addition, because I have all the shot positions, I can also plot how the 10, 3-shot groups would look. The 3-shot groups are simply the shots, in order shot, broken down into groups of three. And I plot them along with the 30 shot group. That looks like this.

View attachment 1736043
I won't get into all the info available in this, but it does enable one to see (top pic) how the mean radius looks compared to the classical extreme spread method. And it also shows (lower pic) what 3-shot groups might look like as opposed to one 30 shot group.

Yoiur going to burn out the barrel testing. Doesn't look like the rifle is capable of consistant small groups.

 

[Ron_R]

Yoiur going to burn out the barrel testing. Doesn't look like the rifle is capable of consistant small groups.

I could, but AR barrels are relatively inexpensive. And I'm shooting very light loads. Don't AR barrels last for about 10,000 or so rounds?

This one only has about 500 through it.

 

[shoot4fun]

I think the question is if short range benchrest shooters can score consistently over many, many shot very small groups, what is preventing you from doing the same?

Exactly!
@Ron_R My point is how do you take those statistics and improve either shooter performance, load performance, rifle performance, etc.
The whole point of shooting on paper is accuracy.
I would have used the most solid bench, rest and bag first.
Next I would determine the most accurate load for that rifle by using either Newberry's OCW or Audette's Ladder testing. Also, chronograph data can provide volumes on what you need to fix in your loading processes.
If, as an EE, you were given a project to design and spec components for would you pick bare minimum components for that job even if you knew some of those components were of dubious ability?
When you specs are used to create the finished product and even one component is subpar does the new finished product do what it was supposed to do?
It's the same for this project. Have you and are you choosing the correct components?

 

[TheOtherZilla]

I am retired electrician. So welcome to the forum. Any electrical problems I can help ya solve give me a shout. 45 years experience in all facets of the trade.

 

[jelenko]

Equipment and technique. Not to mention triggers with breaks measured in ounces. You're thinking my AR with a 12" barrel would shoot just as good as highly customized BR guns?

If your technique is a significant contributor to the groups, then what is the objective of doing the testing?

ETA. shoot4fun beat me to it.

 

[JFrank]

With all due respect this looks like a classic example of over complicating a very simple process.

 

[Webster]

Excellent question!
From 10 3-shot groups, I can calculate the 3-shot group size I have a 95% probability of hitting. In this case it's a dreadful 1.86 MOA. The 90% group size would be 1.7 MOA.

From the actual 30 shot group, the ES group size is 1.89 MOA. Notice how close the overall (ES) group size is to the 95% probability group of 1.86! Not surprising - it's all the same data, just two different ways of looking at it.


Another excellent point!

The shooter is 100% captured in the data. The results are the combination of the shooter, gun, ammo, brace, environment, etc. It's all in the data.

The only way to separate the shooter from the data is to put the gun in a vise such that the shooter is no longer a factor. That's the way Hornady did it.


That's what I do, but there will be group variation. This set was shot in an AR with a 12" CF barrel. I was disappointed with it's performance.

With a precision rifle in a pedestal mount, Hornady shot many 3-shot groups. They found that 3-shot group sizes vary between 60-70% - even with the gun in a pedestal. Based on my 10 groups, my group size variation is 69%.

The problem of NOT using statistics is which of the 10 groups best represent what we can expect from the gun? Given the 10, 3-shot groups I shot, sad though they may be, which one bests represents what I can expect from the gun? Using statistics, I can calculate what can be expected based on the actual group sizes. In this case, I have a 95% probability putting 3 shots in a 1.89 MOA group. Not a very good expectation.


Not following, "consistently small counts"??? Don't all the groups count equally for evaluation of performance? The small group sizes, i.e. 3-shot groups, could not be obtained from the target. I had to extract them from the 30 shot group by the coordinates of each shot.

Bryan Litz talks about all this in an interview on the Ultimate Reloader channel. Bryan talks bout how difficult it is to boil the accuracy of a rifle down to one number. Again, his point is about group size variation. He even goes so far as to say that, and I have to paraphrase, "After shooting so many groups, I've come to realize those outliers are just a part of the statistical possibility."

At some point you have to decide is this the best the rifle can do. How much are you contributing to the group? Looks like your testing one bullet and powder over and over again. You mentioned cheap AR barrels. If the rifle doesn't want to shoot about 1/2-3/4" groups it's telling you a lot about the barrel quality. It won't shoot small groups just because you want it to.