Brier score
Brier score — the mean squared error of probability forecasts; lower is better.
Brier score measures how far stated probabilities sit from what actually happened, across many forecasts.
Brier = (1/N) × Σ (p_i − o_i)²o = 1 if the event happened, 0 otherwise.Why it matters: it is a proper scoring rule — minimised only by reporting your true belief — so a forecaster cannot improve it by exaggerating confidence. Hit rate has no such property.
Common mistake: reading it in isolation. A Brier of 0.21 means nothing until compared against 0.25 (the naive 50% forecaster) and against the de-vigged market price on the same events.
Related
Check any of this against our record
Every signal CONSENSUS publishes carries the bookmaker odds fixed before the event starts and the settled result afterwards — including the drawdowns and the losing runs. The running total is on the front page and every entry is in the log.