Cover of Thinking in Bets

Thinking in Bets

Annie Duke

6 ideas

  1. Resulting: judging decisions by outcomes

    Resulting is the error of inferring decision quality from how things turned out, treating a bad outcome as proof of a bad decision and a good outcome as proof of a good one. Because outcomes are shaped by both skill and luck, the link between decision and result is loose, and grading by results teaches the wrong lessons from both wins and losses.

  2. Pete Carroll's Super Bowl pass call

    In Super Bowl XLIX, Seattle's coach called a pass on second down at the goal line instead of a run, and it was intercepted, costing the game; commentators called it the worst play call in history. Duke argues the call was defensible given clock management and the low historical interception rate, and the backlash came purely from the unlucky result, not from any flaw in the reasoning.

  3. Every decision is a bet on futures

    Any choice is a wager that one possible future will serve you better than the alternatives, placed with limited information and against an uncertain world. Seeing choices as bets forces you to state what you believe, how confident you are, and what you stand to win or lose, rather than pretending the decision is a certainty.

  4. Say 'wanna bet?' and calibrate confidence

    Asking 'wanna bet?' about your own belief exposes how sure you actually are and shifts you from treating beliefs as true-or-false to holding them with a percentage of confidence. Stating beliefs as probabilities lowers the cost of updating, because revising from 70% to 50% feels like refinement rather than admitting you were wrong.

  5. Truth-seeking group with CUDOS norms

    A small group that agrees to reward accuracy over comfort can catch biases an individual cannot see in themselves. It runs on Merton's norms: communism (share all the data), universalism (judge ideas apart from their source), disinterestedness (guard against conflicts of interest, including hiding outcomes while evaluating decisions), and organized skepticism (dissent is expected and welcomed).

  6. We attribute outcomes to protect our self-image

    People fielding outcomes tend to credit their wins to skill and blame their losses on luck, while doing the reverse for others. This self-serving bias blocks learning, because the only way to improve is to sort honestly which outcomes came from your decisions and which came from chance, whatever the result does to your ego.

Save and mark ideas in the app