Good Judgment: How to Improve Your Decision Making and Make Better Choices: Summary & Key Insights

In the Good Judgment Project, we recruited thousands of volunteers to forecast hundreds of real-world events: elections, policy decisions, economic trends. We measured accuracy rigorously, rewarding participants who gave probabilities that matched actual outcomes. Over time, a small group consistently rose above the rest—our superforecasters.

These individuals weren’t specialists with decades of insider training. They were ordinary people—engineers, retirees, teachers—who simply thought better. They approached uncertain questions the way a scientist approaches data: ask precisely, measure carefully, revise constantly. Their secret lay not in intuition but in process. They decomposed problems, checked assumptions, sought diverse sources, and treated beliefs as hypotheses to test, not treasures to protect.

When analysts asked why superforecasters succeeded, it became clear that their success wasn’t magic—it was discipline. They thought in probabilities, not certainties. They saw forecasting as learning. And they proved that accuracy in judgment is attainable when humility meets method. Through their example, we saw that good judgment can be trained, refined, and cultivated.

The mental habits of superforecasters run counter to much of human psychology. People crave certainty and often cling to their first beliefs, interpreting new evidence through the lens of what they already think. Superforecasters resist that pull. They live in a state of perpetual curiosity.

They are open-minded yet skeptical. When they encounter information, they ask, “Is that really true?” and “What would it mean if it were?” They think in probabilities—assigning 60 percent or 40 percent rather than declaring something ‘true’ or ‘false.’ They update continually, adjusting those probabilities as fresh data appears. This mindset reflects Bayesian reasoning: the continuous refinement of beliefs.

Equally important, they value precision in language. When a question asks, “Will North Korea test a nuclear weapon this year?” they clarify the timeframe, define ‘test,’ and consider what counts as confirmation. That clarity guards against the vagueness that ruins forecasting.

The deeper lesson isn’t about statistics, but about humility. Understanding that our brains are fallible opens space for learning. The best forecasters embrace being wrong as the only path to becoming more right next time.