--- title: "Stationary Strategies | Game Theory" description: "Stationary strategies are choices that stay the same over time in Game Theory, helping you analyze repeated play, Markov decisions, and stable behavior." canonical: "https://fiveable.me/game-theory/key-terms/stationary-strategies" type: "key-term" subject: "Game Theory" unit: "Unit 13" --- # Stationary Strategies | Game Theory ## Definition Stationary strategies are game theory strategies that do not change with time or past play. In this course, they show up when you model repeated decisions with a fixed rule based on the current situation. ## What It Is Stationary strategies are strategies in Game Theory that stay the same instead of changing from round to round. If you use one, your action rule does not depend on the earlier history of the game, only on the present situation or state. That makes them useful in repeated games and dynamic settings where the full history could get messy. Instead of asking, "What happened three turns ago?" you ask, "What is the current state, and what does my rule say to do now?" This keeps the analysis cleaner and is a big reason stationary strategies show up in models of long-run behavior. A simple way to think about it is this: a stationary strategy is like following one steady policy. For example, if a player always chooses the same action when the game reaches a given state, that is stationary. If the player changes behavior based on who moved first, what happened last round, or how many rounds have passed, that is not stationary. In game theory, stationary strategies are especially common when the environment is stable and future play looks similar from one round to the next. That is why they connect naturally to Markov decision processes, where the next decision depends on the current state rather than the whole past. They also fit well with reinforcement learning, where an agent tries to learn a policy that works repeatedly across similar situations. Stationary does not mean random or weak. A stationary strategy can still be smart, optimal, or hard to beat. It just means the logic behind the choice does not keep changing with the game history. In zero-sum settings, that can make patterns easier to study, because once a player settles into a steady rule, opponents can sometimes predict and respond to it. The main idea is stability. Stationary strategies give you a way to model players who stick to a rule over time, which is useful when you want to analyze long-run outcomes instead of one-off moves. ## Why It Matters Stationary strategies matter because a lot of game theory is really about repeated choice, not just one isolated move. If the situation unfolds over time, a fixed strategy can let you study whether behavior settles into a pattern, whether outcomes stabilize, and whether one player can exploit another player’s predictability. This term is also a bridge into machine learning approaches to game-theoretic problems. In reinforcement learning, an agent often tries to find a policy that works across many rounds, and a stationary strategy is the cleanest version of that kind of policy. Instead of memorizing every past move, the agent learns what to do from the current state. Stationary strategies also help you compare theoretical models to real strategic behavior. In economics, politics, and competitive systems, players often do not react with perfect memory to every past event. A stationary rule can be a realistic approximation when the setting is stable enough that current conditions matter more than long history. For class work, this term lets you explain why some games are easier to solve with dynamic methods and why others can be reduced to a simpler policy problem. It gives you language for talking about consistency, long-run optimization, and why a repeated game can still have a clear solution even when the number of possible histories grows fast. ## Connections ### Nash Equilibrium A stationary strategy can be part of an equilibrium, but it is not the same thing as one. Nash equilibrium is about no player wanting to deviate given the others' choices, while stationary describes how a strategy is structured over time. A stationary strategy may support equilibrium analysis in repeated or dynamic games. ### Mixed Strategy Mixed strategies use probabilities to randomize actions, while stationary strategies describe whether the rule changes across time or history. You can have a stationary mixed strategy, meaning the probabilities stay the same in a given state. The two ideas often work together in dynamic game models. ### [best response dynamics](/game-theory/key-terms/best-response-dynamics) Best response dynamics tracks how players update choices in reaction to each other. Stationary strategies sit on the other side of that idea, because they do not keep updating based on the full past. If best response dynamics settle down, the process may end up near a stationary pattern. ### [convergence properties](/game-theory/key-terms/convergence-properties) Convergence properties ask whether repeated play or learning eventually settles into a stable outcome. Stationary strategies are a natural target when you want convergence, because a fixed policy can make long-run behavior easier to analyze. If the process does not converge, a stationary strategy may still be a useful benchmark. ## On the AP Exam A problem set question may ask you to decide whether a policy is stationary by checking if it depends on past moves or only on the current state. In a repeated-game or reinforcement-learning setting, you might explain why a fixed rule is easier to analyze than a history-dependent one. You may also be asked to compare a stationary strategy with a mixed strategy or with an adaptive best-response rule, then describe what happens to payoffs over time. If the course uses matrices or state diagrams, look for whether the action rule changes across stages. That is usually the signal that the strategy is not stationary. ## Stationary Strategies vs Mixed Strategy People mix these up because both can involve repeating the same action pattern over time. A mixed strategy is about randomizing among actions with set probabilities, while a stationary strategy is about whether the rule stays the same across time or past history. A strategy can be both stationary and mixed. ## Key Takeaways - Stationary strategies keep the same decision rule over time instead of changing with past play. - They are useful when you want to analyze repeated games, stable environments, or long-run behavior. - A stationary strategy can still be optimal, even though it does not adapt to every earlier move. - These strategies connect closely to Markov decision processes and reinforcement learning, where the current state matters most. - If a player changes behavior based on history, stage number, or earlier rounds, the strategy is not stationary. ## FAQs ### What is stationary strategies in Game Theory? Stationary strategies are strategies that do not change over time or depend on the history of play. In Game Theory, they are often used in repeated or dynamic games where the current state is enough to decide what to do next. The idea is to model a fixed policy instead of a memory-heavy rule. ### How are stationary strategies different from mixed strategies? A mixed strategy randomizes over possible actions using probabilities. A stationary strategy says the rule stays the same across time or past history. You can actually have a stationary mixed strategy if the probabilities stay fixed in the same state, so the two ideas can overlap. ### Why do stationary strategies matter in repeated games? Repeated games create lots of possible histories, which can make analysis messy. A stationary strategy simplifies that by using the same rule each time, or the same rule for each state. That makes it easier to study long-run outcomes, predict behavior, and compare strategies. ### How do stationary strategies show up in machine learning? In reinforcement learning, an agent often learns a policy that maps the current state to an action. If that policy does not depend on the full past, it is stationary. That is one reason stationary strategies connect so well to Markov decision processes and algorithmic decision-making. ## Related Study Guides - [13.3 Machine learning approaches to game-theoretic problems](/game-theory/unit-13/machine-learning-approaches-game-theoretic-problems/study-guide/XAYTjevapksB2y7n) ## About This Document Canonical Fiveable pages are available as Markdown at the same path plus `.md`. - [llms.txt](https://fiveable.me/llms.txt): index of Fiveable's sections and URL patterns - [llms-full.txt](https://fiveable.me/llms-full.txt): complete subject and unit listing - [MCP server](https://fiveable.me/mcp): call Fiveable as tools instead of fetching pages (`https://fiveable.me/api/mcp`) - [MCP server for AP teachers](https://fiveable.me/mcp/teachers): a teacher's classes, assignments and AP-rubric grading (`https://fiveable.me/api/mcp/teacher`) ## Structured Data ```json {"@context":"https://schema.org","@graph":[{"@type":"LearningResource","@id":"https://fiveable.me/game-theory/key-terms/stationary-strategies#resource","name":"Stationary Strategies | Game Theory","url":"https://fiveable.me/game-theory/key-terms/stationary-strategies","learningResourceType":"Concept explainer","educationalLevel":"AP® / High School","about":{"@id":"https://fiveable.me/game-theory/key-terms/stationary-strategies#term"},"audience":{"@type":"EducationalAudience","educationalRole":"student"},"dateModified":"2026-07-03T02:21:36.475Z","isPartOf":{"@type":"Collection","name":"Game Theory Key Terms","url":"https://fiveable.me/game-theory/key-terms"},"publisher":{"@type":"Organization","name":"Fiveable","url":"https://fiveable.me"}},{"@type":"DefinedTerm","@id":"https://fiveable.me/game-theory/key-terms/stationary-strategies#term","name":"Stationary Strategies","description":"Stationary strategies are game theory strategies that do not change with time or past play. In this course, they show up when you model repeated decisions with a fixed rule based on the current situation.","url":"https://fiveable.me/game-theory/key-terms/stationary-strategies","inDefinedTermSet":{"@type":"DefinedTermSet","name":"Game Theory Key Terms","url":"https://fiveable.me/game-theory/key-terms"}},{"@type":"FAQPage","mainEntity":[{"@type":"Question","name":"What is stationary strategies in Game Theory?","acceptedAnswer":{"@type":"Answer","text":"Stationary strategies are strategies that do not change over time or depend on the history of play. In Game Theory, they are often used in repeated or dynamic games where the current state is enough to decide what to do next. The idea is to model a fixed policy instead of a memory-heavy rule."}},{"@type":"Question","name":"How are stationary strategies different from mixed strategies?","acceptedAnswer":{"@type":"Answer","text":"A mixed strategy randomizes over possible actions using probabilities. A stationary strategy says the rule stays the same across time or past history. You can actually have a stationary mixed strategy if the probabilities stay fixed in the same state, so the two ideas can overlap."}},{"@type":"Question","name":"Why do stationary strategies matter in repeated games?","acceptedAnswer":{"@type":"Answer","text":"Repeated games create lots of possible histories, which can make analysis messy. A stationary strategy simplifies that by using the same rule each time, or the same rule for each state. That makes it easier to study long-run outcomes, predict behavior, and compare strategies."}},{"@type":"Question","name":"How do stationary strategies show up in machine learning?","acceptedAnswer":{"@type":"Answer","text":"In reinforcement learning, an agent often learns a policy that maps the current state to an action. If that policy does not depend on the full past, it is stationary. That is one reason stationary strategies connect so well to Markov decision processes and algorithmic decision-making."}}]},{"@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Game Theory","item":"https://fiveable.me/game-theory"},{"@type":"ListItem","position":2,"name":"Key Terms","item":"https://fiveable.me/game-theory/key-terms"},{"@type":"ListItem","position":3,"name":"Unit 13","item":"https://fiveable.me/game-theory/unit-13"},{"@type":"ListItem","position":4,"name":"Stationary Strategies"}]}]} ```