What is Expert Political Judgment about?

Expert Political Judgment is Philip Tetlock's report on a twenty-year study of political forecasters. Starting in 1984, Tetlock collected predictions from nearly three hundred experts — economists, political scientists, intelligence analysts, journalists — and then tracked whether those predictions came true. The results were damning: experts performed barely better than chance, and in many cases worse than simple statistical extrapolations or informed amateurs. The book is the primary source for the now-famous claim that political and economic experts are poor forecasters.

The study's most striking finding was not that experts were wrong, but that their accuracy depended heavily on how they thought rather than what they knew. Tetlock borrowed Isaiah Berlin's distinction between the fox and the hedgehog. Hedgehogs know one big thing: they have a single organizing theory — Marxism, market fundamentalism, realism — and they use it to explain almost everything. Foxes know many small things: they draw on multiple frameworks, update their views more readily, and are comfortable with uncertainty. In Tetlock's data, foxes outperformed hedgehogs in prediction, often substantially, especially over longer time horizons.

The book has two levels. The first is empirical: here is what the data show about expert accuracy. The second is psychological and philosophical: why are experts poorly calibrated, why does the public continue to trust overconfident forecasters, and what would a culture of epistemic accountability look like? Tetlock explores how experts protect their self-image through hindsight bias ("I knew it all along"), creeping determinism, and belief system defenses — the rhetorical moves that let a confident but wrong prediction be reframed as "almost right."

Expert Political Judgment is an academic book more than a popular one — denser and more qualified than Tetlock's later Superforecasting. But it is also more methodologically rigorous. Readers willing to work through the research design and scoring methodology will come away with a detailed and intellectually honest account of what expertise does and does not provide in domains characterized by genuine complexity and feedback lags.