Forecasting skill is real, measurable, and learnable

The book's founding fact is the Good Judgment Project, a forecasting tournament run for the U.S. intelligence community's IARPA research arm, in which thousands of amateur volunteers predicted the outcomes of real-world questions — will there be a coup in country X, will a treaty be signed by a given date — that were later scored against what actually happened.

A small group, roughly the top 2%, consistently beat everyone else, including professional intelligence analysts with access to classified information. Tetlock's team called them superforecasters, and crucially, their edge wasn't a one-off lucky streak — it persisted across years and across different topics, which is the statistical signature of genuine skill rather than random variance.

This directly rebuts the fatalistic view (voiced by pundits who dislike being graded) that predicting complex human affairs is basically impossible. It's hard, but a measurable slice of people are demonstrably better at it, and — the book's real payload — their methods can be taught.