Conduct a randomness audit of outcomes you’re explaining

Before attributing an outcome to skill, strategy, or character, estimate how much of it could be chance.

Why it works

Narrative construction attributes outcomes to agents: someone’s talent, foresight, or decision-making. This attribution is psychologically satisfying because it creates a learnable lesson ("do what they did"). But many outcomes, especially in complex, noisy environments, depend heavily on factors outside anyone’s control. Explicitly estimating the randomness component prevents the narrative from crowding it out entirely.

How to do it

1. After explaining an outcome, estimate: "What percentage of this outcome could plausibly be attributed to luck, timing, or context outside anyone’s control?"
2. List two or three specific random or contingent factors that were present.
3. Revise the lesson accordingly: what is actually learnable vs what was fortunate circumstance?

Evidence

Attribution theory and outcome bias research show that people systematically over-attribute good outcomes to skill and bad outcomes to situational factors (or vice versa depending on identity involvement). Taleb’s contribution is emphasizing the scale of randomness in complex systems. (observational)

Estimating the "randomness percentage" is necessarily rough; the tool’s value is directional — introducing the question at all — rather than precise.

Common mistake

Applying the randomness audit only to other people’s successes ("they got lucky") and not to your own.