Which is more indicative of load accuracy potential in a particular rifle?
a) one 3 shot group chosen for bragging rights
b) one 5 shot group chosen for bragging rights
c) the average of all 3 shot groups shot with a particular powder, bullet and seating depth
d) the median and standard deviation of all 3 shot groups shot with a particular powder, bullet and seating depth
e) the average of all 3 shot groups shot with a particular powder-bullet combination regardless of seating depth.
f) the median and standard of all 3 shot groups shot with a particular powder-bullet combination
To me, a) and b) are anomalies that don't represent anything but all the stars and planets being aligned for the shooter and don't represent the true performance of the shooter, the rifle and the load. It can't be claimed to be repeatable.
c) and e) would allow some amount of confidence if the number of groups measured were 20 or more.
d) and f) would allow an additional value if the number of groups measured were 20 or more because you could also calculate the percentage of groups you could expect to be with limits (such as the size range of groups you could expect to fall within 90% of the groups measured).
I measure all 3 and 5 round groups that I shoot and dutifully record them.
I use 3 round groups to test loads recording bullet and weight, powder and grains loaded to 0.1 grain, seating depth to 0.001, primers, and case length trim to 0.001. I generally shoot 6 groups at a minimum and only use the average to judge potential performance.
When I find loads that the rifle shoots consistently, I concentrate on taking larger samples.
I also find that gathering all groups shot with a powder-bullet combination provides additional information on what combinations are really the best performers, regardless of seating depth, velocities and trim length.
There is a difference in average group sizes between 3 and 5 round groups, possibility caused more by shooter variations and potential barrel heating than anything else.
My most accurate rifles show the following measured results. So which is the most accurate?
Savage 12 LPR 6.5mm Creedmoor
Best Ammo Avg. (5 - 3 rnd groups) = 0.161 --------- (5 rnd groups) = 0.302
Overall Average (169 - 3 rnd groups) = 0.259 - (5 rnd groups) = 0.435
Median (169 - 3 rnd groups) 0.249 StDev = 0.096
Group size range (from the smallest 5% to the largest 5%) = 0.233 to 0.265
Best powder-bullet combination (IMR 4350 142 gr TMK - 31 grps) avg. = 0.223
Median = 0.215 Std Dev = 0.086
.308 Savage 10 FP
Best Ammo Avg. (5 - 3 rnd groups) = 0.198 ---------(5 rnd groups) = 0.429
Overall Average (962 - 3 rnd groups) = 0.362 - (5 rnd groups) = 0.472
Median (962 - 3 rnd groups) 0.337 StDev = 0.152
Group size range (from the smallest 5% to the largest 5%) = 0.327 to 0.347
Best powder-bullet combination (IMR 4350 168 gr SMK- 31 grps) avg. = 0.223
Median = 0.286 Std Dev = 0.063
.223 Les Baer Super Varmint
Best Ammo Avg. (5 - 3 rnd groups) = 0.149 --------- (5 rnd groups) = 0.283
Overall Average (839 - 3 rnd groups) = 0.259 - (5 rnd groups) = 0.423
Median (839 -3 rnd groups) 0.266 StDev = 0.119
Group size range (from the smallest 5% to the largest 5%) = 0.257 to 0.275
Best powder-bullet combination (N140 69 TMK - 13 grps) avg. = 0.215
Median = 0.205 Std Dev = 0.080