Pick/Ban Elo Model Alignment

This page compares historical veto choices with the model's greedy map-Elo recommendation at each step. "Average Model Regret" is the Elo-gap difference between the observed choice and that recommendation; lower means closer model alignment. The leakage-safe rebuild uses ratings strictly before the recorded match completion cutoff, excludes ratings produced by that same match, and defaults a missing team/map rating to 1000; completion time is the proxy because match start is not stored. Ban comparisons are counterfactual because banned maps are not played. These rankings show model alignment, not proven coaching quality or causal win lift. Read the evaluation scope and limitations.