[!todo] Seed note. A starting point, not a finished note yet.
Superforecasters are the top performers Philip Tetlock identified in the Good Judgment Project, the team that won IARPA’s 2011-2015 geopolitical forecasting tournament against four university competitors and reportedly beat intelligence analysts with classified access. They are not credentialed experts; they are ordinary people with a method. The defining habit overlaps exactly with Reference Class Forecasting: they start from the outside view, asking “how often do things of this sort happen in situations of this sort,” then adjust in small, frequent increments as new evidence arrives, scoring themselves with Brier scores to stay honest. Tetlock’s earlier work found that average expert prediction was barely better than chance, which is what made the existence of a reliably better group notable. Documented in Superforecasting by Tetlock and Gardner, and seeded from Reference Class Forecasting.