Loading module…
Loading module…
Module 1 of 6
Every previous decision-making framework in this curriculum, however sophisticated, has treated the world as something you act upon — a system of forces, a set of probabilities, a landscape to be navigated. But some of the most important situations you face are not like that. They are situations where the outcome depends not only on what you do but on what others do, and where those others are choosing their actions by reasoning about what you will do — including reasoning about your reasoning about them. When you negotiate, compete, bargain, cooperate, threaten, or promise, you are not acting on a passive world; you are interacting with other minds that are simultaneously acting on you, each side's best move depending on the other's. These are games in the technical sense, and the discipline for reasoning about them is game theory.
Game theory is the mathematics of strategic interaction — of situations where rational actors' outcomes depend on each other's choices. Developed by von Neumann, Morgenstern, Nash, and others, it has become one of the most powerful and widely applied frameworks in economics, biology, political science, and strategy, precisely because so much of life is strategic interaction. But this course is subtitled "for Real-World Operators," and the emphasis is deliberate: the goal is not mathematical elegance but practical power — the ability to look at a real situation of competition, negotiation, or cooperation, recognize its strategic structure, and reason your way to better moves. Game theory's real value to an operator is less in the equations than in the way of seeing: the disciplined habit of thinking through the other side's incentives and anticipated responses before you act, so that forethought — as Buxton's epigraph has it — wins.
By the end of this module you will understand what makes a situation a game and the elements every game has (players, strategies, payoffs, information); the crucial distinction between zero-sum and non-zero-sum games and why it changes everything; the concept of a dominant strategy and why you should never play a dominated one; the Nash equilibrium — the foundational solution concept — and both its power and its limits; and the operator's core habit of reasoning strategically by anticipating others' rational responses. This is the foundation on which the specific games of the rest of the course are built.
Not every decision is a game. When you decide alone against a passive environment — what to eat, how to invest against an impersonal market, whether to carry an umbrella — you are not in a game, because no other mind is choosing its actions in response to yours. A situation becomes a game, in game theory's technical sense, when the outcome depends on the choices of multiple decision-makers, each of whose best choice depends on what the others choose. The defining feature is strategic interdependence: your best move depends on what others do, their best move depends on what you do, and each of you knows this and reasons about the others while choosing. This mutual, reasoning interdependence is what distinguishes a game from a mere decision, and recognizing when you are in one is the first competence, because strategic situations call for strategic reasoning that ordinary decision-making lacks.
Every game has four elements, and identifying them is how you analyze any strategic situation. The first is the players: the decision-makers whose choices affect the outcome. Identifying the players correctly matters — missing a player (a hidden stakeholder, a future entrant, a third party) means missing part of the strategic structure, while including irrelevant parties clutters the analysis. The players are the minds whose choices interact, and the game is defined over them.
The second element is the strategies: the complete set of choices available to each player — the moves each can make. In simple games these are discrete options (cooperate or defect, raise or fold, enter or stay out); in complex ones they are elaborate plans specifying what to do in every contingency. A key subtlety: a strategy in game theory is a complete plan of action, not just a single move — it specifies what the player will do in every situation that might arise, including in response to every possible move by the others. Identifying the real strategy set available to each player — the actual options, not just the obvious ones — is essential, because the whole analysis is about which strategies the players will choose, and missing an available strategy (a creative option, an outside alternative) can flip the analysis.
The third element is the payoffs: the value each player receives from each possible combination of strategies — the outcomes, measured in whatever the players actually care about (money, utility, survival, reputation, satisfaction). Payoffs are what the players are trying to maximize, and they are the engine of the analysis: each player chooses their strategy to get the best payoff given what the others do. A critical point that the rest of the course will develop: payoffs must be measured in the players' true values (their utilities, per the decision-making sector's Module 2), not in naive proxies like money, because players act on what they actually value, and misreading their payoffs — assuming they care only about money when they also value reputation, fairness, or spite — is one of the most common ways game-theoretic analysis goes wrong in the real world.
The fourth element is the information: what each player knows when they choose — about the game's structure, the other players' payoffs, and the other players' moves. Information structure profoundly shapes games: whether players move simultaneously (neither knowing the other's choice) or sequentially (later players seeing earlier moves); whether players know each other's payoffs (complete information) or not (incomplete information); whether past moves are observable. Much of the strategic richness of real situations — and several later modules of this course — comes from information: what you know, what they know, what you know they know, and the games of signaling and inference that asymmetric information creates. Identifying the information structure is essential because the same players, strategies, and payoffs yield completely different games depending on who knows what when.
The practical discipline is to analyze any strategic situation by explicitly identifying these four elements: Who are the players? What strategies does each have? What are the payoffs to each from each combination of strategies (in their true values)? What does each know when they choose? This decomposition turns a vague sense of "a competitive situation" into a structured game that can be reasoned about, and it is the first move of every game-theoretic analysis. Many real strategic errors come from getting one of these elements wrong — missing a player, overlooking a strategy, misreading a payoff, or misjudging the information — so the discipline of laying them out explicitly is itself protective, forcing the completeness that intuition skips.
Key Insight: A situation is a game when the outcome depends on multiple decision-makers whose best choices depend on each other — strategic interdependence, where each reasons about the others while choosing. Every game has four elements: players (the interacting decision-makers), strategies (each player's complete set of available choices/plans), payoffs (the value to each from each combination, measured in their true values not naive proxies), and information (what each knows when choosing — simultaneous vs. sequential, complete vs. incomplete). Analyzing any strategic situation begins with explicitly identifying these four, which forces the completeness intuition skips and prevents the errors of missing a player, strategy, payoff, or information asymmetry.
The single most important distinction in game theory for a real-world operator is between zero-sum and non-zero-sum games, because it determines whether you are in a situation of pure conflict (where the other's loss is your gain and cooperation is impossible) or a situation with room for mutual gain (where cooperation can make both parties better off) — and misreading which one you are in is one of the most common and costly strategic errors, leading people to fight when they should cooperate or to trust when they should compete.
A zero-sum game is one in which the players' interests are purely opposed: whatever one player gains, another loses, so the total across all players is fixed (it sums to zero, or more generally is constant), and there is no possibility of mutual gain. Dividing a fixed pot is zero-sum: every dollar you get is a dollar the other doesn't, so more for you is exactly less for them. Pure competitions for a fixed prize — a single job, a single contract, a chess game (one wins, one loses) — are zero-sum. In a zero-sum game, the interests are perfectly antagonistic, cooperation is impossible (there is nothing to cooperate toward, since any gain to one is a loss to the other), and the correct posture is pure competition: maximize your share, which necessarily minimizes theirs, because that is the only structure the game allows.
A non-zero-sum game is one in which the players' interests are not purely opposed — where the total is not fixed, so outcomes exist that are better (or worse) for everyone, and mutual gain (or mutual loss) is possible. Most real interactions are non-zero-sum: trade (both parties gain from a good deal, or the deal wouldn't happen), most negotiations (there is usually a range of deals better for both than no deal), most business relationships (a healthy partnership creates value both share), and most of social life. In a non-zero-sum game, the total value depends on the players' choices — cooperation can expand the total (making everyone better off) while conflict can shrink it (making everyone worse off) — so the game is not purely about dividing a fixed pot but partly about how large the pot will be, which cooperation and conflict determine. This changes the correct posture fundamentally: pure competition is often self-defeating (it shrinks the pot both share), and the sophisticated play involves cooperation to expand the total alongside competition over the division, holding both at once.
The error of misreading the game type is pervasive and expensive in both directions. Treating a non-zero-sum game as zero-sum — approaching a situation of potential mutual gain as pure conflict — is the more common error, and it destroys value: the negotiator who treats a deal with a large mutually-beneficial zone as a pure fight over a fixed pot poisons the relationship, misses the mutual gains, and often reaches no deal or a worse deal for both, having competed where they should have cooperated to expand the pie before dividing it. This "zero-sum bias" — the tendency to see fixed-pie conflict where mutual gain is possible — is a documented and damaging habit of mind, and it wrecks partnerships, negotiations, and relationships by turning positive-sum situations into fights. The opposite error — treating a zero-sum game as non-zero-sum, seeking cooperation and mutual gain where the interests are genuinely purely opposed — is less common but also costly: the person who seeks a win-win in a genuinely zero-sum competition (a fixed single prize only one can have) is naive, and will be beaten by an opponent who correctly recognizes the pure conflict and competes accordingly, while they waste effort seeking a cooperation the game does not permit.
The practical discipline is to diagnose the game type before choosing your posture: is this situation zero-sum (purely opposed interests, fixed total, cooperation impossible) or non-zero-sum (interests not purely opposed, total variable, mutual gain possible)? And crucially, recognize that most real situations are non-zero-sum and often have a larger cooperative potential than the parties perceive, so the default suspicion should be that mutual gains exist and the zero-sum framing is an error — while remaining alert to the genuinely zero-sum situations where cooperation is impossible and pure competition is correct. Many situations are also mixed — partly cooperative (both gain from a deal existing) and partly competitive (they compete over the terms) — and the sophisticated operator holds both aspects at once, cooperating to expand the total while competing over the division, which is exactly the structure of most negotiations (Module 6). Getting the game type right is the master diagnostic, because it determines whether your fundamental posture should be competition, cooperation, or the mixed dance of both.
Key Insight: The master distinction is zero-sum (interests purely opposed, total fixed, one's gain is another's loss, cooperation impossible — correct posture: pure competition) versus non-zero-sum (interests not purely opposed, total variable, mutual gain possible — cooperation can expand the pot, conflict can shrink it). Most real interactions are non-zero-sum, and the pervasive costly error is zero-sum bias: treating positive-sum situations (trade, negotiation, partnership) as fixed-pie fights, destroying the mutual gains and often the relationship. Diagnose the game type before choosing your posture — default to suspecting mutual gains exist, while staying alert to genuinely zero-sum pure conflicts — because most situations are mixed, cooperative over the pot's size and competitive over its division at once.
The first analytical tool for solving a game — figuring out what rational players will do — is the concept of dominance, which identifies strategies that are unambiguously good or bad regardless of what the others do. Dominance is the most straightforward game-theoretic reasoning, and it yields two rules that are simple, powerful, and often violated: always play a dominant strategy if you have one, and never play a dominated strategy.
A dominant strategy is one that is your best choice regardless of what the other players do — it produces a higher payoff than any of your other strategies against every possible combination of the others' choices. If you have a dominant strategy, the analysis is easy: play it, because it is best no matter what anyone else does, so you don't even need to predict the others' moves — whatever they do, this strategy serves you best. Dominant strategies are rare (most games don't offer one), but when they exist, they are decisive: a rational player always plays their dominant strategy, and if you have one, you should play it without agonizing over what the others will do, because their choices don't change your best move.
A dominated strategy is the opposite: one that is worse than some other strategy of yours regardless of what the others do — there is another strategy that beats it against every possible combination of the others' choices. The rule for dominated strategies is equally clear: never play a dominated strategy, because you always have another strategy that does at least as well and sometimes better, no matter what the others do, so playing the dominated one is simply leaving value on the table for no possible benefit. Playing a dominated strategy is a clear error, and identifying and eliminating your dominated strategies is the first step of playing any game well — you never want to choose an option that is beaten by another of your options in every scenario.
The power of dominance extends beyond your own choices to predicting others' choices and simplifying the game through iterated elimination of dominated strategies. Since rational players never play dominated strategies, you can eliminate the others' dominated strategies from consideration — they won't play those — which may then make some of your strategies dominated (given that the others won't play their eliminated options), which you then eliminate, which may make more of the others' strategies dominated, and so on. This iterated elimination can dramatically simplify a game, sometimes solving it entirely (reducing it to a single strategy for each player) by repeatedly removing options that no rational player would choose. The reasoning is: I won't play my dominated strategies; you won't play yours; given that you won't play yours, some of mine that weren't dominated before now are, so I eliminate those too; given that I won't play those, more of yours become dominated; and the elimination cascades until only the strategies that survive remain. This is a powerful way to reason about what rational players will do, and it embodies the core game-theoretic move of reasoning about others' rationality (Section 5).
The practical value of dominance is both direct and cautionary. Directly, it gives you two firm rules — play dominant strategies, never play dominated ones — that resolve many strategic choices cleanly and catch a common class of errors (people do play dominated strategies, out of confusion, emotion, or failure to think it through). But there is a cautionary note that connects to the whole course's realism: dominance analysis assumes players are rational and have the payoffs you think they have, and in the real world both assumptions can fail — a player may not be rational, or their true payoffs (Section 1) may differ from what you assumed (they value spite, or fairness, or reputation in ways that change which strategy is actually dominant for them). So while "never play a dominated strategy" is a sound rule for your play (given your true payoffs), predicting others by assuming they'll avoid dominated strategies requires that you have their payoffs and rationality right — and the sophisticated operator uses dominance analysis while remaining alert to the possibility that the other player's real payoffs or rationality differ from the model, which would change what is dominated for them. Dominance is the first and cleanest tool, but like all the tools, it is only as good as the payoffs and rationality assumptions it rests on.
Key Insight: Dominance identifies unambiguously good or bad strategies: a dominant strategy is best regardless of what others do (play it without needing to predict them), and a dominated strategy is worse than another of your options regardless of what others do (never play it — it leaves value on the table for no benefit). Iterated elimination of dominated strategies — since rational players avoid dominated strategies, eliminate theirs, which may make more of yours dominated, cascading — can dramatically simplify or solve a game. The rules are firm for your own play given your true payoffs, but predicting others by dominance requires having their payoffs and rationality right — so use it while staying alert that their real payoffs (spite, fairness, reputation) or rationality may differ from your model.
When a game has no dominant strategies and dominance elimination doesn't solve it, the central tool for predicting what rational players will do is the Nash equilibrium, the most important solution concept in game theory, for which John Nash won the Nobel Prize. Understanding it — both its power and its limits — is essential, because it is the workhorse concept that the analysis of most games relies on, and because misunderstanding it (treating it as a prescription for the best outcome, which it often is not) leads to serious errors.
A Nash equilibrium is a combination of strategies — one for each player — such that no player can do better by unilaterally changing their own strategy, given what the others are doing. In other words, at a Nash equilibrium, each player's strategy is a best response to the others' strategies, so no one has an incentive to deviate: given what everyone else is doing, you are already doing the best you can, and so is everyone else. It is an equilibrium precisely because it is self-stable — no one wants to move, given what the others are doing — so if the players reach it, they stay there, each held in place by the fact that deviating would only make them worse off given the others' choices.
The concept's power is that it predicts the stable outcomes of a game — the combinations of strategies that, once reached, no one has reason to abandon. To find where a game will settle, look for its Nash equilibria: the strategy combinations where each player is best-responding to the others. This is a powerful predictive tool because rational players, reasoning about each other, tend toward these self-stable points — any non-equilibrium combination has at least one player who could do better by deviating, so it is unstable, while an equilibrium holds. The Nash equilibrium is thus the answer to "where will this game settle?" for a wide class of games, and computing it (finding the strategy combinations where everyone is best-responding) is the core technique of game-theoretic analysis.
But the Nash equilibrium has crucial properties that the operator must understand, because misunderstanding them causes errors. First, a Nash equilibrium is not necessarily good — it is stable, not optimal. The equilibrium is where the game settles because no one can improve by unilateral deviation, but the outcome can be bad for everyone, worse than some other outcome they could reach if they jointly changed — the individually-stable point can be collectively terrible. The next module's Prisoner's Dilemma is exactly this: its Nash equilibrium (both defect) is worse for both players than the non-equilibrium outcome (both cooperate), yet it is where rational individual players land, because each is best-responding. So the Nash equilibrium tells you where the game settles, not where the players would be best off — and confusing "stable" with "good" is a fundamental error. Second, a game can have multiple Nash equilibria, and then the concept alone doesn't tell you which one occurs — additional reasoning (focal points, dynamics, coordination — Module 3) is needed to predict which equilibrium the players reach. Third, equilibria can be in mixed strategies — where players randomize over their options with certain probabilities — which is the equilibrium concept extended to games (like matching pennies or rock-paper-scissors) where any predictable pure strategy would be exploited, so the stable play is to randomize unpredictably.
The practical use of the Nash equilibrium for an operator is twofold. As a predictive tool, it tells you where a strategic situation will likely settle — find the equilibria, and you know the stable outcomes toward which rational play tends, which lets you anticipate how a competition or negotiation will resolve if everyone plays rationally. As a diagnostic tool, recognizing that a game's equilibrium is bad for everyone (as in the Prisoner's Dilemma) tells you that you are in a situation where individual rationality produces collective harm, and that escaping the bad equilibrium requires changing the game — altering payoffs, adding enforcement, building trust, or changing the information — rather than just playing it better, because playing the game as given lands everyone at the bad equilibrium. This diagnostic use is where much of game theory's real-world power lies: it identifies when you are trapped in a bad equilibrium and points toward changing the game's structure as the escape, which the rest of the course develops. The Nash equilibrium is the foundational concept, but its deepest lesson is that stable is not the same as good, and that the operator's power often lies in changing the game so its equilibrium becomes one they want.
Key Insight: A Nash equilibrium is a combination of strategies where no player can do better by unilaterally deviating — each is best-responding to the others, so it is self-stable and predicts where a game settles. But its crucial properties prevent errors: it is stable, not optimal (a game's equilibrium can be bad for everyone, worse than an outcome they'd reach by jointly deviating — the Prisoner's Dilemma), games can have multiple equilibria (needing further reasoning to predict which occurs), and equilibria can require mixed (randomized) strategies. Its power is predictive (find the equilibria to anticipate where rational play settles) and diagnostic (recognizing a bad equilibrium tells you to escape by changing the game — payoffs, enforcement, trust, information — not by playing it better).
Underlying all the tools is a habit of mind that is game theory's deepest gift to the operator: the discipline of strategic reasoning — thinking through the other players' incentives and anticipated responses before you act, and specifically reasoning at least one level deeper than the naive actor who considers only their own move. This habit, more than any equation, is what game theory trains, and cultivating it transforms how you approach every interactive situation.
The core of strategic reasoning is anticipating the other side's rational response to your move, and choosing your move in light of that anticipated response. The naive actor asks only "what is my best move?" considering the situation as it is now; the strategic actor asks "what is my best move, given how the other side will rationally respond to it?" — recognizing that their move will provoke a response, that the response will change the situation, and that the right move is the one that leads to the best outcome after the other side responds, not the one that looks best before they respond. This is the essence of forethought (Buxton's epigraph): looking ahead to the other side's reaction and choosing accordingly, rather than acting on the immediate situation and being surprised by the reaction. The chess player who considers only their own move and not the opponent's reply loses to the one who thinks "if I move here, they'll respond there, and then I'll be worse off, so I should move elsewhere" — and life's strategic situations reward the same forward-reasoning discipline.
This reasoning proceeds through levels, and the operator must calibrate how many levels deep to think. Level zero is not reasoning strategically at all — acting on impulse or on the immediate situation. Level one is anticipating the others' response to your move ("if I do X, they'll do Y"). Level two is anticipating their anticipation of your move ("they'll expect me to do X, so they'll prepare Y, so I should do Z"). And so on — each level reasons about the other's reasoning about your reasoning. In principle this regresses infinitely (I think about what you think about what I think...), and the Nash equilibrium is in a sense the fixed point where the levels stabilize (each correctly anticipating the other). But in practice, real players reason to limited depth, and a documented finding is that most people reason only one or two levels deep — so the operator who reasons one level deeper than their counterpart gains an edge, anticipating a response the counterpart didn't anticipate you'd anticipate. The practical calibration is to reason deeply enough to stay one level ahead of your actual counterpart (not infinitely deep, which is both impossible and unnecessary against limited-depth opponents), which means gauging how strategically sophisticated the other side actually is and thinking one level beyond them.
Strategic reasoning also includes the crucial recognition that you can influence the game, not just play it — that your moves change the other side's incentives and therefore their responses, and that the most powerful strategic moves are often those that change the game's structure in your favor rather than just playing the existing game well. This is the deepest level of strategic reasoning: not "what is my best move in this game?" but "how can I change this game — the payoffs, the information, the sequence, the options — so that its rational outcome is one I want?" Committing publicly to a course of action (changing your own payoffs so a threat or promise becomes credible), sharing or concealing information (changing what the other knows), moving first or waiting (changing the sequence), or creating new options (changing the strategy sets) are all moves that alter the game itself, and they are frequently more powerful than any move within the fixed game, because they change the equilibrium rather than just navigating to it. The rest of the course develops these game-changing moves — credible commitments, signaling, mechanism design — and they all flow from the strategic-reasoning insight that the operator is not merely a player in a fixed game but an actor who can reshape the game.
The habit of strategic reasoning is game theory's most portable and valuable gift, because it applies to every interactive situation whether or not you ever compute an equilibrium. Before any negotiation, competition, or interaction, the strategic reasoner asks: Who are the players and what do they really want (their true payoffs)? How will they rationally respond to what I'm considering? What are they anticipating from me? How can I change the game to my advantage rather than just playing it? This disciplined anticipation of others' incentives and responses — thinking one level deeper than the naive actor and one level deeper than your actual counterpart — is what "thinking strategically" means, and it is the foundation on which all the specific game analyses of this course rest. Master the habit, and you gain the forethought that wins.
Key Insight: Game theory's deepest gift is the habit of strategic reasoning — anticipating the other side's rational response to your move and choosing in light of it, rather than acting on the immediate situation and being surprised by the reaction. It proceeds through levels (anticipating their response, their anticipation of your move, and so on); since most people reason only one or two levels deep, the operator who reasons one level deeper than their actual counterpart gains an edge — calibrated to stay just ahead, not infinitely deep. Deepest of all: recognizing you can change the game — its payoffs, information, sequence, options — so its rational outcome favors you, which is often more powerful than any move within the fixed game. This portable habit applies to every interaction whether or not you compute an equilibrium.
Consolidating the module into a usable capability: you should now be able to take a real strategic situation, recognize it as a game, analyze its structure, and reason toward better moves using the fundamental concepts. This closing section assembles the fundamentals into a working method for the real-world operator and traces it through an example, so the concepts become a practical approach to strategic situations rather than abstract theory.
The operator's method for a strategic situation. (1) Recognize the game: notice that the outcome depends on multiple decision-makers whose choices interact, so strategic (not just individual-decision) reasoning is required. (2) Identify the four elements: Who are the players (including hidden or future ones)? What strategies does each actually have (including creative options)? What are the payoffs to each, in their true values (money, but also reputation, fairness, spite, relationship)? What does each know when they choose (the information structure)? (3) Diagnose the game type: is this zero-sum (pure conflict, compete) or non-zero-sum (mutual gain possible, cooperate to expand the pot while competing over division), and recognize that most situations are non-zero-sum and often more cooperative than they appear (beware zero-sum bias). (4) Apply dominance: identify your dominant strategy (play it) and dominated strategies (never play them), and eliminate the others' dominated strategies to simplify and predict. (5) Find the equilibria: where does the game settle (each player best-responding)? — and critically, is the equilibrium good or bad for you, and are there multiple equilibria? (6) Reason strategically: anticipate the others' rational responses, think one level deeper than your counterpart, and — most powerfully — ask how you can change the game (commitments, information, sequence, options) so its equilibrium favors you. This method turns a vague strategic situation into a structured analysis pointing toward better moves.
Tracing the method through an example. Two competing firms are deciding whether to cut prices to gain market share. Recognize the game: each firm's best move depends on the other's, so it's a game. Four elements: players (the two firms, and perhaps potential entrants and customers); strategies (cut price or hold); payoffs (profit, but also long-term market position and the true value each places on share versus margin); information (do they move simultaneously or can one observe the other? do they know each other's costs?). Diagnose the type: this is non-zero-sum — if both hold prices, both keep healthy margins (mutual gain), while if both cut, both lose margin (mutual loss), so it is not a fixed pie but a situation where cooperation (both holding) expands the total both share. Apply dominance: often, "cut price" is individually tempting regardless of the other (if they hold, I cut and steal share; if they cut, I must cut to not lose share) — making "cut" a dominant strategy for each, which is the Prisoner's Dilemma structure of the next module. Find the equilibrium: both cut (each best-responding), which is the Nash equilibrium — and recognize it is bad for both, worse than both holding, a bad equilibrium that individual rationality produces. Reason strategically: since playing the game as given lands both at the bad equilibrium (a price war), the powerful move is to change the game — build the conditions (repeated interaction, credible signals of restraint, differentiation that reduces direct competition) that transform the one-shot Prisoner's Dilemma into a repeated game where cooperation (both holding) can be sustained (Module 2), or differentiate so the firms aren't in a pure price competition at all (the positioning of the strategy sector). The method reveals that the naive move (cut to win share) leads to a mutually destructive equilibrium, and that the real strategic leverage is in changing the game to escape it — exactly the kind of insight game theory provides that naive competition misses.
The foundation for what follows. This module has built the fundamentals — games and their elements, zero-sum versus non-zero-sum, dominance, Nash equilibrium, and strategic reasoning — that the rest of the course applies to specific, important classes of games. The Prisoner's Dilemma and the problem of cooperation (Module 2), coordination games (Module 3), signaling and information games (Module 4), auctions and mechanism design (Module 5), and the application to business and negotiation (Module 6) are each a specific strategic structure that recurs constantly in real life, analyzed with the fundamentals built here. The through-line is the operator's orientation: game theory's value is not the mathematics but the way of seeing — recognizing strategic situations, analyzing their structure, anticipating others' rational responses, and, most powerfully, changing the game so its outcome favors you. Carry the fundamentals forward as the lens through which the specific games become clear, and remember the deepest lesson: the skilled operator is not just a good player of games but a shaper of them, and forethought — the disciplined anticipation of how strategic situations will unfold — is what wins.
Key Insight: The operator's method for any strategic situation: recognize the game, identify the four elements (players, strategies, payoffs in true values, information), diagnose the type (zero-sum vs. non-zero-sum, beware zero-sum bias), apply dominance (play dominant, never dominated, eliminate to simplify), find the equilibria (and judge whether they're good or bad for you), and reason strategically (anticipate responses, think one level deeper, and change the game so its equilibrium favors you). Traced through a price-war example, it reveals that the naive move lands both firms at a mutually destructive equilibrium and that the real leverage is changing the game to escape it — the insight naive competition misses. Game theory's value is the way of seeing, and the skilled operator is not just a player of games but a shaper of them.
This module built the fundamentals of game theory for the real-world operator: what makes a situation a game and its four elements (players, strategies, payoffs, information); the master distinction between zero-sum and non-zero-sum games; the tool of dominance and the rules to play dominant and never dominated strategies; the Nash equilibrium as the foundational solution concept, stable but not necessarily good; and the portable habit of strategic reasoning — anticipating others' responses, thinking one level deeper, and changing the game rather than just playing it. You can now recognize and analyze strategic situations, which every subsequent module applies to a specific important class of games.
The next module takes up the most famous and consequential game of all: the Prisoner's Dilemma, the game whose Nash equilibrium is bad for everyone, which captures the fundamental tension between individual and collective rationality that pervades business, politics, and life. Prisoner's Dilemma & Cooperation develops why individually rational players reach a mutually destructive outcome, and — crucially — how cooperation can nonetheless emerge and be sustained, especially through repetition, reputation, and the strategies (like tit-for-tat) that make cooperation the rational choice even among self-interested players. Where this module gave you the tools to analyze any game, the next module uses them on the single most important game there is, and shows how to escape the bad equilibrium that individual rationality otherwise traps everyone in.