This article distills the most trading-relevant ideas from Thinking in Bets by Annie Duke. The book’s core value is its framework for separating decision quality from outcome quality in environments where luck, hidden information, and incomplete feedback are always present. For trading, that makes it a decision-process book more than a psychology book.
Poker, Not Chess (Duke)
Duke’s central analogy is that real decisions are more like poker than chess. In chess, all information is visible and the better move can often be evaluated directly. In poker, as in trading, important information is hidden and luck affects short-term outcomes.
This matters because traders often judge themselves as if they are playing chess:
- one losing trade means the idea was bad
- one winning trade means the process was right
- uncertainty is treated like a flaw instead of a condition
Trading is closer to poker. Good decisions can lose. Bad decisions can win.
Resulting (Duke)
The strongest concept from the book is resulting: judging the quality of a decision by the quality of its outcome.
| Bad evaluation habit | Better evaluation habit |
|---|---|
| Winner = good decision | Good decision = sound process under uncertainty |
| Loser = bad decision | Bad outcome can still come from a good process |
| Review P&L first | Review assumptions, probabilities, and alternatives first |
| Learn from what happened | Learn from what was knowable at the time |
For traders, resulting is destructive because it trains confidence and doubt off short-term variance rather than process quality.
IMPORTANT
A good process does not guarantee a good outcome on one trial. It improves the odds across many trials.
Beliefs as Bets (Duke)
Duke’s practical reframe is to treat beliefs as bets. A belief is not “true” or “false” in the emotional sense. It is a claim held with some level of confidence under uncertainty.
This changes trader thinking in useful ways:
- conviction becomes probabilistic rather than absolute
- disagreement becomes information rather than identity threat
- certainty gets replaced by ranges and scenarios
- confidence can be expressed as degree, not certainty
This is especially useful in discretionary trading where overstatement of conviction often drives over-sizing and stubbornness.
Wanna Bet? (Duke)
Duke’s “wanna bet?” test is a calibration tool. If you had to put money behind a belief, how confident would you really be? The question exposes whether a trader is speaking with justified conviction or casual overconfidence.
Applied to trading:
- “I know this will hold” becomes “How much would I actually risk on that?”
- “This has to break out” becomes “What probability would I assign, and what would invalidate it?”
- “I am sure I am right” becomes “What odds would I offer or accept?”
This is a clean way to reduce dogmatism and force more honest sizing.
Luck, Skill, and the Quality of Decisions (Duke)
Trading outcomes contain both skill and luck. Duke’s framework forces a trader to keep those mixed rather than pretend they can always be separated cleanly trade by trade.
Useful implications:
- do not confuse a hot streak with proof of edge expansion
- do not confuse a drawdown with proof of lost ability
- evaluate process using samples, not isolated outcomes
- expect feedback to be noisy, delayed, and emotionally misleading
Truthseeking vs Being Right (Duke)
Duke emphasizes truthseeking over defending identity. This matters because smart people are often better at rationalising than at updating. In trading, intelligence can amplify bias when it is used to defend a position rather than reassess it.
The stronger posture is:
- seek disconfirming evidence
- downgrade certainty quickly when new information appears
- separate self-worth from being right on a trade
- treat revision as strength, not weakness
Open-Mindedness in a High-Variance Domain (Duke)
Probabilistic thinking requires a style of open-mindedness that is more disciplined than vague flexibility. Duke’s version means holding views strongly enough to act, but loosely enough to update.
That balance is important in trading because two failures are common:
- Rigid conviction: refusing to update because changing your mind feels like defeat
- Aimless flexibility: changing views constantly with no probabilistic structure
The correct middle ground is conviction with explicit uncertainty.
Backcasting and Premortem Thinking (Duke)
One of the more practical decision tools associated with Duke’s framework is to imagine that a future decision has already failed and then work backward to identify why. This improves planning because it surfaces hidden assumptions before capital is committed.
For traders, this can be used pre-trade:
- how could this thesis fail even if the setup looks strong?
- what am I likely to ignore because I want the trade?
- what would make me look back and say this was avoidable?
This is a useful counterweight to confirmation bias and narrative intoxication.
What Belongs in a Trading Wiki (Duke)
The durable trading takeaways from Thinking in Bets are:
- separate decision quality from outcome quality
- think in probabilities, not certainties
- treat beliefs as bets with position-sized conviction
- use “wanna bet?” to calibrate confidence honestly
- train truthseeking rather than ego-defence
- evaluate process over distributions, not single trades
Sources
- Annie Duke, Thinking in Bets