What a zero-sum game is
A zero-sum game is a situation where one person's gain equals another person's loss. Add up all the gains and losses and the total is zero. That simple rule is the definition.
Common examples:
- Poker or chess. If one player wins, the other loses.
- A fixed amount of money split between two people. If you get more, the other gets less.
- Some auctions and betting situations.
Zero-sum is a model. It assumes the pie cannot grow. People compete to divide what already exists.
How to think about it
The key idea is conservation. Value moves from one player to another. There is no mutual benefit. That makes thinking about strategy easier in some ways. If you can measure your opponent's loss, you know your gain.
Because the interests are exactly opposite, optimal play often focuses on preventing the opponent from getting a better result rather than trying to create value.
Formal idea in one sentence
Each outcome gives payoffs to players. Sum the payoffs across players. If every possible outcome sums to zero, the game is zero-sum.
For two players, we often write the payoff to player A as x and the payoff to player B as -x.
Key concepts
- Payoff: The reward or cost a player gets.
- Value of the game: The expected payoff a player can guarantee with optimal play.
- Strategy: A plan or rule for making decisions.
- Pure strategy: Always make the same move in a given situation.
- Mixed strategy: Randomize between moves according to fixed probabilities.
- Minimax theorem: In two-player zero-sum games, the player who wants to maximize their minimum gain can do as well as the other player can force them down. That leads to a stable expectation when both play optimally.
Minimax in plain words
If you must assume the worst about your opponent, choose the strategy that gives the best of those worst-case outcomes. Your opponent does the same. In many zero-sum games this gives a point where neither can improve by changing strategy. That is the game's value.
How zero-sum games are solved
- Small, simple games: Draw a payoff matrix and look for best responses.
- Mixed strategies: Use algebra to find probabilities that make the opponent indifferent.
- Linear programming: Turn the problem into a linear optimization and solve with standard methods.
- Backward induction: For sequential games with perfect information, solve from the end back to the start.
Computer programs use these methods to beat humans at chess, poker, Go, and other games.
Why they matter
Zero-sum models are useful when resources are fixed or when competition cancels cooperation. They are a good first approximation in many fields:
- Economics: Pricing wars or market share fights can look zero-sum in the short run.
- Politics: Elections and certain negotiations are often treated as zero-sum.
- Military and security: One side's advantage is the other side's loss.
- Sports and games: Most sports and many games are effectively zero-sum.
Knowing a situation is zero-sum guides strategy. If it really is zero-sum, cooperation will not increase total payoff. That points toward defensive play and tactics to limit opponent gains.
When the model breaks
Many situations are not zero-sum:
- Trade where both sides gain.
- Projects that create new value.
- Negotiations where adding options lets both parties benefit.
If you treat a non-zero-sum situation as zero-sum, you will miss opportunities to cooperate and create value.
Common misconceptions
- Zero-sum means cruel or simple. It only describes the payoff structure.
- Zero-sum only applies to two players. It can apply to many players if total payoffs sum to zero.
- Real life is never zero-sum. Some situations are close enough that the model helps.
Quick examples you can test
- Coin flip bet: If you bet $1 on heads in a fair coin flip, your expected value is zero. It is zero-sum between the two bettors.
- Market share with fixed customers: If total customers are fixed, gaining a customer reduces others. That is approximately zero-sum.
Final takeaways
- Zero-sum games model pure competition.
- Strategy focuses on defense and guaranteeing a minimum outcome.
- The minimax idea gives a clean way to find optimal play.
- Check whether the real situation truly keeps the pie fixed. If not, look for cooperative or value-creating moves.
Related terms: game theory, minimax, Nash equilibrium, mixed strategy, payoff.