2019 年 62 巻 2 号 p. 64-82
In this paper, baseball is formulated as a finite non-zero-sum Markov game with approximately 6.45 million states. We give an effective dynamic programming algorithm which computes equilibrium strategies and the equilibrium winning percentages for both teams in less than 2 second per game. Optimal decision making can be found depending on the situation—for example, for the batting team, whether batting for a hit, stealing a base or sacrifice bunting will maximize their win percentage, or for the fielding team, whether to pitch to or intentionally walk a batter, yields optimal results. Based on this model, we discuss whether the last-batting team has an advantage. In addition, we compute the optimal batting order, in consideration of the decision making in a game.
