I did a small experiment at 1000 yards recently . Fired six shots in a row at the end of a match string. 3 shots had a short bearing surface length and three shots had a bearing surface length .009" longer. I use the Tubb comparators at both ends of the bullet to get total bearing surface length (BSL). I alternated short, long, short and made no scope or aiming adjustments. The wind was from 4 o'clock and slowly increasing during these 6 shots...don't think they had any effect on elevation. Here is the result I got:
The odd numbered shots have short BSL and the even ones have long BSL. This was a small test at long range. But the results seem to clearly point to a relationship between bearing surface length and impact elevation.
I'm still collecting data and these results have caused me to come up with a bunch of new questions to explore. Is this effect seen at shorter distances? What is the maximum variance that effectively has no/minimal elevation dispersion? Do base-to-ogive and BSL have a strong correlation? Strong enough so that the quicker base-to-ogive measure can effectively substitute for a BSL measure? Are some brands of bullets better than others with this measurement? Lots of work to do, but some of my preliminary testing indicate:
1. All brands of bullets show BSL variation. When comparing Sierra, Hornady, and Berger, there were good and bad samples from each company. Each brand showed a standard bell curve pattern of variance. The amount of variance seems to be more about the lot # and type of bullet than which company you select. I'm collecting lots of samples and will put them into a spreadsheet to calculate SDs.
2. Only did a small test (like 100 bullets): measured both BSL and base-to-ogive on the bullets. I did not see as strong of a correlation as I expected. Using only this small test result, I would conclude that you need to measure BSL and base-to-ogive does not capture all the variance in bearing surface.
That's my .02 for now.