Game theory has endless applications and can be very useful to you in the many competitions of life. As an example, I want to tell you the story of Yul Kwon. Yul competed in the CBS reality show Survivor back in 2006. On Survivor, contestants are stranded on an island and compete for a $1 million prize. While survival is important, strategy is even more important. because contestants must navigate twists in the game and survive tribal council, where each episode, one player is voted out by the other contestants.
Once there are only three people left, a jury composed of the players they voted off votes for a winner of the game. Ewell prepared for Survivor by studying game theory. He was particularly interested in the work of Robert Axelrod.
a political scientist at the University of Michigan. In the 1980s, Axelrod was studying the evolution of cooperation in humans. He wanted to explain why we would evolve to cooperate with one another at all. As part of his research on the topic, he set up these tournaments where computer programmers were invited to write a program which could play the prisoner's dilemma.
It set up the dilemma as a repeated game. Each program would face off with each other in a round-robin format where they played 200 games of the Prisoner's Dilemma in each round. The idea of playing it 200 times was that perhaps cooperation in the Prisoner's Dilemma made sense if players knew they would have to play it again right afterwards.
The quick rewards of confessing seemed small if you could have big rewards from cooperation 200 times. Many programmers wrote very complicated programs, which considered the whole history of the games played so far, and had complicated models for predicting their opponent's behavior. But the winner of the tournament was Anatole Rapoport, whose program started with a strategy of cooperation, but then it just copied whatever the opponent did in the previous game.
He called it tit-for-tat. If the opponent continued to cooperate, his program kept cooperating. If the opponent defected, then his program would defect in the next game. But if the opponent switched back to cooperation, tit-for-tat was happy to return to cooperation as well.
Through these tournaments and his other research, Axelrod devised these rules for success in cooperative games. Be nice. Cooperate. Never be the first to defect. Be provokable.
If your opponent double crosses you, don't be a pushover about it. Return defection for defection. Cooperation for cooperation. Don't be envious.
Many don't just want to win, they want to crush their opponents. But this obsession with your opponent's score tends to cloud your ability to maximize your own score. Axelrod recommends focusing on your own score instead.
And don't be too clever. To win the trust of allies, you can't be tricky. Clarity is essential for others to cooperate with you. This was the strategy Yule followed for Survivor.
A critical moment in the game came when there were just 12 players left. The players were split evenly between two tribes, which meant that immunity challenges were critical. Whichever tribe lost the challenge would have to vote someone in their own tribe out, and when all the players came together in an inevitable merge, they would be outnumbered by the other tribe in a game where having the majority on your side is essential.
But then, the host, Jeff Probst, offered players an opportunity. If they wanted, any player could mutiny and join the other tribe now. Two people. took him up on this offer, Candace and Jonathan. Ewell's tribe was crippled by this mutiny, and they were now badly outnumbered.
But as luck would have it, his group of four managed to win every challenge going forward, and the other tribe ended up having to send three of their members home before the tribes merged. When the remaining players merged into one tribe, things still looked grim for Yule. His alliance was outnumbered five to four, and it looked like they would each be voted off one by one. But Yule had managed to find an immunity idol, which gave him a one-time power to cancel any votes cast for him to leave. A lesser player would have played it at the next Tribal Council, survived one more round, and then gotten voted off anyway.
But Yule devised a brilliant strategy, which meant he never actually had to play the idol. He went to Jonathan and convinced him to switch sides. He showed him that he had this idol, and Jonathan agreed to flip, completely shifting the balance of the game. At least that's what they show you. on the show.
But Ewell's strategy was much better than that. Ewell convinced Jonathan to switch with game theory. Ewell told Jonathan that no matter what, his group of four was going to write down his name at the next tribal council.
This isn't much of a threat since five votes still beats four, but then Ewell showed Jonathan he had the idol. This was risky because if the other side knows Yule has the idol. They can just vote someone else out on his side instead.
But again, Ewell was one step ahead. He told Jonathan that he would randomly choose one of the four to give it to at each tribal council. So the other side would never know who had it.
So there would always be a chance Jonathan would get voted out. This was a strategy of mutually assured destruction. Ewell was deploying a strategy that was sure to cause Jonathan to lose the game.
Because even if the idol swapping didn't work, he promised that none of the four on his side would vote for Jonathan to win at the end. Ewell laid out the math Jonathan was facing. There was a three-fourths chance he would survive the next vote.
Because there was a one-fourth chance his team would vote for the idol holder. And if he survived that, there was a two-thirds chance he would survive the next round after that, giving Jonathan a one-half or 50% chance of making it to the final five, successfully defeating Yule's alliance and making it to the final five with his current alliance. But if Jonathan just switched and joined Yule's side, his chances of making the final five were essentially guaranteed.
You will make sure that Jonathan's... dominant strategy was to defect. And Jonathan defected.
Yule never had to play his idol. And that play paid off. We're now down to two people. Yule, who dominated strategically, like maybe nobody else ever has. Ozzy, who dominated physically, fair to say, like nobody ever has.
It comes down to one vote left. The winner of Survivor Cook Islands.