However, in practice I use the very same cast lead bullets in both, seated the same distance. So in my ammo, there is a 0.10 difference in the OAL.
Of course. And you could get a 0.10" difference in OAL in just about any revolver cartridge you wanted without switching to a different caliber just by playing with the seating depth.
I guess what I'm saying is that:
1. The quintessential accuracy round in the .38spl is the flush seated wadcutter which has a longer jump to the throat/rifling than a round loaded with a conventional bullet. The jump doesn't seem to hurt accuracy.
2. The two rounds, can differ in terms of OAL, but in practice, they are often loaded to virtually identical OAL. In my experience, that is commonly the case with factory ammunition. If one wants to assess the relative accuracy of the two rounds and try to correlate it to bullet jump, one needs to insure that there really is a significant bullet jump difference in the two particular loadings being checked. Since that's rarely even noticed/mentioned with regard to this topic, the likelihood is that any anecdotal evidence relating to .38spl having poorer accuracy than .357Mag in the same revolver is probably not related to bullet jump. That's because the rounds likely weren't even checked for bullet jump difference since everyone "knows" that the .38spl is a shorter round.
I'll throw in a third that I didn't mention earlier. While it's common to find reloaders talking about tailoring the seating depth of rifle rounds for maximum accuracy, finding a similar discussion relating to pistol rounds is not that easy. I think the primary reason is that there are too many other variables that affect revolver accuracy more significantly to make seating depth experiments worthwhile.
In short, any way you cut it, I think it's going to be difficult to make a case for bullet jump being a significant contributor to revolver accuracy in the vast majority of cases.
