How does "Judge decisions by the process, not the result" work?

Outcome bias — judging decision quality by what happened rather than by what was known before the decision — is pervasive and damaging. It causes people to reverse correct processes that got unlucky and to repeat incorrect processes that happened to succeed. Expected value thinking separates these: a decision is good if it correctly used available information to maximize expected payoff, regardless of which outcome the probability dice rolled.

How does "Calculate the expected value of gathering more information" work?

More information is not always worth its cost. The expected value of information (EVOI) is the probability that the new information would change your decision, multiplied by the gain from making the better choice. If the probability you would change your decision is low, or if the decision consequence is small, the information is worth less than the time cost of obtaining it. This prevents the common pathology of endless research to avoid deciding.

How does "Adjust raw expected value for risk aversion on large stakes" work?

Raw expected monetary value ignores diminishing marginal utility: the pain of losing $10,000 is not just ten times the pain of losing $1,000; it can be much larger if the loss would meaningfully damage your situation. Utility theory formalizes this: for large stakes, convert monetary values to utility before computing EV. Practically, this means being willing to pay a premium to reduce variance when the downside would be catastrophic.

How does "Accept positive-EV decisions even when they feel uncomfortable" work?

Loss aversion causes people to reject positive-expected-value gambles because the pain of a likely small loss outweighs the pleasure of a less likely large gain — even when the math clearly favors the bet. The discipline of accepting positive-EV decisions consistently is what produces good outcomes in aggregate over a lifetime of decisions. Individual outcomes are noise; repeated positive-EV plays are the signal.

How does "Look for decisions with asymmetric upside — large potential gain, small defined loss" work?

Traditional EV analysis weights gains and losses symmetrically, but many opportunities have a fundamental asymmetry: the downside is capped (the cost of a book, the time for a cold email, the fee for a course) while the upside is large and uncertain. These are optionality plays — Nassim Taleb’s "antifragile" positions — where the math is favorable without needing to know the exact probability of success.

How does "Keep a decision journal to score your EV estimates" work?

EV thinking is only as good as the quality of the probability and outcome estimates feeding it. Without feedback, estimates drift toward whatever is emotionally convenient. A decision journal creates accountability: you record your estimates before the outcome, then revisit after. This feedback loop is the same mechanism that makes calibration practice effective — seeing systematic errors allows deliberate correction.

decision theory

How do you use expected value thinking to make better decisions under uncertainty?

The math of rational choice under uncertainty, its real limits, and how to use it anyway

Short answer

Expected value thinking multiplies each possible outcome by its probability and sums the results, giving a single number that represents the average payoff of a decision. It is the mathematical foundation of rational decision-making under uncertainty — well grounded in decision theory — but it has real limits: probabilities are often uncertain, outcomes are not always quantifiable, and raw expected value ignores risk aversion that can be legitimate.

Expected value is the weighted average of all possible outcomes, where the weights are probabilities. It sounds like a purely technical concept, but it is also a practical attitude: the willingness to evaluate a decision by its long-run average rather than by any single outcome. Most decisions people find hardest — ones with uncertain large upside, or low-probability catastrophic downside — become substantially clearer when analyzed this way. The practices below turn EV from an abstract formula into a working decision tool.

The practices (7)

Why it works

The human mind defaults to imagining one future — usually the intended outcome — rather than a distribution. Writing out scenarios with explicit probabilities forces engagement with alternatives that would otherwise be invisible. Each scenario also makes the decision decomposable: you can examine whether your probability estimates are defensible rather than just trusting a gut feeling about "how it will go."

How to do it

1List the four to six most meaningfully different outcomes of your decision.
2Assign a probability to each, ensuring they sum to 100%.
3Assign a value (in whatever units matter: money, time, satisfaction) to each outcome.
4Multiply each probability by its value and sum — the highest EV option is the baseline recommendation.

Evidence

Observational

Scenario planning and explicit probability elicitation are established components of structured decision analysis, which has been shown to outperform unaided intuition in complex multi-attribute decisions.

Honest caveat: Formal decision analysis evidence is strongest in high-stakes organizational settings; its benefit in everyday personal decisions is plausible but less directly studied.

Common mistake: Listing scenarios until you have covered "everything," ending up with twenty imprecise entries that make probabilities arbitrary and the calculation useless.

Go deeper on this practice →

Practice this with IX Coach

Reading about a practice changes nothing on its own. IX Coach turns these into a guided, adaptive routine — discerning where you are in real time and walking the practice with you, session after session.

Start with IX Coach →