Failure Tolerance

principle

In high-variance work, ordinary failure is evidence about an attempt, not a verdict on the person making it. Failure tolerance means preserving the confidence to keep participating while still learning from what went wrong.

A good venture capitalist can watch eight startups fail without becoming less confident. That sounds irrational—until you remember that venture investing is built around many failures and a few disproportionate successes.

E1

Separate the outcome from the ability to try again

In a high-variance domain, any single result contains too much chance to serve as a reliable judgment of your ability. Treating every loss as personal invalidation lets noisy outcomes control whether you make the next attempt. The trained response is not blind optimism; it is refusing to let the expected losing field disappear behind the visible winners. This also guards against being fooled-by-randomness in the opposite direction: a bad streak can be as misleading about underlying skill as a lucky one.

E1

Tolerance must not become permission

Failure tolerance breaks when it protects a repeated method rather than the capacity to keep trying. If each miss is waved away as inevitable, you invite normalization-of-deviance: evidence that should change the process instead becomes part of the scenery. Preserve confidence, but keep outcomes available for diagnosis.

Keep two scores

For your next uncertain project, record the outcome and the next attempted change separately. After a rejection, write one sentence about what you will alter, then make the next attempt before using the result to revise your opinion of your overall ability.

Episodes that teach this