Cognitive Biases and Fixes

Favoring evidence that supports your belief.

Fix: actively seek disconfirming evidence.

Judging likelihood by ease of recall.

Fix: check data/base rates when possible.

First number seen pulls your estimate.

Fix: make an independent estimate first.

Continuing because you already invested.

Fix: decide using future costs/benefits only.

Learning only from winners.

Fix: include failures and missing data in the analysis.

Losses feel larger than gains.

Fix: compare options using consistent, pre-set criteria.

Choices change when phrased differently.

Fix: reframe in neutral terms and compare.

Overestimating accuracy.

Fix: use ranges and track calibration over time.

Low skill can inflate confidence.

Fix: seek feedback and objective benchmarks.

After the fact it feels inevitable.

Fix: write predictions before outcomes.

Underestimating time/cost.

Fix: use reference classes and add buffer.

Preferring current state.

Fix: ask "Would I choose this if starting fresh?"

Overvaluing what you own.

Fix: price it as if you did not own it.

One good trait colors all judgments.

Fix: rate traits independently.

Over-assigning behavior to character.

Fix: consider situational constraints first.

Success = me; failure = circumstance.

Fix: run blameless postmortems with data.

Bad weighs more than good.

Fix: track positives explicitly to balance perception.

Recent events dominate.

Fix: look at longer time windows.

Overvaluing now vs later.

Fix: pre-commit and reduce future friction.

Overweighting experts/leaders.

Fix: ask for reasoning and check independent sources.

Following the crowd.

Fix: separate popularity from correctness; test directly.

Favoring your group.

Fix: use standardized criteria and diverse reviewers.

Judging decisions by results alone.

Fix: evaluate decision quality given info at the time.

Overestimating influence.

Fix: separate controllables from luck; focus on process.

Familiar feels true/better.

Fix: deliberately compare alternatives on merits.