Game Theory Isn’t Really About Games, Its the Science of Decision Making
- Games of conflict are studied, through a mathematic and scientific lens, to find the best strategies for competition given human behavior.
- Likewise, games of cooperation are studied in the same way, but to find the best strategies regarding fairness and cooperation.
Game Theory: The Science of Decision-Making
FACT: Game theory can be considered “the science of strategy”.Game theory is not game studies (the study of games).
Game Theory and Oligopoly: Crash Course Economics #26
What Types of Games Are Studied in Game Theory
A good example of a game that is studied in game theory (in competitive game theory at least) is the classic “zero-sum game” “the prisoner’s dilemma” (see image and video below).
The game involves some A or B choices rather than complex rulesets based on chance or skill. The behavior of people over time is analyzed, and the results used to understand decision-making better in areas like business, economics, and politics where complex social dynamics rule the day.
Understanding the Prisoner’s Dilemma (the classic Zero-Sum game theory game)
The image below shows a simple example of the prisoner’s dilemma regardding war, but lets use the more common example of two thiefs who are being interrogated.
The two thiefs go into a negotiation blind (they don’t know the deal the other will make), they can choose to cooperate (not rat on each other) or not (rat on each other).
If both teams cooperate, they both get 2 years in jail (see game theory terms).
If one team defects (rats) and the other cooperates (doesn’t rat), then the one who cooperates gets shafted (in this case they get 10 years in jail).
If both teams don’t cooperate then no-one gets shafted (neither get 10), but no one gets the payout (in this case they both get 5 years in jail).
The game is then played over and over and decision making behavior is observed and analyzed as players learn from past results.
The math tells us that the best bet, as we can never know what the other person will do, is to always rat (to always not cooperate). As, if the other thief is virtuous we walk, and if they are a rat then we get 5 and not 10.
This is all fun and games, until we apply it to say a Cold War. As, in a Cold War, the implication is that both sides should press the button (as if you press the button you may win, but if you don’t then the other side presses the button… you lose). Problem here is that, no one really wants anyone pressing any dang button.
Luckily, there are many [often more complex] versions of this game, and other game-theory games tend to be of this style.
If we apply this to something like international politics we can start to understand why although nations tend to want the best case “both teams cooperate scenario”, a little bit of fear and distrust can easily lead to the non-cooperation scenario.
So aside from studying a small subset of very specific types of games, game theory has little to do with games, and everything to do with better understanding human behavior.
This is important to grasp, as the title “game theory” makes it sound like “games” are the subject. The theory of how to beat a person at a game of skill, or how to increase your odds in a game of chance, is only related in the very loosest of senses.
FACT: There are also cooperative games (as explained in the first video on the page). These allow us to study how to best cooperate, rather than how to best compete. An example given in the first video is a cookie baking game that tells you how to divide payouts between members when one member is a stronger player. These sorts of games have been studied since Pascal first invented probability theory and before.
International Relations 101 (#7): The Prisoner’s Dilemma
TIP: Game Theory, its all fun and games… until you realize that it is more about math, business, and international relations than Call of Duty.