The Prisoner's Dilemma at Your Game Night

Two suspects are arrested. The police separate them and offer each the same deal: betray your partner and walk free while they serve three years — but if you both betray each other, you both serve two. If you both stay silent, you each serve one. That's the prisoner's dilemma, the most famous thought experiment in game theory, and you don't need a police station to see it play out. You need a game night.

Why betrayal is "rational"

Look at the deal from one prisoner's perspective. If your partner stays silent, betraying them gets you zero years instead of one — better. If your partner betrays you, betraying them back gets you two years instead of three — also better. Betrayal wins in both scenarios, so a purely self-interested player always betrays. The trap: when both players follow this bulletproof logic, they both end up worse off than if they'd trusted each other. Individually rational, collectively terrible.

Where you've already played this

Monopoly alliances. Two players agree not to build hotels on adjacent streets. Each turn, both quietly calculate whether breaking the pact first is worth it. It usually is — which is why Monopoly alliances have the shelf life of a banana.

Risk truces. "You don't attack from Ukraine, I won't touch Alaska." Both players honor it right up until the exact moment one of them can win by not honoring it.

Splitting the bill. Ten friends agree to split evenly. Each individually reasons that ordering the expensive steak costs them only one-tenth of its price. When everyone reasons this way, the bill explodes. Economists literally call this the "diner's dilemma."

The twist: repetition changes everything

Here's where it gets hopeful. In 1980, political scientist Robert Axelrod ran a tournament where computer programs played the prisoner's dilemma against each other hundreds of times. The winner wasn't a ruthless betrayer or a saintly cooperator. It was Tit for Tat — a four-line strategy: cooperate first, then copy whatever your opponent did last round.

Tit for Tat wins because repeated games have memory. Betray someone at game night and you gain one game — and lose the next ten, because nobody allies with you again. Your reputation is the mechanism that makes cooperation rational. This is why the dilemma feels different with strangers than with friends: with friends, you're never playing just once.

Using the dilemma on purpose

Want to feel the tension directly? Play "Split or Steal" at your next party: two players each secretly choose to split or steal a pot of points (or candy). Split/split shares it, steal/split takes it all, steal/steal gets nothing. Run ten rounds with rotating pairs and watch alliances, grudges and reputations form in real time. It's the entire field of game theory in a bowl of M&Ms.

The takeaway

The prisoner's dilemma explains why one-shot interactions tend toward selfishness and repeated ones toward cooperation. Game night is a repeated game. So is your office, your friend group and your marriage. Play accordingly — and if you must betray someone in Risk, at least accept that the next three games are going to be personal.