RBHarter, that's the stuff I'm talking about. What's intuitive doesn't always occur because there are sometimes forces at work that we didn't realize, so we scratch out heads and do some more shooting and ponder "Now, what caused it to do that?".
Thinking in terms of second derivatives instead of fX=whatever helps me imagine pressure rise/volume and how it relates to bullet acceleration and deformation. It's not just "curve" we should consider, but the area under that curve that does the work, while the shape of the curve tells how and when the pressure is applied.
Here's another thing I did which has influenced my thinking: I have taken some bullets cast from the same pot of alloy and done two different things with them, then shot them in the same rifle with very different results. First group was gas-checked, lubed, sized to throat entrance diameter, and loads worked up to find best powder and charge at highest velocity that would hold 1 MOA at 100 yards. Other group was run through a push-through sizer right after water-dropping and wet-patched with rag paper. The patched bullets were lubed with a little soft beeswax/Vaseline mixture. Both bullets fired out of a .308 Winchester bolt rifle, ten-twist. The best I could manage with the checked/lubed bullets after a lot of work was 2300 fps. I pushed the paper-patched bullets with a variety of powder burn rates as fast as pressure for the cartridge would allow (based on extractor scuffs, primer cratering, and case head expansion), usually easily achieving MOA accuracy and often better. I topped out at over 2700 fps with a 172-grain bullet if that is any indication of how much peak pressure my loads were making. Ok, I repeat: SAME BULLETS FROM SAME PILE, SAME RIFLE. Minus gas check and plus a couple wraps of paper, the difference was night and day. My conclusions? First, the concept that the forces of high RPM acting on casting defects because we can't cast a bullet that's up to the task is complete and utter hogwash. Second, that a given alloy can withstand a LOT more pressure and steepness of launch pressure curve than I keep reading about, provided that land engagement isn't a limiting factor, and a few other things. When the jacket is removed, the pressure curve becomes super-critical. I think cast bullet accuracy limits have more to do with bullet/land-groove engagement under high acceleration than is commonly thought.
You mentioned slower powders doing weird things in auto-loaders, I have too. Not only high port pressure, but erratic port pressure throwing the barrel harmonics all over the map. The M1A is a good teacher of many things.
