Interesting results above, and they are pretty much in line with Western's 1954 tests when average ES is corrected for number of shots per group. The big problem with using the average ES of five 5-shot group ESs is statistical reliability (statistical significance) - it just ain't there with such a small number of groups. Statistical significance of the average ES of five 10-shot group ESs is much greater. Twice the ammunition produces much more reliable grouping results.

Another way to look at it is this - the standard deviation (SD) of a number of average ESs of five 5-shot groups will be approximately 13% of the mean group ES (from computer statistical modeling results). For five 10-shot groups, the corresponding SD will be <5% of the mean group ES. This is because that firing a single 10 shot group is the equivalent of firing 252 5-shot groups. That is correct, 252 5-shot groups. From a single group of 10 shots, you can make 252 different possible 5-shot group combinations. This is why I said that firing a small number of 5-shot groups for accuracy testing is a waste of time and ammunition if someone is really serious about making comparisons among various guns and/or ammunition types. All it's good for is zeroing your sights.

BTW - different types/brands of ammunition may rank differently in grouping performance when fired from different guns, i.e., some guns just seem to "like" different types/brands of ammunition better than others. In Frost's book (mentioned earlier) he also discusses this. In my case, my Winchester 52B has a definite preference for CCI standard velocity, which it shoots better than any other, including all brands of match ammunition I have tried. Federal American Eagle (40 grain HV) is also a pretty decent performer in my 52B despite its economy price.