Game Theory: A Simple Strategy That Will Change Your Life Forever

A simple, repeatable strategy—start cooperative, retaliate when wronged, and forgive to restore cooperation—beat far more complicated approaches in repeated versions of the classic “prisoner’s dilemma.” The practical punchline is that long-term success in real relationships, workplaces, and even international conflicts often depends less on “winning” in the moment and more on shaping incentives over time.

The setup begins with a domestic fairness problem: two roommates alternate dish duty (Sunday vs. Wednesday). After a few weeks, one roommate stops doing their scheduled dishes, letting the pile grow. The other roommate eventually does the dishes to prevent the situation from spiraling, but the pattern repeats—then worsens—until it becomes clear that the issue isn’t a one-off lapse. The dilemma becomes strategic: should the responsible roommate keep doing the dishes to maintain order, or stop and let consequences force change? In game theory terms, the incentive to defect (avoid the work) can produce outcomes that are worse for both sides.

Game theory is presented as a mathematical way to model decisions where outcomes depend on what others choose—whether in cooperation (teams, partnerships, alliances) or in non-cooperative settings where each party pursues its own interest. A one-shot version of the dish problem resembles the “Golden Balls” game show: with only one decision and no future interaction, the dominant move is to steal, because it pays off regardless of the other person’s choice. But real life rarely ends after one round. Relationships, contracts, and conflicts continue, and the “game” becomes iterated—played again and again under uncertainty.

That shift drives the key experiment. In 1980, political scientist Robert Axelrod ran computer tournaments of an iterated prisoner’s dilemma. Each strategy program played 200 rounds against every other strategy (and against a copy of itself), earning points based on whether it cooperated or defected. Fourteen strategies entered the first tournament, and Axelrod added a random baseline. Many competitors were complex—probing, exploiting, or mixing moves—but the consistent winner across repeated runs was “Tit for Tat.”

Tit for Tat begins by cooperating. After that, it copies the opponent’s previous move: it keeps cooperating as long as the other side cooperates, defects immediately after defection, and then returns to cooperation once the opponent cooperates again. Axelrod’s surprise wasn’t just that it won—it won with a blend of traits: “nice” enough to avoid unnecessary conflict, “retaliatory” enough to discourage exploitation, “forgiving” enough to restart cooperation, and “clear” enough that others can predict and respond.

A second tournament introduced a more realistic twist: the number of rounds was unknown, removing a clean endgame. Tit for Tat still prevailed. The broader takeaway is not that cooperation is naive, but that it can be rational when paired with proportional consequences and the expectation of future interaction. The limitations are acknowledged—real-world systems include many players, shifting incentives, asymmetric power, and human emotion—but the guiding lesson remains: every interaction creates precedent, and over time, strategies that balance kindness with accountability tend to produce the best overall outcomes.