Best way to determine what a group really measures

[Mulligan]

I am down the path a ways comparing field measured targets with what I measure at home with a group measuring device from Neil Jones and OnTarget software.

Here is the crux of the situation.
I believe...... now I know this may not be entirely true in every situation, but it is close.

When a potential record target is shot at a benchrest match (long range and short range), the Match Director and target crew have to mail the target or targets and backers if short range to the sponsoring organizations records leadership. I think the Regional Honcho is involved in the case of the NBRSA? Then that person, sends the target or targets to each of the committee members who have been blessed/trained/appointed/ or coerced into taking their time to measure targets in some official way. Along the path each target or targets go until they have crossed the desk of each official measurer and returned to the group guru for the sponsoring organization. I think most organizations use 3-5 folks to measure the group/ groups in question and then use the average of those measurements as the official measurement.

I have NEVER measured a group at home after a match..... even the matches I have been the Match Director at, that measured at home what it did at the range.
Last Saturday, I shot a club level short range benchrest fun shoot. Warm-up plus 5 targets at 100 yards. I was shooting my PPC.

In an effort to separate the fly do-do from the pepper, I went to measuring groups with my good calipers, cheap calipers, and On-Target software, all to no avail. In no way shape or form could I get any two of the three systems to correlate. With me running the Neil Jones group measuring device on two of my calipers, the groups measured so close as to be really insignificant, but the field measurement, me on any of my calipers, or the On-Target software could not agree.

How do I (we) establish a base line for what is acceptable with so much potential error? Each of the sponsoring organizations use a committee and the average of those extra super ninja experts is the accepted "real" measurement. I am grateful for these folks volunteering their time and energy to do this, I really am. However what this says is there really is error in our system and we accept it.

All of this is in an effort to develop a method to test the accuracy of eTargets. How do we know if a measurement is real or not when the best system we have is to have 3-5 experts measure it and take an average of that? Again, I really am grateful to the field measurers and the experts working behind the scenes so we can all play the games we like to play. I have never argued a group size at a match and I hope I never do. I have questioned it more than once and then remembered that someone is volunteering their time to do the group measuring and would likely be doing something else and I am happy again to live in America and be playing the game. So, many thanks to those kind soles who preform this work.

I really want to establish a good way to compare "real" to system or field measured so I can get back to to the work at hand..... evaluating eTargets for use in recording group size and score.

Is there a TRUSTED software program that does this? Is a committee of gurus the answer?

Any help is appreciated.

Kind Regards
CW

BTW

A couple of weeks ago, one of the local BR (both LR and SR) shot 6 targets at 100 yards on a small frame built for the shot marker system. He then had several SR benchrest shooters measure the groups independently and compared the data.
The extreme spread of the committee was on average about .024". The shot marker average group size differed from the committee average by about .005" . What is the truth?
CW



Edit, I should have added that we are going to be shooting several 600 yard benchrest matches in the coming months and I want to compare the measuring systems from real matches with several systems running and lots of targets to get a solid data set.
Shipping a bunch of targets all over the country would be very expensive and time consuming, not to mention all the time involved in actually measuring hundreds of targets if we could find folks to do this.

CW

 

[BoydAllen]

I think that it would be a good idea to add a little information to your thread. How much difference was there between the various methods that you used to measure the same group? Another interesting number would be the amount of variation in the measurements of the same group there is measured by multiple persons, using the type of tool that is normal for short range group matches. I have had a Jones tool on a dedicated dial caliper, for many years. I picked it up out of an estate sale. When I have compared my measurements of the same target using it to those made with bare calipers, the caliper measurements were slightly larger. With a caliper I measure to the edge of the black line around the bullet hole or holes. The number that I subtract from the maximum width of the group is a careful measurement of a single bullet hole in the same paper, shot on the same backer.

 

[DaveTooley]

I was on the IBS records committee for a lot of years. We all used electronic calipers with original Sweeny reticles. Much like Neil Jones. Those reticles have measured every record target ever submitted to the IBS. Our tolerance for group targets whether single or aggregate was .010". Larger than that and we start over. I don't recall us ever having to remeasure targets. We were that consistent.
It is impossible to get a truly accurate measurement just using calipers. Not all bullet holes of the same caliber measure the same. Ask the short range guys about that. An in tune gun will shoot cleaner rounder holes.
At a match the number isn't as important as consistency. Measure big-measure small, just measure the same every time.

 

[Mulligan]

I was on the IBS records committee for a lot of years. We all used electronic calipers with original Sweeny reticles. Much like Neil Jones. Those reticles have measured every record target ever submitted to the IBS. Our tolerance for group targets whether single or aggregate was .010". Larger than that and we start over. I don't recall us ever having to remeasure targets. We were that consistent.
It is impossible to get a truly accurate measurement just using calipers. Not all bullet holes of the same caliber measure the same. Ask the short range guys about that. An in tune gun will shoot cleaner rounder holes.
At a match the number isn't as important as consistency. Measure big-measure small, just measure the same every time.

Dave, thanks for taking the time to respond. This is really good stuff.


At a match the number isn't as important as consistency.
^^^ this is absolutely true^^^^^^^

CW

 

[quest450]

FWIW I had my four targets measured by IBS scoring committee this winter
My 4 target agg measured by four judges at the gun range was 1.598"
Official measurement was 1.607" with three additional signatures
That's only 0.009" difference which is totally acceptable in my book
But forsheets and giggles I just measured it again to the best of my ability using Balistic - X app and I came up with 1.551" so there's 0.056" discrepancy between human and a machine and that's not acceptable IMO.

 

[Clancy]

I have checked the on target software time and time again and never seen more than just a few thousandths. Like 3 thou difference in that and calipers and that’s hard to hold or see.

Make sure when you set the 1” reference that you are exactly sure on the size and even the side of the line that you’re measuring.

I mean that with all due respect just saying that can really extrapolate.

 

[quest450]

I did, I’m able to zoom in really close on my iPhone to set the reference I think most of my errors is in overlaying th actual bullet holes, they are kinda dark and difficult to get it just right even when I zoom in

 

[BoydAllen]

FWIW I had my four targets measured by IBS scoring committee this winter
My 4 target agg measured by four judges at the gun range was 1.598"
Official measurement was 1.607" with three additional signatures
That's only 0.009" difference which is totally acceptable in my book
But forsheets and giggles I just measured it again to the best of my ability using Balistic - X app and I came up with 1.551" so there's 0.056" discrepancy between human and a machine and that's not acceptable IMO.

Having measured a few targets myself, both with the Jones tool and with bare calipers, I do not see how it would be possible to be off that much with either, which makes me suspect the program, or how it was used.

 

[quest450]

Most likely user error
I just double checked the inner circle of the X is a touch over 1” and I assumed it was an inch, there’s one problem

 

[jackieschmidt]

At most Matches, we are not that concerned about the actual scoring procedure as long as they are consistent from target to target and competitor to competitor.

Several years ago, there was a match held in southern Louisiana that had two different people scoring the targets. That was not a good idea.

Of course, anytime there is what appears to be a record target or aggregate shot, you want to be sure that the procedures are set to a National Standard, such as used by the NBRSA and IBS.

This can come into play at some very inopportune times. Several years ago, there was a Nationals held that on several days, had some really great conditions conducive to records being set.

In one yardage, two shooters shot below the national record, one came in first, the other second. The aggs between the two were very close to each other. The first place shooter got the HOF point.

However, when the targets came back from the records committee, they flipped. The shooter that came in second by the range measurements was found, by the records committee members, to have actually beat the first place shooters agg by a few thousandths.

So, who really came in first, and who should have been awarded the HOF point.

It was finally decided that the range measurements would be used.

Now, this was the story that came down. I can’t attest to it being true or Benchrest fokelore. But it does illustrate a scenario that could crop up when dealing with very small numbers and human short comings.

In short, we do the best we can. We are still measuring bullet holes in paper targets, trying to ascertain numbers within .001 inch.

As for the electronic targets and the accuracy of the actual group measurements, I have no opinion.

 

[jepp2]

so there's 0.056" discrepancy between human and a machine and that's not acceptable IMO.

This a personal comment about myself, not a comment on you.

I measure lots of targets for my own personal benefit. When I scan them in using a quality flatbed scanner, and make my OnTarget reference much larger (around 8 inches), I get very high quality results. When I take a picture of a target too large to fit on my scanner, I am not able to obtain nearly as good of results. It is a challenge having the picture taken exactly in the center of the target perfectly perpendicular to the target face, and capturing the image exactly to scale (across the entire width/length). Ballistic-X provides me with a more higher error result.

 

[David Milisock]

At most Matches, we are not that concerned about the actual scoring procedure as long as they are consistent from target to target and competitor to competitor.

Several years ago, there was a match held in southern Louisiana that had two different people scoring the targets. That was not a good idea.

Of course, anytime there is what appears to be a record target or aggregate shot, you want to be sure that the procedures are set to a National Standard, such as used by the NBRSA and IBS.

This can come into play at some very inopportune times. Several years ago, there was a Nationals held that on several days, had some really great conditions conducive to records being set.

In one yardage, two shooters shot below the national record, one came in first, the other second. The aggs between the two were very close to each other. The first place shooter got the HOF point.

However, when the targets came back from the records committee, they flipped. The shooter that came in second by the range measurements was found, by the records committee members, to have actually beat the first place shooters agg by a few thousandths.

So, who really came in first, and who should have been awarded the HOF point.

It was finally decided that the range measurements would be used.

Now, this was the story that came down. I can’t attest to it being true or Benchrest fokelore. But it does illustrate a scenario that could crop up when dealing with very small numbers and human short comings.

In short, we do the best we can. We are still measuring bullet holes in paper targets, trying to ascertain numbers within .001 inch.

As for the electronic targets and the accuracy of the actual group measurements, I have no opinion.

Goes to show how much BS there is in the process. One of the multitude of reasons not to compete.

I was a printing pressman for years. By changing the plate and blanket packing you can change the size of a 1" printed circle several thousands of an inch in either direction. The size can vary maybe .001 throughout a long run with the same packing.

If you're serious about measuring then all targets used need to be certified and from the same lot. With a size scale imprinted on the target.

The only way I know to measure a group accurately is to utilize a vector graphics application. Create perfect bullet diameter circles with the .001 outline set to the inside of the bounding box and center a .001" outlined X in each circle.

Scan the target and adjust for accuracy of size, place the bullet sized circles in the center of each hole. Draw a perfect circle encompassing the center strike of each bullet hole. The resulting large circle is the group size. The computer always measures the circles to the bounding box.

 

[Mulligan]

On a recommendation from @BoydAllen yesterday, I scanned a target with a small machinists ruler in the “photo”.

The combination of a very accurate scale (ruler) and taking angles from cameras out of the equation, the results using the On-Target software is very good.

CW