The Framework
Game theory is the mathematics of strategic interaction — situations where the outcome of your decision depends on the decisions of others. Most real-world conflicts, negotiations, markets, and social dynamics are games in this sense.
The dominant strategy framework provides a systematic method for reasoning through these situations without requiring perfect prediction of opponents' behavior.
Core principle: A dominant strategy is one that produces a better outcome than any alternative, regardless of what the opponent does. If you have one, use it. If your opponent has one, you can predict their action with certainty.
Dominance Types
Strict Dominance
Strategy A strictly dominates Strategy B if A produces a better outcome than B in every possible scenario. When strict dominance exists, the decision is trivial — play the dominant strategy and eliminate the dominated one from analysis.
Weak Dominance
Strategy A weakly dominates Strategy B if A is at least as good as B in every scenario, and strictly better in at least one. Weaker signal, but still eliminates options.
Iterated Elimination
The power of the framework compounds through iteration. Once you eliminate a dominated strategy from your opponent's available actions, their remaining options change — and new domination relationships may emerge.
Iterated elimination of dominated strategies (IEDS) can sometimes solve an entire game, reducing the apparent complexity of dozens of possible action combinations to a single predicted outcome.
Application Protocol
Step 1: Map the game. Define who the players are, what moves are available, and what the payoff structure looks like. In real conflicts, this step is often incomplete — you must estimate opponent payoffs, which is itself a strategic skill.
Step 2: Search for your dominant moves. For each of your available strategies, ask: is there any opponent action against which this strategy performs worse than an alternative? If no, it's dominant. Play it.
Step 3: Eliminate opponent dominated strategies. What moves will your opponent rationally never make? Remove these from your mental model of their behavior. This constrains the space of futures you need to plan for.
Step 4: Iterate. After eliminating, re-examine the reduced game. New dominant and dominated strategies may have become visible.
Step 5: Identify equilibria. Where iterated elimination doesn't fully solve the game, look for Nash equilibria — states where no player can improve their outcome by unilaterally changing strategy. These are stable attractor states in adversarial dynamics.
Practical Constraints
Real-world games differ from formal game theory in important ways.
Incomplete information: You rarely know opponent payoffs precisely. Estimation here is a core skill. Use behavioral signals, history, stated incentives, and structural position to infer what opponents value.
Continuous action spaces: Many real games have infinitely many possible moves, not the discrete 2×2 matrices of textbook examples. The principle still applies, but requires gradient-style reasoning rather than tabular elimination.
Repeated games: One-shot game logic breaks down in long-horizon repeated interactions. Cooperation, reputation, and reciprocity become viable strategies. The folk theorem of repeated games says that almost any outcome can be sustained as an equilibrium if the game repeats indefinitely. Factor this in when operating in ongoing relationships.
Multi-player dynamics: Games with many players introduce coalition formation, signaling, and coordination problems that binary analysis cannot capture. The dominant strategy lens remains useful, but must be supplemented.
The Adversarial Posture
Beyond solving your own strategic problem, this framework enables an adversarial move: manipulating an opponent's perceived payoff matrix to change their dominant strategy.
If you can make cooperation appear to dominate defection in your opponent's perception — through credible commitments, demonstrated capabilities, or altered incentive structures — you shift their rational behavior without defeating them directly.
This is the game-theoretic basis for deterrence, reputation effects, and credible threats.
Common Failure Modes
Assuming symmetric information: Believing the opponent has the same mental model of the game that you do. They don't. Their payoff structure may be entirely different from your inference.
Static analysis in dynamic games: Applying one-shot reasoning to games with evolving structures. The dominant strategy in round one may become dominated in round three.
Ignoring mixed strategies: In some games, the optimal play is to randomize between strategies with specific probabilities. Predictability is a vulnerability — a fully deterministic strategy in an adversarial context can be exploited once the pattern is learned.