Game Theory and Strategic Choices

Most decisions in earlier lessons treated the world as a fixed backdrop you optimise against. But often the most important part of the backdrop is other people, who are deciding at the same time and reacting to you. Pricing against a competitor, negotiating a deal, splitting shared work, or agreeing to disarm are all situations where the best move depends entirely on what the other side does.

Game theory is the plain study of these situations. It does not require heavy mathematics to be useful. Its core lesson is a habit: before you choose, think one move ahead about how others will respond to your choice — and how you would respond to that.

Strategic interaction

A situation is strategic when your payoff depends not only on your action but on others' actions too. The distinguishing question is simple: does my best choice change depending on what they choose? If yes, you cannot decide well by looking only at your own preferences. You have to model theirs.

The prisoner's dilemma

The classic illustration: two suspects are questioned separately. Each can stay silent (cooperate with the other) or talk (defect). If both stay silent, both get a light sentence. If both talk, both get a heavy one. But if one talks while the other stays silent, the talker walks free and the silent one takes the worst outcome.

Here is the trap. Whatever the other person does, each individual is better off talking. So both talk — and both end up worse off than if they had both stayed silent. Individually rational choices produce a collectively bad result. This is the heart of the dilemma, and it explains price wars, arms races, and overused shared resources.

Nash equilibrium

A Nash equilibrium is a combination of choices where no player can do better by changing their own choice alone, given what everyone else is doing. In the prisoner's dilemma, both defecting is the equilibrium: neither can improve by unilaterally switching to silence. Notice that an equilibrium need not be the best outcome for the group — it is just stable, in the sense that no one has a private reason to deviate.

Repeated interaction

The gloomy conclusion softens when the game repeats. If you will face the same person again and again, defecting today invites retaliation tomorrow, and cooperating can pay off over time. When the future matters enough, cooperation becomes a rational strategy, not just a hopeful one.

Two neighbouring cafes each decide whether to keep prices normal or cut them to steal customers. If both hold, both earn steadily. If both cut, both lose margin for no net gain in share. If one cuts while the other holds, the cutter wins the week. Each is tempted to cut — so both do, and both bleed. Recognising this as a repeated prisoner's dilemma, one owner signals restraint, holds prices, and matches only if the other cuts. Over months the pattern settles into the better shared outcome.

Not every conflict is a prisoner's dilemma, and treating them all as one is a mistake. Some situations are pure coordination, where both simply want to match each other and there is no temptation to defect. Others are genuinely one-shot with no future, where a warmer strategy can be exploited. And where communication and binding agreements are possible, parties can escape the dilemma directly. The tool is diagnosis, not a script that says always cooperate or always defect.

The scenario was formulated around 1950 by Merrill Flood and Melvin Dresher, researchers at the RAND Corporation. The mathematician Albert Tucker gave it the memorable prisoner framing and the name. The underlying solution concept, the equilibrium in which no player gains by changing alone, was developed by John Nash, whose 1950 work on non-cooperative games later shared in a Nobel Prize in economics.

Whether cooperation can emerge among self-interested players was tested by political scientist Robert Axelrod around 1980. He invited experts to submit strategies for a repeated prisoner's dilemma and ran them against one another in a computer tournament. The winner, submitted by Anatol Rapoport, was strikingly simple: tit for tat — cooperate on the first move, then do whatever the opponent did last time. It was nice, retaliatory, forgiving, and clear, and it outperformed far more elaborate schemes. Axelrod described the results in his 1984 book The Evolution of Cooperation.

• "Game theory needs advanced maths." The core intuitions require none; the habit of thinking a move ahead is the point.
• "The rational move is always to defect." In repeated interaction, cooperation is often the rational strategy.
• "A Nash equilibrium is the best outcome." It is only stable; it can be worse for everyone than an alternative.
• "Every conflict is a prisoner's dilemma." Many are coordination or other games with different logic.

Given a short scenario, decide whether it is a true prisoner's dilemma, and if repetition would change the smart strategy.

Think Like a Maester: Before you choose, ask how the other side will respond to your choice — and whether you will meet them again.

1. What makes a situation strategic rather than a simple optimisation?
2. Why do both players defect in a one-shot prisoner's dilemma despite it leaving both worse off?
3. What does it mean for a set of choices to be a Nash equilibrium?
4. How does repeated interaction change the incentive to cooperate?
5. What were the features of the tit-for-tat strategy that won Axelrod's tournament?

When others are deciding too, your best move depends on theirs. The prisoner's dilemma shows how individually rational choices can leave everyone worse off, and a Nash equilibrium can be stable yet poor for the group. But repetition changes the game: retaliation and reward make cooperation rational, as Axelrod's tournaments and the success of tit for tat demonstrated. The practical skill is to think one move ahead.

• Risk, Trade-offs and Opportunity Cost
• Reversible and Irreversible Decisions