State a prediction before looking at the outcome

Commit to a prediction before the result is known to prevent the narrative from rewriting your memory of your forecast.

Why it works

Hindsight bias — the sense that you knew it all along — is driven partly by narrative construction: once an outcome is known, memory of the pre-outcome state is revised to match it. Writing down a prediction before the result is known creates an external record that resists this revision. It makes the gap between pre-outcome uncertainty and post-outcome apparent inevitability visible and measurable.

How to do it

1. Before any event whose outcome matters to your learning, write your prediction with an explicit probability.
2. Record the date.
3. After the outcome is known, compare the prediction to what happened — without revising your memory of your thinking.

Evidence

Hindsight bias is among the most replicated findings in cognitive psychology. Fischhoff’s original research showed that people consistently believe they would have known the outcome before it happened. Written pre-outcome predictions provide the only reliable corrective. (observational)

Sources

• Fischhoff (1975), "Hindsight ≠ Foresight," Journal of Experimental Psychology

Common mistake

Recording the prediction ambiguously so it can be read as correct regardless of outcome — a prediction must specify a direction and a probability to be checkable.