"Operators should consider that many exercises required to be hand flown are well-known and repetiitve and may not measure the true underlying skill level outside of those specific exercises. Rather, operators should use safety system data and industry best practices to identify opportunities to present both expected and unexpected scenarios requiring manual flight that better measure and improve proficiency in FPM skills during MFO." Parenthetical omitted.
No kidding. It's almost as if some maneuvers we regularly validate are actually assessments of control and performance, particularly certain worst-case-scenarios in terms of raw pilot ability, and
not necessarily the most likely thing that is going to happen on a line operation. That's not to say the maneuver isn't worth validating—far from it—but it's insufficient to simply tick the box.
Anyway, proper proficiency at all levels of automation—from none whatsoever, to a flight director, to an autopilot, to an FMC telling an autopilot where to take the jet—should be the objective, as opposed to the "always use highest level of automation" that seems to have been the zeitgeist. Your program should give pilots opportunities to exercise all of those levels, be proficient in all of them, give them scenarios in which selecting a different level (sometimes higher, sometimes lower, sometimes none!) is a good outcome. And policy should empower pilots to do the same.
Anyway: