From Incan Gold to Dominion: Unraveling Optimal Strategies in Unsolved Games

While some games have inherent optimal strategies, strategies that will win no matter how the opponent plays, perhaps more interesting are those that do not possess an objective best strategy. In this paper, we examine three such games: the iterated prisoner’s dilemma, Incan Gold, and Dominion. Through computer simulations, we attempted to develop strategies for each game that could win most of the time, though not necessarily all of the time. We strived to create strategies that had a large breadth of success; that is, facing most common strategies of each game, ours would emerge on top. We begin with an analysis of the iterated prisoner’s dilemma, running an Axelrod-style tournament to determine the balance of characteristics of winning strategies. Next, we turn our attention to Incan Gold, where we examine the ramifications of different styles of decision making, hoping to enhance an already powerful strategy. Finally, we unpack Dominion, attempting to uncover the most potent combination of cards that can lead to almost universal success.