Developing Expert Political Judgment: The Impact of Training and Practice on Judgmental Accuracy in Geopolitical Forecasting Tournaments
by Welton Chang, Eva Chen, Barbara Mellers, Philip E. Tetlock · Judgment and Decision Making (Cambridge University Press)
An open-access paper from the Good Judgment Project. It tests the CHAMPS KNOW training, a module under one hour on base rates, comparison classes and debiasing, and finds it improved Brier scores by 6–11% over four tournament years. The paper sets out the module's principles. This is the primary evidence that forecasting can be taught, and it lays out the actual training principles (CHAMPS KNOW). It is more actionable than the broader Mellers et al. papers.
More resources on Forecasting & Calibrated Thinking
Good Judgment Open
A free public forecasting platform run by Good Judgment Inc., the company Tetlock co-founded. You make probability forecasts on political, economic and technology questions, get scored against the crowd, and join sponsored challenges. Top performers can be invited to become Superforecasters.
Calibrate Your Judgment
A free interactive trainer from ClearerThinking.org. You answer question sets with a confidence level, and it shows how often your 60%, 80% or 95% answers are actually right, tracking your progress. It targets overconfidence directly. Calibration is a skill you build through practice, and this is the best-known free drill for it. It complements the reading. Requires a free account.
Edge Master Class 2015: A Short Course in Superforecasting
Full transcripts of a five-session seminar Philip Tetlock gave in 2015, with Daniel Kahneman, Robert Axelrod and others joining the discussion. It covers forecasting tournaments, accountability for accuracy, ways to aggregate forecasts, and the limits of prediction.
Expert Political Judgment: How Good Is It? How Can We Know? (New Edition)
Tetlock's 20-year study of about 28,000 predictions made by political experts. He scores their calibration and discrimination against simple baselines and shows why "foxes" (eclectic thinkers) beat "hedgehogs" (committed to one big idea). The 2017 edition adds a preface covering the later tournament research. This is the research foundation that Superforecasting simplifies, for readers who want the method and the statistics. It overlaps with the popular book but works at a much deeper, technical level, so it doesn't add choice paralysis.