DAATANByline authors and outlets scored on their own directional forecasts (author_lean), resolved against outcomes — the same accountability the user leaderboard gives predictors. Lower Brier is better; 0.25 is a coin flip.
Shadow scoring: this track record is informational only and does not feed the live forecast estimate. Skill is a conservative rating (discounts low-sample noise); Brier is the raw resolved-outcome accuracy of each source's own directional lean.