EconLearn

Dominant Strategy vs Mixed Strategy

Dominant Strategy and Mixed Strategy are related concepts in AP Economics that students often mix up. A dominant strategy is a strategy that results in the highest payoff for a player regardless of the strategies chosen by other players. A mixed strategy is a plan to randomize over your moves with fixed probabilities, used when always making the same predictable choice would be exploited. Here is how they compare side by side.

Dominant Strategy

In game theory, a dominant strategy is the best course of action for a player in a game, no matter what the opponents do. If a player has a dominant strategy, they will always choose it. Not all games have a dominant strategy for each player.

Mixed Strategy

A pure strategy commits you to one move every time. A mixed strategy assigns a probability to each move, so the opponent cannot predict you and cannot tailor a response. Games where any predictable pattern gets punished, such as a penalty kick or a pitcher choosing pitches, often have no equilibrium in pure strategies at all, and the only stable outcome is both sides randomizing. The equilibrium probabilities have a surprising property: each player mixes in the way that leaves the opponent indifferent between their own options. John Nash proved that every finite game has at least one equilibrium once mixed strategies are allowed, which is why a game that looks unsolvable in pure strategies still has an answer.

Find the rival's equilibrium mix by setting your own expected payoffs equal: EV(your move 1) = EV(your move 2), where EV = the sum of (probability of each rival move × your payoff).

Dominant Strategy vs Mixed Strategy: What to Play When No Single Move Is Always Best

Dominant StrategyMixed Strategy
What the player actually doesPlays one and the same move every timeRandomises across moves with set probabilities
When it is availableOnly when one move beats the alternatives against every rival choiceAlways available, and needed when no single move can be safely predicted
Does the rival's choice matter?No, the move is best whatever the rival doesYes, the probabilities are chosen around the rival's incentives
How you find itCompare your payoffs cell by cell, holding the rival's move fixedSet your probabilities so the rival is left indifferent between their own moves
Cost of being predictableNone, since the rival cannot punish you for itHigh, because a pattern is exactly what a rival exploits
Where it shows upThe prisoner's dilemma, where confessing beats staying silent either wayPenalty kicks, bluffing, tax audits and any game of pursuit

A dominant strategy makes the rival irrelevant; mixing exists because the rival is not

A dominant strategy is one you would still play if the rival announced their move first. Most games do not hand you one. Take matching pennies, payoffs written as (row player, column player). Row wins when the coins match and Column wins when they differ, so heads with heads gives (1, -1), tails with tails gives (1, -1), and either mismatch gives (-1, 1). Check Row's options one rival move at a time. Against Column playing heads, Row earns 1 from heads and -1 from tails, so heads is better. Against Column playing tails, Row earns -1 from heads and 1 from tails, so tails is better. Row's best move flips with the rival's, which is what having no dominant strategy looks like written out. No pair of fixed moves is stable either, because in whichever cell you land the loser can switch and turn the result around. What holds instead is randomising. If Column plays heads with probability q, Row's heads is worth 2q - 1 and Row's tails is worth 1 - 2q, and those two are equal only at q = 0.5. Each player flipping fairly leaves the other with no pattern to attack, and that is the equilibrium of the game.

Being predictable is free in one game and expensive in the other

Suppose Row gets lazy in the coin game and plays heads seven times out of ten. Column notices and plays tails every round, collecting 1 on the seven rounds where Row shows heads and losing 1 on the three where Row shows tails, for a return of 0.7 minus 0.3, or 0.4 a round. Row has handed over a steady loss by being readable. Now run the same test on a prisoner's dilemma, where confessing is dominant. Row could publish that strategy in a newspaper and lose nothing, because no reply the rival can make turns confessing into a mistake. That difference is the practical point of the comparison. It also gives an order of operations for exam questions. Read the /glossary/payoff-matrix and first cross out anything that is a /glossary/dominated-strategy, since a rational player never touches one. If a move then survives for a player whatever the rival does, that player has a dominant strategy and the answer is short. If no such move exists and no pair of fixed choices is stable, the question is pointing at mixing, and the method is always the same: choose probabilities that leave your rival with nothing to gain from either of their own options.

Frequently asked questions

What is the difference between a dominant strategy and a mixed strategy?

A dominant strategy is a single move that beats your alternatives no matter what the other player does, while a mixed strategy is a deliberate randomisation over several moves with chosen probabilities. Dominance means the rival's choice does not affect yours, and mixing is what you fall back on precisely when it does.

Does every game have a dominant strategy?

No, and most do not. Plenty of games give neither player a move that is best against everything the rival might do, which is why the Nash equilibrium idea is needed at all. Once randomisation is allowed, though, every finite game has at least one equilibrium.

Why would a player randomise on purpose?

Because in games where one side wants to match and the other wants to differ, any predictable pattern can be read and exploited by the opponent. Randomising with the right probabilities makes the rival indifferent between their options, which removes any profit they could make from guessing correctly.

Get AP Econ exam tips in your inbox

Occasional emails with study tips, new interactive graphs, and exam-season reminders. Free, no spam.

No spam. Unsubscribe anytime. Read our privacy policy.

Keep track of what you have studied

A free EconLearn account adds progress tracking, your quiz history, and achievements. Studying here is free either way, and there is nothing to pay for as a student.

Create a free account

Already have one? Sign in

Last updated

← Back to the glossary
AP® is a trademark registered by the College Board, which is not affiliated with, and does not endorse, EconLearn.