Hi,
OK, a newbie on this forum… though I have done a lot of homework on long-range shooting. I need some statistical help on sampling. I had a college 300-level statistics class about 30 years ago, but I’m afraid that that is not helping out here! :'(
Here is the main issue: I am fairly new to long-range shooting. I bought a Savage Model 12 Long Range Precision in 6.5 Creedmoor with
I have a very heavy rifle rest, and a very nice person at the 1000 yard range in Missoula took me to his house on the way home from a match, and made me a flat aluminum piece that screwed into the front of my rifle it can fit very snugly with the rifle rest. (I now know it is better to have the stock of the rifle have a completely flat and wide surface, but that is maybe for my next rifle …)
At a match late in 2013 in Missoula I had a group size of 21â€. I think I can do a good deal better than that, but much of this is with your help. (BTW, one of the guys there just shot a 2.68†group in the light gun category, which is of course better than any heavy gun record. The record certification is pending: link). Finally, I should note that my son and I have had some long-range shooting classes.
I am an insanely busy technical professional so I don’t have any time to handload, at least not for at leat a few years. Ergo, upon the advice from some who owns the same rig as mine and who has shot on the US national team (lets call him “MMâ€), I ordered some semi-custom ammo. This company, which we’ll label “CC†(after the Unix C compiler), offered a load development service and then ammo based on it. I bought 1200 rounds from them and still have about 700 left. So I have to use them as best as I can. Ergo my questions to this forum. I paid for a load development service (~$250) then for the actual ammo (~$2.10/round). The rounds used a Berger Hybrid match with a G7 BC of .317 plus 8% that CC claimed from adding a point to the rounds.
Unfortunately, when I was at my local range last year, the CC ammo had a very big MV spread: at least 40fps (sample size of ~10; I know this is low) and there were not just one or two outliers. (I have to clean my garage to find my record book to get the exact numbers…) Very sad: I know from Litz’s book that this is pretty awful. And the chronometer seems to be very solid.
So I decided to do some measurements on the CC ammo, with a lot of homework ahead of time. Actually hired my 18-year-old son and his friend, both with very good math skills. (I have a BS in Math, but way too many years….) We measured length(in) and weight(g) of about 400 rounds. The idea here is that likely the MV depends on one or both of these. Originally some of the shooters at my 1000 yard range thought that this Berger bullet was heavily dependent on seating depth. But later they thought it was not dependent on seating depth. So maybe the MV depends only on the weight, which one presumes is dependent on the powder charge (brass and bullet weighing consistently). At least those are my novice suppositions.
Below is info on this. But my main point in posting to this forum is to ascertain which rounds I should use to test MV. Originally, the choice of my son and I was to choose an “almost-max†and “almost-min†in Len and Wt. “Almost†means not the outlier, but not far away whereby there are a lot of samples near it. So then we would have the cross product {almost-MIN, almost-MAX} X {almost-MIN,almost-MAX} for 4 sample sets. Probably I would be looking at 10-16 samples/set. The idea is to get the biggest cluster at each end of both variables. That should make it clear if the MV is dependent on one or both of those.
The other way of choosing a sample set is randomly. Methinks we should do as per above, with MAX an MIN, so to have a reasonable set for each combo of {MIN,MAX} (here “reasonable†includes the notion that this is high-quality match ammo, not garden-variety plinking stuff). But my son’s friend and now he think that we should do this random way. I have not had statistics as recently as my son and his friend have; but, of course, that does not mean that they have more common sense and/or wisdom on this issue. So inquiring minds want to know….
So whaddaya think? Below are some tables from my measurements. FWIW the length was calculated with a Hornady Electronic Caliper 6â€, and for weight a Frankford Arsenal DS0750. Check out the reviews and specs there. Both were very repeatable.
But I need help in deciding how to choose the sample set from the CC cartridges. The local range I go to has a permanently-mounted Oehler 35P so we are probably pretty good here. I do know that when I brought a 300’ landscape steel measure it was within 1†from the end of the shooting bench to the middle of the target holder. So they seem to be very careful here. And even if it were set up not 100% über-perfect (e.g., the sensors a teeny bit too far away from each other) that would only affect the accuracy of the measurement, not its precision. I.e., the measured MV would have been spread out just as much.
To add more data into the mix, I also measured 20 rounds each from other 6.5 Creedmoor rounds:
1. Hornady Max 140-AMAX,
2. Nosler Ballistick Tip Trophy Grade 140BT, and
3. Nosler Custom Competition Match Grade 140 HPBT.
These were ones I could find at Midway (could not find any elsewhere). Besides measuring them now, I plan on firing a box of each of these to see the variance in MV for a full box and of course how, if any, it correlates with Weight and Length.
And, while this is not 6.5 Creedmoor, I bought my .308 hunting rifle and tried out some round. After a fouling shot the next 4 rounds with Federal Gold Medal 2343, {2524, 2523,2520,2550}. I know this is a small sample size, but this is a very very small STDEV. Kinda puts the 6.5 Creedmoor (a round supposedly only for competition) so shame.
So, then, here below are my data. Assuming that the seating depth does not matter on the 140 Berger Hybrid, the main issue here is the weight. I think… and welcome your input.
Summary Data
Here, “STD%†means the STDEV/MEAN. This normalizes the groupings, kinda, I think.
Sample Size Length STD% Weight STD%
CC Berger Hybrid Match 140 391 0.67064 0.44433
Hornady Max AMAX 140 20 0.1302 0.23629
Nosler Ballistic Tip Trophy Grade 140 BT 20 0.10878 0.19487
Nosler Custom Competition Match Grade 140 HPBT 20 0.16847 0.28146
Of course, the CC rounds were supposedly customized to the harmonics of my rifle (yes, I shipped it there). So, even though its STD% is greater, it may in theory perform much better than the STD% may indicate. Maybe. I’ll have a lot better data on that once I go to the range and measure MV with a well-chosen set of rounds. But, before I do, I need advice on how to take a sampling from my 391 CC cartridges. Please give me advice on how to choose well!
And, for that matter, I am wondering why someone getting factory match ammo would not go ahead and measure and/or weigh each round to try to group similar ones. At least if that made a difference with MV. Thoughts on this?
Finally, FWIW I have not yet measured another 300 or so rounds; I may well be able to do this before I go shoot. But the difference in the stats for 391 vs ~700 would presumably be minimal. And it does not affect the methodology by which I choose my sample rounds to shoot and measure MV.
Thanks in advance for any help here. I know that you guys are handloaders, but you are very well aware of all the variables that can affect MV in a round.
OK, a newbie on this forum… though I have done a lot of homework on long-range shooting. I need some statistical help on sampling. I had a college 300-level statistics class about 30 years ago, but I’m afraid that that is not helping out here! :'(
Here is the main issue: I am fairly new to long-range shooting. I bought a Savage Model 12 Long Range Precision in 6.5 Creedmoor with
- Sightron SIII 8-32x56 MOA-2 scope
- Ken Farrell base (30 MOA tapered)
- Ken Farrell scope rings
- Flatline Ops sight level
- Center Shot Rifles muzzle brake (had a local custom rifle guy thread the bore and set this all up right)
I have a very heavy rifle rest, and a very nice person at the 1000 yard range in Missoula took me to his house on the way home from a match, and made me a flat aluminum piece that screwed into the front of my rifle it can fit very snugly with the rifle rest. (I now know it is better to have the stock of the rifle have a completely flat and wide surface, but that is maybe for my next rifle …)
At a match late in 2013 in Missoula I had a group size of 21â€. I think I can do a good deal better than that, but much of this is with your help. (BTW, one of the guys there just shot a 2.68†group in the light gun category, which is of course better than any heavy gun record. The record certification is pending: link). Finally, I should note that my son and I have had some long-range shooting classes.
I am an insanely busy technical professional so I don’t have any time to handload, at least not for at leat a few years. Ergo, upon the advice from some who owns the same rig as mine and who has shot on the US national team (lets call him “MMâ€), I ordered some semi-custom ammo. This company, which we’ll label “CC†(after the Unix C compiler), offered a load development service and then ammo based on it. I bought 1200 rounds from them and still have about 700 left. So I have to use them as best as I can. Ergo my questions to this forum. I paid for a load development service (~$250) then for the actual ammo (~$2.10/round). The rounds used a Berger Hybrid match with a G7 BC of .317 plus 8% that CC claimed from adding a point to the rounds.
Unfortunately, when I was at my local range last year, the CC ammo had a very big MV spread: at least 40fps (sample size of ~10; I know this is low) and there were not just one or two outliers. (I have to clean my garage to find my record book to get the exact numbers…) Very sad: I know from Litz’s book that this is pretty awful. And the chronometer seems to be very solid.
So I decided to do some measurements on the CC ammo, with a lot of homework ahead of time. Actually hired my 18-year-old son and his friend, both with very good math skills. (I have a BS in Math, but way too many years….) We measured length(in) and weight(g) of about 400 rounds. The idea here is that likely the MV depends on one or both of these. Originally some of the shooters at my 1000 yard range thought that this Berger bullet was heavily dependent on seating depth. But later they thought it was not dependent on seating depth. So maybe the MV depends only on the weight, which one presumes is dependent on the powder charge (brass and bullet weighing consistently). At least those are my novice suppositions.
Below is info on this. But my main point in posting to this forum is to ascertain which rounds I should use to test MV. Originally, the choice of my son and I was to choose an “almost-max†and “almost-min†in Len and Wt. “Almost†means not the outlier, but not far away whereby there are a lot of samples near it. So then we would have the cross product {almost-MIN, almost-MAX} X {almost-MIN,almost-MAX} for 4 sample sets. Probably I would be looking at 10-16 samples/set. The idea is to get the biggest cluster at each end of both variables. That should make it clear if the MV is dependent on one or both of those.
The other way of choosing a sample set is randomly. Methinks we should do as per above, with MAX an MIN, so to have a reasonable set for each combo of {MIN,MAX} (here “reasonable†includes the notion that this is high-quality match ammo, not garden-variety plinking stuff). But my son’s friend and now he think that we should do this random way. I have not had statistics as recently as my son and his friend have; but, of course, that does not mean that they have more common sense and/or wisdom on this issue. So inquiring minds want to know….
So whaddaya think? Below are some tables from my measurements. FWIW the length was calculated with a Hornady Electronic Caliper 6â€, and for weight a Frankford Arsenal DS0750. Check out the reviews and specs there. Both were very repeatable.
But I need help in deciding how to choose the sample set from the CC cartridges. The local range I go to has a permanently-mounted Oehler 35P so we are probably pretty good here. I do know that when I brought a 300’ landscape steel measure it was within 1†from the end of the shooting bench to the middle of the target holder. So they seem to be very careful here. And even if it were set up not 100% über-perfect (e.g., the sensors a teeny bit too far away from each other) that would only affect the accuracy of the measurement, not its precision. I.e., the measured MV would have been spread out just as much.
To add more data into the mix, I also measured 20 rounds each from other 6.5 Creedmoor rounds:
1. Hornady Max 140-AMAX,
2. Nosler Ballistick Tip Trophy Grade 140BT, and
3. Nosler Custom Competition Match Grade 140 HPBT.
These were ones I could find at Midway (could not find any elsewhere). Besides measuring them now, I plan on firing a box of each of these to see the variance in MV for a full box and of course how, if any, it correlates with Weight and Length.
And, while this is not 6.5 Creedmoor, I bought my .308 hunting rifle and tried out some round. After a fouling shot the next 4 rounds with Federal Gold Medal 2343, {2524, 2523,2520,2550}. I know this is a small sample size, but this is a very very small STDEV. Kinda puts the 6.5 Creedmoor (a round supposedly only for competition) so shame.
So, then, here below are my data. Assuming that the seating depth does not matter on the 140 Berger Hybrid, the main issue here is the weight. I think… and welcome your input.
Summary Data
Here, “STD%†means the STDEV/MEAN. This normalizes the groupings, kinda, I think.
Sample Size Length STD% Weight STD%
CC Berger Hybrid Match 140 391 0.67064 0.44433
Hornady Max AMAX 140 20 0.1302 0.23629
Nosler Ballistic Tip Trophy Grade 140 BT 20 0.10878 0.19487
Nosler Custom Competition Match Grade 140 HPBT 20 0.16847 0.28146
Of course, the CC rounds were supposedly customized to the harmonics of my rifle (yes, I shipped it there). So, even though its STD% is greater, it may in theory perform much better than the STD% may indicate. Maybe. I’ll have a lot better data on that once I go to the range and measure MV with a well-chosen set of rounds. But, before I do, I need advice on how to take a sampling from my 391 CC cartridges. Please give me advice on how to choose well!
And, for that matter, I am wondering why someone getting factory match ammo would not go ahead and measure and/or weigh each round to try to group similar ones. At least if that made a difference with MV. Thoughts on this?
Finally, FWIW I have not yet measured another 300 or so rounds; I may well be able to do this before I go shoot. But the difference in the stats for 391 vs ~700 would presumably be minimal. And it does not affect the methodology by which I choose my sample rounds to shoot and measure MV.
Thanks in advance for any help here. I know that you guys are handloaders, but you are very well aware of all the variables that can affect MV in a round.