Repeated Game Theory vs One-Shot Game Theory

This article explores the fundamental differences between one-shot and repeated game theory, detailing how the frequency of interaction transforms player strategies, incentives, and outcomes. While one-shot games often lead to conflict and mutual defection due to a lack of future consequences, repeated games introduce the concepts of reputation, retaliation, and long-term cooperation, fundamentally changing how rational decision-makers behave.

The Core Difference: Time Horizon

The primary distinction between one-shot and repeated game theory lies in the time horizon of the interaction.

• One-Shot Games: Players interact exactly once. Because there is no future, players have no reason to worry about the long-term consequences of their actions. Decisions are made purely to maximize immediate payoffs.
• Repeated Games: Players interact multiple times. This ongoing relationship means that a choice made today will influence how opponents behave tomorrow. Players must balance immediate gains against potential future rewards or punishments.

Incentives and the Evolution of Cooperation

The shift from a single interaction to a sequence of interactions fundamentally changes the incentives of the players, particularly in scenarios like the Prisoner’s Dilemma.

In a one-shot Prisoner’s Dilemma, the dominant strategy for both players is to defect (betray the other). Since there is no future interaction, cheating yields a higher individual payoff regardless of what the opponent does, leading to a suboptimal Nash equilibrium where both players lose out on the benefits of mutual cooperation.

In a repeated Prisoner’s Dilemma, cooperation can become the rational choice. Because players interact repeatedly, they can employ conditional strategies to enforce cooperation. The threat of future retaliation discourages defection in the present.

Key Strategies in Repeated Games

Repeated games allow for complex, history-dependent strategies that are impossible in one-shot scenarios. The most notable strategies include:

• Tit-for-Tat: A player starts by cooperating and then simply mimics the opponent’s previous move in every subsequent round. This strategy promotes cooperation while punishing defection immediately.
• Grim Trigger: A player starts by cooperating, but if the opponent defects even once, the player defects for the rest of the game. This extreme threat of permanent punishment enforces strict cooperation.

The Role of Reputation and Trust

In one-shot games, trust and reputation do not exist because players have no history and no future. In repeated games, reputation is a valuable asset.

A player who consistently cooperates builds a reputation for trustworthiness, which encourages others to cooperate with them. Conversely, a reputation for defection leads to exclusion or punishment. The mathematical foundation for this is described by the Folk Theorem, which states that in infinitely repeated games, any mutually beneficial outcome can be sustained as a Nash equilibrium if players value future payoffs sufficiently.

Finite vs. Infinite Horizons

The difference between one-shot and repeated games also depends heavily on whether the repetition has a known end point:

• Finitely Repeated Games: If players know the exact number of rounds (e.g., the game will end after 10 rounds), cooperation often collapses. Through a process called backward induction, players realize that defection is the dominant strategy in the final round. Anticipating this, they defect in the second-to-last round, and the logic cascades backward, making the repeated game behave like a series of one-shot games.
• Infinitely (or Indefinitely) Repeated Games: If the end of the game is unknown or if the game could theoretically continue forever, backward induction is impossible. Cooperation remains highly sustainable because players can never be sure when the “last round” will occur to exploit their opponent without consequence.