“I checked OAL when I ran the batch, and that's not the problem; instead, I have discovered that diameter of the case just below the shoulder is 0.003 too big; this is consistent across most of my cases”
I have a leaver policy, I purchase a die, then! I apply the leaver policy, I leaver the way I founder, others grind, polish, lap and hone etc., bad habits, then there is the grinding the bottom of the die and if that dies not ‘get it’ they go for the shell holder and grind on it for a while, no one is accountable, RCBS should issue a ‘void warranty’ stamp for those that bubba-rise dies, back to no one is accountable.
Let us assume the die had not been screwed up (rendered scrap), let us assume the die is capable of returning a case to minimum length/full length sized (not bumped). Let us assume the case does not have more resistance to sizing than the RCBS press and die can overcome, that leaves us with determining if the case was returned to minimum length/full length sized before the ram was lowered. Before a reloader lowers the ram to remove a case it is possible to determine if the case whipped the press or if the press won and whipped the case. If the case won the case did not get stuffed into the die, the amount of the case that did not get stuffed into the die will be holding the die off the shell holder, the gap between the shell holder and bottom of the die can be measured in thousandths with a feeler gage.
The feeler gage can also be used in increase the ability of the die and press to overcome case resistance to sizing and increase the length of the case between the shoulder/datum and case head.