Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts (Book Digest)

Thinking in Bets by Annie Duke presents a framework for making better decisions in an uncertain world by recognizing that nearly every decision is a bet on the future. 

The book's key insight is that life is more like poker than chess: We make decisions with incomplete information and luck plays a significant role. The quality of a decision can't be judged solely by its outcome. 

Success comes from accurately representing what you know and don't know, creating systems that help you think more clearly despite natural human biases, and reacting rationally to uncertainty.

Life Is Poker, Not Chess

Duke challenges the common assumption that good decisions always lead to good outcomes and bad decisions always lead to bad outcomes. This false equivalence, which she calls "resulting," is one of the most damaging habits in decision-making. 

Good decisions sometimes lead to bad outcomes, such as when a football coach makes a play call that is right on the odds but leads to a bad result due to luck. 

Humans are apt to think that good decisions always lead to good outcomes because we have the wrong model for decision making. We think that life is a game like chess, where you have complete information and, if you're better than your opponent, you'll almost certainly win. 

But life isn't like chess; it’s like poker. In poker: 

• You're dealing with incomplete information

• Luck plays a significant role in outcomes

• Good decisions can lead to bad outcomes, and vice versa

• You can't simply work backward from results to determine decision quality

Using the wrong model for decision making causes us to miss opportunities to improve. We take credit for successes that were mostly luck and blame failures on factors outside our control, even when our decision-making was flawed. 

The first step to better decision making is to recognize that "I'm not sure" is often the most accurate answer. When we express our beliefs as probabilities rather than certainties, we open ourselves to new information and give ourselves the chance to react productively to an uncertain world.

Truth Seeking in Groups

We can't overcome cognitive biases simply by knowing about them. Duke found that even smart people—including herself—are susceptible to motivated reasoning, confirmation bias, and self-serving interpretations of results. 

The solution isn't just individual discipline; it's creating a group environment that actively promotes truth-seeking.

Duke learned this firsthand in her poker career. She joined a group of world-class poker players who became her "learning pod." This group followed a set of rules: When discussing hands, they would focus on decision quality rather than results, challenge each other's thinking, and reward intellectual honesty.

Not all groups naturally promote better thinking—in fact, most make it worse. Groups can become echo chambers that reinforce existing beliefs rather than challenging them. More than 80% of sociologists self-identified as left of center and this political homogeneity has led to confirmation bias in research topics, peer review processes, and conclusions.

Effective decision groups follow principles captured by the acronym CUDOS, drawn from sociologist Robert Merton's work:

Communism (of information): Share all relevant details with the group, especially uncomfortable ones. Poker players describing hands work from a template that prevents them from cherry-picking information that supports their preferred narrative. 

Universalism: Apply the same standards of truth to every idea, no matter where it comes from or how you feel about it. Good ideas can come from unexpected sources, and bad ideas can come from people we respect.

Disinterestedness: Recognize and counteract conflicts of interest. Our brains have built-in conflicts—we want to confirm existing beliefs, avoid admitting errors, and take credit for good outcomes. One powerful technique is outcome-blind analysis: Companies can evaluate sales strategies before learning whether deals closed and hiring decisions before seeing how employees perform.

Organized Skepticism: Approach beliefs with an attitude of "I'm not sure" rather than "I'm certain." When groups embrace uncertainty, dissent becomes easier. Instead of confrontational declarations like "You're wrong," conversations start with "I'm not sure about that." This creates psychological safety for questioning ideas without attacking people.

Countering Emotional Bias

Even with perfect processes, we still face a fundamental problem: The deliberative parts of our brain are constantly overtaxed. We can't simply decide to think more carefully about everything. 

Instead, we need strategies that work within our human constraints. Here are three:

Meeting your future self: One powerful intervention is making future consequences feel more immediate. Financial companies now offer tools that show clients age-progressed images of themselves during retirement planning. By making future-you feel like a real person who exists now, it triggers the regret from making a bad decision before you make the decision, not after. 

Avoiding ticker-watching: Monitoring short-term results can lead to bad decisions. Berkshire Hathaway stock showed tremendous long-term gains. But if you zoomed in on any given month, you'd see alarming ups and downs. Short-term fluctuations mean almost nothing for long-term performance, yet we often evaluate our lives this way and experience strong emotions as a result of these swings.

In poker, these swings can lead to emotional hijacking called going on "tilt." The emotional center of the brain shuts down the thinking part of the brain, leading to catastrophic decision-making. 

Duke learned to think of her entire career as "one long poker game" where individual hands didn't matter. Similarly, we should evaluate our lives over longer timeframes rather than obsessing over short-term fluctuations.

Ulysses contracts: These are precommitments that prevent future-you from making predictable mistakes (named after Ulysses tying himself to the mast to resist the Sirens' call). Examples include automatically allocating money to retirement before you can spend it or removing junk food from your house so midnight-you can't mindlessly eat it.

Scenario Planning

Traditional planning assumes we can predict the future or that current conditions will persist. Both assumptions are dangerous given pervasive uncertainty.

Duke argues for thinking probabilistically about multiple possible futures using scenario planning. The basic approach involves three steps:

1. Identify possible futures: Imagine the range of outcomes that could occur, not just the most likely one

2. Assign probabilities: Make your best guess at the likelihood of each outcome

3. Plan for each: Consider how you'd respond in each case

Duke helped a charitable organization use scenario planning to improve its budgeting and grant process. Previously, they spent more time on large grant applications. This ignored the fact that a large grant application might have a very low probability of success, making it worth less effort than its large size made it seem. 

They shifted to scenario planning, calculating the expected values for each grant by multiplying the amount of the grant by the probability of winning. A $100,000 grant with a 25% success rate has an expected value of $25,000. A $40,000 grant with a 70% success rate has an expected value of $28,000, making it more valuable despite the smaller size of the grant. 

This simple change produced multiple benefits:

• Better resource allocation based on expected value rather than grant size

• Motivation to improve probability estimates by gathering more information

• Feedback loops to calibrate their predictions over time

• More realistic budgeting and reduced financial volatility

Backcasting: When you've identified a positive goal, imagine that you’ve achieved it and work backwards by asking “What steps led to this outcome?" This forces you to think through the full decision tree that leads to success rather than just the immediate next step. 

Premortems: You can flip things around: Imagine you’ve failed and ask why. This is beneficial because it can be hard to dissent or express negative views. But if you frame the exercise as "Okay, we failed—why did we fail?" everyone feels free to identify potential problems so you can plan for them.

Conclusion

Thinking in Bets offers a practical framework for making better decisions in an uncertain world. The key shift is recognizing that life is poker rather than chess because it’s subject to limited information and randomness.

Duke's most important insight is that "it's all one long game." Any single outcome tells us very little. What matters is improving our decision-making process over time, being honest about what we do and don't know, and creating systems that help us think more clearly despite the cognitive limitations and biases we all share.

The promise isn't perfection; we'll still make mistakes and experience bad luck. But by thinking in bets, we can make decisions we're less likely to regret and learn more effectively from both success and failure.

- - -

If you liked this article, check out our other book digests in our series Required Reading for Salespeople. You can also check out the P.S.I. Selling Content Page for more insights on sales communication, strategy, and leadership.

Want to build a sales process that proves value and a team that can execute? Get in touch.

For more about the author, check out Mike's bio.