3 ideas
Most decision quality is determined before any options are weighed, by the mental frame that defines what the problem is, which reference points apply, and what counts as a possible solution. Because frames operate invisibly, the discipline is to surface and deliberately reframe a decision — testing alternative framings — rather than accepting the first frame that comes to mind.
People are systematically overconfident in their estimates, producing 90% confidence intervals that capture the true value far less than 90% of the time. The corrective mechanism is to widen your ranges until they feel uncomfortably broad, then test calibration against known answers to recalibrate how much you actually know.
Good outcomes can follow bad decisions and bad outcomes can follow good ones, so judging decision quality by results alone teaches the wrong lessons. Effective learning requires evaluating the decision process at the time it was made — the information used, alternatives considered, and reasoning applied — separately from the luck-influenced outcome.