RougeLeader,
Brass is more plastic (malleable) and less elastic than steel. When a charge gets high enough, cases can resist ejection because the steel has expanded beyond the yield point of the brass but not beyond its own yield point, so it has then returned to its own original shape afterward, thereby clamping the expanded, less elastic brass in place. Many of the old-timers depended on this for revolver load development (read anything by Elmer Keith on load development). They would load up to the point the cases began to resist ejection, then back the charge weight down 5%.
The above is unique to revolvers because mushroomed and flattened primers and
other pressure signs often fail to appear in revolvers before sticky ejection does. the steel is thinner at some parts of the chamber than others, so resistance to ejection is treated as a revolver pressure sign and therefore an indication you should back down. If you do not, but instead keep increasing the charge weight, the steel will give way first where it is thinnest, as FITASC's photos showed.
The reasons old load data are suspect are several. First, many old manual authors did not use pressure testing at all, instead of looking for pressure signs in a particular production firearm whose chamber(s) might be looser than others, and therefore needing more powder to reach a pressure sign. Moreover, pressure signs on a case usually tell you more about what the case can tolerate than what the gun can tolerate, the stick ejection being an exception. Second, manufacturers of and formulations for powders can and do change over time. An attempt is made to keep a new powder compatible with the original, but that may not always be exact. Almost all primer formulations have changed in some measure since 1989, and primers can have some effect on pressure. Many cartridge cases have at least slightly different capacities than they did even a decade ago due to outsourcing and changes in plants and tooling, forcing loads to be worked up again. Pressure measuring has advanced immensely since old data was developed, and now, instead of just a peak value, the whole pressure curve is revealed and in some instances, this has resulted in anomalies being identified that has changed load data. Probably the most notable was the warning from Alliant issued July 25, 2008, to stop loading Blue Dot powder in .357 Magnum with 125 grain bullets and to stop loading it in the .41 Remington Magnum at all. Old load data often has these loads Alliant says can be hazardous.
You are right about the time frame of firing events. Years ago, H.P. White Laboratories found they could not detect movement in a bullet before pressure reached 10,000 psi, despite bullet pull being only tens of pounds. That high number was due to the dynamics of the event. In the quarter millisecond or so it takes to go from primer pressure to 10,000 psi, the inertia of the bullet plays a significant roll in limiting its acceleration. Yet, if you fire just a primer with no powder, you can unseat a bullet into the throat of the gun. It's just that it doesn't happen as quickly. Given enough time, everything equilibrates toward steady-state behavior.