TB: How to Decide (Duke)

Core Thesis

Only two things determine how your life turns out: luck and the quality of your decisions. You can only control one of these. Most decision-making failures are failures of process, not intelligence — and the outcome of a decision tells you almost nothing about whether the decision was good.

Key Takeaways

Separating decision quality from outcome quality

  • The quality of the outcome casts a long shadow over our ability to assess the quality of the decision that produced it. A bad outcome from a good decision is just bad luck; a good outcome from a bad decision is just good luck.
  • Resulting (judging decisions by outcomes) is the primary cognitive error in retrospective evaluation.
  • Ask habitually: "If I were wrong, why would that be?" This forces engagement with contrary evidence before it arrives.

When to go fast vs slow

  • When a decision is hard, that's often a signal it's actually easy — the closeness of the options is what's slowing you down. Close options are approximately equivalent; just choose.
  • The happiness test: will this decision's outcome affect your happiness in a month? A year? If no, spend little time on it.
  • "Free roll" decisions: asymmetry between upside and downside makes the answer obvious — go fast.

Decision stacking

  • The lower the cost to quit, the faster you should move. Design reversibility into decisions where possible.
  • Decision stacking: make lower-impact, easier-to-quit decisions that teach you something before the high-impact, one-way decision. Rent before you buy.
  • "Knowing when the hammer is good enough is a meta-skill worth developing."

Reducing uncertainty

  • "He who knows only his own side of the case knows little of that." — Mill.
  • We don't have a common language for talking about decision quality, which is part of why we don't improve at it.

Mental Models

  • Inversion — instead of asking "will this work?", ask "under what conditions would this fail?"
  • Expected Value — Duke's framework formalizes expected value reasoning for everyday decisions

Source note

How to Decide (Duke)

Backlinks