Thinking in Bets

The Trap: Resulting

Duke introduces a concept she calls "resulting"—our tendency to evaluate a decision based solely on its outcome. Russell Wilson's infamous goal-line interception in Super Bowl XLIX became a case study in bad decision-making. But was it?

The Seahawks had one timeout, 26 seconds, and the ball on the 1-yard line. A pass play would stop the clock if incomplete, preserving time for another play. A run that failed would cost a timeout. New England's defensive alignment suggested a pass was the right call. Malcolm Butler made an improbable interception—a play he had never made before.

[!INSIGHT] Resulting is the cognitive error of conflating decision quality with outcome quality. A good decision can have a bad outcome, and a bad decision can have a good outcome—because luck and incomplete information are always in play.

Duke's research reveals that resulting isn't just a sports problem. Investment committees routinely fire managers after down years, regardless of whether their strategy was sound. CEOs lose jobs after quarterly misses caused by macroeconomic shifts entirely outside their control. We build our entire accountability systems around outcomes, not decisions.

The 10,000-Hand Sample Size

In poker, a single hand tells you almost nothing about a player's skill. You can make every wrong move and still win against a superior player. Duke notes that professional poker players need approximately 10,000 hands before they can confidently assess whether they're playing well or running hot.

This has profound implications for life decisions. How many job changes constitute a pattern? How many relationships before you can evaluate your romantic decision-making? How many product launches before a CEO's strategy can be fairly judged?

“*"Life is not like chess. In chess, you can see all the pieces, and if you lose, it's because you made a mistake. In poker and in life, you can make the right decision and still lose. The key is to stop punishing yourself for bad outcomes that resulted from good decisions.”

Thinking in Bets: Rewiring for Probability

The core insight of Duke's framework is deceptively simple: every decision is a bet. Not a gamble—a bet. A bet is a choice made under uncertainty, where you're wagering resources on a particular future outcome based on your current beliefs.

When you choose a career path, you're betting that this field will remain viable, that you'll be competitive in it, and that it will bring satisfaction. When you marry someone, you're betting on decades of compatibility through circumstances neither of you can predict. When a CEO acquires a company, they're betting on synergies, market conditions, and integration success.

The 70% Confidence Problem

Duke cites research showing that when people say they're "100% certain" about something, they're right only about 80% of the time. When they say they're 70% confident, accuracy drops to around 60%. We are systematically overconfident in our predictions.

[!INSIGHT] Professional poker players rarely express certainty. They speak in ranges: "I'm about 65% confident this opponent is bluffing." This probabilistic thinking protects against both overconfidence and analysis paralysis.

The solution isn't to eliminate uncertainty—it's to embrace it explicitly. Duke recommends practicing "confidence calibration" by regularly writing down predictions with probability estimates, then tracking accuracy over time. The goal is to align your expressed confidence with actual accuracy.

The Buddy System: Surviving Truth-Seeking

Perhaps the most practical chapter in Duke's book addresses how to create environments where honest decision evaluation is possible. In poker, players form "groups"—small communities that review hands together, debate decisions, and hold each other accountable to intellectual honesty.

These groups work because they establish three critical norms:

1. Agreement to debate decisions, not judge character — The focus stays on the play, not the player
2. Celebration of accuracy, not confirmation — Members are rewarded for correct analysis even when it contradicts the group
3. Permission to express uncertainty — Saying "I don't know" is treated as wisdom, not weakness

[!NOTE] Duke's research found that decision-making groups without these norms consistently performed worse than individuals deciding alone. Poorly structured groups amplify biases rather than correcting them.

Corporate environments often fail at this. Performance reviews judge outcomes, not decision processes. Political dynamics reward certainty over accuracy. Admitting uncertainty is perceived as weakness. Duke argues that organizations serious about better decisions must deliberately construct psychological safety around probabilistic thinking.

Implications for Leadership and Life

The betting framework reshapes how we approach regret, hindsight bias, and even happiness. Duke points out that regret is often misplaced—we regret outcomes, not decisions. But you can only control the decision.

Leaders who internalize this thinking become more willing to take calculated risks, more forgiving of their own failures, and better at extracting lessons from both successes and failures. They stop asking "Why did this fail?" and start asking "Given what I knew then, was this the right call?"

The Hindsight Trap

Hindsight bias is particularly destructive because it convinces us that past events were more predictable than they actually were. After the 2008 financial crisis, countless commentators claimed they "saw it coming." But Duke notes that if you examine predictions made in 2006-2007, virtually no one accurately predicted the crisis's timing, scope, and triggers.

“*"Hindsight bias is the enemy of learning from the past. It makes us think we understood what was happening when we didn't, which prevents us from understanding why things actually unfolded as they did.”

This has critical implications for how organizations conduct post-mortems. If you assume the outcome was inevitable, you'll learn nothing. If you reconstruct the decision environment as it actually existed—with all its uncertainty—you might learn something valuable.

The most transformative idea in Thinking in Bets is that you can make all the right decisions and still lose—just as you can make terrible decisions and win. The goal is not to eliminate uncertainty or guarantee outcomes. The goal is to maximize the probability of good outcomes over thousands of decisions across a lifetime. This requires separating your ego from results, building systems for honest feedback, and learning to think in probabilities rather than certainties. As Duke puts it: life is poker, not chess. The question is whether you'll learn to play accordingly.

Sources: Duke, Annie. Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts. Portfolio, 2018; Tetlock, Philip E., and Dan Gardner. Superforecasting: The Art and Science of Prediction. Crown, 2015; Kahneman, Daniel. Thinking, Fast and Slow. Farrar, Straus and Giroux, 2011.