A couple logs back concluded with:

Next up: is Z hysteresis small enough to run paths with varying Z vs a succession of descending waterline paths?

(i.e. parallel finishing)

Yes.

The second pocket has other issues, but this isn't about that.