Cooperation wins, even if you lose every time

The most famous problem in game theory is called “The Prisoner’s Dilemma“, and you’ve likely seen it play out in game shows over the years. In game shows, it often comes down to the final two players with a secret vote to split the pot or try to take more for themselves, and the logic behind making the “right” choice can be very challenging.

The full explanation of how this dilemma works is fairly complicated (Investopedia has a great post about it), but it goes something like this:

• You and someone else are in separate rooms, and each offered a deal.
• If you both cooperate, you each get $3.
• If you both defect, you each get $1.
• If you cooperate but your opponent defects, they get $5 and you get $0.

Clearly the best overall strategy is to both cooperate, but the best individual strategy is to always defect. If you defect you always come out ahead, because if:

• Your opponent also defects, you get $1 instead of $0.
• Your opponent cooperates, you get $5 instead of $3.

It’s always better to defect, right?

The long game

For a single game, yes, you’re best off defecting. However, in the long run you’re best off cooperating quite a bit. Derek Muller at Veritasium has a fascinating video about this, in which they put a bunch of strategies to the test where they match up different techniques over hundreds of games. The video is nearly 30 minutes long, but I found it to be wildly insightful (as did millions of other viewers on YouTube):

If you don’t have time to watch the video right now, after dozens of different strategies competed in thousands of match-ups, things ended up very counter-intuitively.

• The selfish strategies that almost always defect are able to win every game, but end up with relatively little money in the end.
• The generous strategies that almost always cooperate end up losing almost every game, but end up with quite a bit more money in the end.

In the long run, the generous strategy they share in the video called “tit for tat” didn’t win a single individual matchup, but dominated every other strategy in terms of total value across all of their games.

The lessons from the video and the tests showed that the best strategies did three things, and I found them to be good ideas for life.

1. Be nice.
2. Be forgiving.
3. Don’t be a pushover.

The “tit for tat” strategy would cooperate as much as possible (“be nice”), but if another strategy defected they’d defect right back (“don’t be a pushover”), but they would be willing to cooperate again in future moves (“be forgiving”).

Take the optimistic view in life and assume the best from people, but don’t be a total pushover. In the end, though, being forgiving is likely the best path for everyone.