“Declaring that you know the answer to a question is an invitation to be wrong. Saying ‘I think’ or ‘I believe’ opens the door to being right.” — Annie Duke
A bet is any decision made under uncertainty where you commit resources — time, money, energy, reputation — based on your beliefs about what is likely to happen. By this definition, virtually every meaningful decision you make is a bet.
When you take a job offer, you’re betting that the role will be rewarding. When you recommend a restaurant, you’re betting your credibility that the other person will enjoy it. When you argue a position in a meeting, you’re betting your reputation on being right. These bets are constant and unavoidable. The question is whether you make them consciously or carelessly.
Most people make them carelessly — because they don’t recognize them as bets at all.
Annie Duke describes a powerful mental trick: whenever you form a belief or make a claim, imagine someone immediately saying “Wanna bet?”
This simple question transforms how carefully you think. If someone offers to stake real money on whether your belief is correct, you suddenly:
The act of committing real resources focuses the mind. Gamblers call this “putting skin in the game.” It’s why predictions made publicly with stakes attached are dramatically more accurate than casual opinions.
Duke argues that all beliefs — not just formal decisions — function as bets. When you hold a belief, you’re implicitly wagering that your mental model of the world is accurate. You’re betting that acting on that belief will serve you better than revising it.
The problem is that most beliefs don’t feel like bets. They feel like facts. And once a belief feels like a fact, we stop questioning it.
The process works like this:
This is why confirmation bias is so powerful. Once we believe something, we seek evidence that confirms it and discount evidence that challenges it. The original belief — which may have been formed hastily or incorrectly — is now armor-plated.
Treating beliefs explicitly as bets breaks this cycle. It keeps them provisional, revisable, and honest.
The goal of thinking in bets is not to become uncertain about everything — it’s to become accurately uncertain. This means expressing beliefs with the degree of confidence they actually deserve, no more and no less.
Most people are systematically overconfident. Research consistently shows that when people say they’re “90% sure” of something, they’re right only about 70% of the time. This miscalibration is everywhere: in weather forecasts, medical diagnoses, financial projections, and everyday conversation.
Improving calibration takes practice. Concrete steps:
One of the most underrated skills in decision-making is the willingness to say “I’m not sure.” In most organizational and social settings, uncertainty is treated as weakness. Leaders are expected to project confidence. Experts are expected to have answers. Saying “I don’t know” is read as incompetence.
But this norm is deeply counterproductive. Leaders who admit uncertainty make better-calibrated decisions. Experts who acknowledge the limits of their knowledge are more trustworthy, not less. Teams that tolerate “I’m not sure” as a legitimate answer seek more information and deliberate more carefully before committing.
“We can train ourselves to express our beliefs as estimates rather than certainties. ‘I think,’ ‘I believe,’ ‘I’m not sure but’ — these phrases are accuracy flags, not weakness flags.” — Annie Duke
Small linguistic shifts signal and reinforce honest calibration:
These aren’t hedges that weaken your position. They’re signals that your beliefs are based on evidence and reasoning, not bluster — and that you’re open to updating if better evidence arrives.
The final insight of this chapter is both liberating and sobering: you never make decisions in the present. You make decisions about the future — and the future is always uncertain.
When you decide to accept a job offer, you’re not betting on what the job is right now. You’re betting on what your experience of the job will be over the next years. When you choose a health treatment, you’re betting on how your body will respond to something that hasn’t happened yet. The present is where you act; the future is where you’ll find out whether you were right.
This is why demanding certainty before deciding is always a mistake. Certainty is never available. The choice is always between making an explicit, calibrated bet or making an implicit, uncalibrated one. The second doesn’t feel like a bet — but it is, and it’s usually worse.