Abstract One drawback with using plan recognition in adversarial games is that often players must commit to a plan before it is possible to infer the opponent’s intentions. In such cases, it is valuable to couple plan recognition with plan repair, particularly in multi-agent domains where complete replanning is not computationally feasible. This paper presents a method for learning plan repair policies in realtime using Upper Confifidence Bounds applied to Trees (UCT). We demonstrate how these policies can be coupled with plan recognition in an American football game (Rush 2008) to create an autonomous offensive team capable of responding to unexpected changes in defensive strategy. Our realtime version of UCT learns play modififications that result in a signifificantly higher average yardage and fewer interceptions than either the baseline game or domain-specifific heuristics. Although it is possible to use the actual game simulator to measure reward offlfline, to execute UCT in real-time demands a different approach; here we describe two modules for reusing data from offlfline UCT searches to learn accurate state and reward estimators