This makes a lot of sense to me. I think this explains exactly what I am seeing happen. But wouldn't a donut be noticeable on close inspection? Wouldn't an expander ball push irregularities to the outside of the neck?You implied that you could seat the bullet in with increments, and these measure low in runout. That you only get a step change in runout with a final seating increment. The seating stem is not doing that.
Donut area comes into play -when you size that donut into play.
I imagine one of the bullet bands is in interference with donut brass at your particular seating depth.
The stick & click part is the stem popping off the ogive (release of wedging) due to mismatch angles and high seating force.
Seating force (relative feel on press handle) seems consistent from case to case and definitely less than other cartridges that I load for that aren't freshly annealed.
I did experiment with an empty case and putting a little imperial on the drive bands of the bullet. This nearly eliminated the induced runout from the seating.