Satisficer vs Maximizer

The satisficer/maximizer distinction comes from psychology, but CF endorses satisficing on its own technical grounds. A maximizer treats every detail as relevant: any change to any factor should nudge the overall conclusion, so the goal becomes optimizing all factors into one best score. CF identifies this as the weighted-factor mindset — and as the same error Bayesians make when they update a credence on every scrap of evidence. It breeds perfectionism and chases gains that don’t matter.

A satisficer instead defines what is good enough and accepts any option that passes. CF argues this is correct, not merely pragmatic. Most factors already have plenty of excess capacity: once an elevator lifts well past the family’s weight, extra capacity adds zero value. So most degree-gains are worthless, and effort poured into maximizing them is wasted — better spent on the factor that actually limits success (see focusing on the constraint).

This grounds CF’s yes-or-no evaluation. Mapping data to pass/fail is many-to-one, so conclusions are resilient: small changes within the margin of error leave the verdict unchanged. Maximizing demands a one-to-one mapping where every wiggle forces a new conclusion — fragile, and akin to chaos-theory sensitivity. CF therefore reframes decision-making as binary evaluation against goals: differences matter only when they flip success into failure.