Peter Sellis (Snap/Discord) frames PM evaluation as a fit function between an individual and their specific context — role, team, product stage — rather than an intrinsic, portable trait of the person alone. A PM who is struggling may not be a bad PM; they may be a bad match for this particular place. See Power-Law Distribution of PM Quality for why PM quality is so unevenly distributed in the first place, and Keeper Test for a contrasting person-only evaluation lens. Before concluding someone is the problem, try changing the team, product surface, or scope first.