Test success principles for survivorship bias

When This Fires

• After extracting a "lesson" or "principle" from a successful outcome
• When building personal operating rules based on what has worked in the past
• When tempted to generalize from a single success to a universal approach
• Complements Three-pass pattern spotting (counterexample checking for patterns) with the success-specific survivorship filter

Common Failure Mode

Over-generalizing from one success: "My presentation went great because I didn't prepare slides — so I should never use slides." The presentation succeeded because the topic was conversational, the audience was informal, and you knew the material deeply. Under different conditions (technical audience, formal setting, unfamiliar material), no slides would fail. The principle ("no slides = great presentation") was extracted from one data point without testing robustness.

The Protocol

(1) When you extract a principle from a success, write it as a testable claim: "Starting projects on Mondays leads to better outcomes." (2) Apply the counterfactual: "If I started a project on Monday but the project scope was different / the team was different / the deadline was different, would I still expect a better outcome?" (3) If yes across multiple condition variations → the principle is robust. It's a genuine causal mechanism worth preserving. (4) If no / uncertain → the principle is context-dependent. It worked this time because of specific conditions, not because of the principle itself. Note the conditions under which it worked rather than extracting a universal rule. (5) For every success principle you keep, also identify the conditions under which it would fail. This prevents blind application to contexts where the principle doesn't hold.