Accuracy At Risk
DANGER Warning
I assume everyone knows that when you switch from .38 Special to .357 Mag rounds (or .44 Special to .44 Mag, for that matter) that IF you feel any chambering resistance you should not fire the round. Choking the case mouth with lead will raise peak pressure of the magnum load by making the bullet harder to release. The result can be catastrophic cylinder failure.
Hodgdon data
The Hodgdon data for .38/.357 for 231/HP38 are clearly reversed. A look at their target loads in 700X and PB show the proper order of things; bigger charges in the bigger case. It's got more room under the bullet, after all. The .38 Special should start with 3.0 grains, which is about equal to the old standby 2.7 grains of Bullseye commonly used for wadcutters in match shooting. 3.5 grains should be maximum for a hollow base wadcutter in .38 Special. If you get too much powder under these soft swaged bullets, the skirts tend to be blown open by the pressure at the muzzle, spoiling accuracy. 3.5 to 4.0 is about right for the .357 case.
The above is verified by old Winchester load data for .38 Special that is 2.9 to 3.3 grains of 231 with the HBWC, and 3.0 to 3.4 grains with the BBWC. I've e-mailed Hodgdon about this error, and if they have any explanation that doesn't involve a flight of fancy, I'll post it here.
Leading
I always ran .357 cases and 10% more powder when shooting revolver bullseye matches back in the 80's, and had excellent results.
The issue of cylinder lead is individual to the gun. If you look through catalogs by reamer makers, you still find revolver lead removing reamers. These are special reamers used to carve chamber lead build-up out of revolvers, and were a standard armorer's tool when revolvers were more common as match guns. They have to be used carefully to avoid marring the chambers and safer cleaning methods exist now, but the fact these reamers exist at all tells you this was a common a issue for revolvers.
The cause of the leading has mainly to do with the chamber throat and bullet diameter. Most match accurized revolvers use maximum diameter chambers, which helps accuracy by letting the bullet size down entering the bore rather than bumping up in it. But any time the space around a lead bullet is bigger than the bullet, whether in cylinder or bore, even just momentarily while the bullet bumps up, leading is caused by gas blow-by. It just makes a good deal more trouble for accuracy in the bore than it does in the chambers.
The hollow base skirt is one strategy for getting a better seal in an overly wide space, but I find a solid wadcutter that is oversize does better. With lead bullets and light loads, you can safely shoot quite a bit oversized, and sizing bullets to equal the chamber throat diameter often works best to minimize leading. Especially when the case is shorter than the chamber. This assumes you have maximum size chambers.
The best strategy, IMHO, is finding a bullet diameter that produces peak accuracy in your particular gun. That may or may not mean a little periodic lead removal, but cleaning bother is secondary to accuracy for me. Fortunately, loads that shoot best tend to lead least. For my Smith K-frame target revolver with its reamed chambers, the Lee tumble lube bullets shoot best. The ones from my 6-cavity mold average about .3595" with my alloy (an approximation of Lyman #2), and the gun sandbags groups with them that average half the size of my swaged BB WC loads or even the best match grade commercial HB WC loads. No leading worth mentioning occurs.
It turns out, BTW, from talking with others, not to be uncommon for revolvers with reamed cylinders to prefer bullets about 0.002" over groove diameter. I suspect that just gets them closer to throat diameter in many instances, but haven't verified that by measuring. It's just food for thought.