Learn it → Apply it → Save as Skill
A stable outcome is not the same as a good outcome. A Nash equilibrium is the point where every player is doing the best they can, given what everyone else is doing — even when a different, better-for-everyone outcome was sitting right there unused.
The Dr. Wallace's Favorite Mental Models library is designed to turn useful ways of thinking into reusable tools. Each model gives you a fast visual overview, deeper interactive learning, practical application exercises and an AI-ready Markdown skill you can save for future use.
Instead of merely reading about mental models, you can gradually build your own AI skills library: a collection of portable reasoning skills that can help ChatGPT, Claude, Grok and other AI assistants apply the mental models you value.
When two or more people are each choosing a strategy, and each one's best choice depends on what the others choose, look for the combination where nobody can do better by switching alone. That combination is what actually tends to happen — whether or not it's the best outcome available.
Notice: both shops end up Low/Low, earning less than they could have at High/High, because neither can move to High alone without losing money.
The model is most useful whenever two or more parties are choosing strategies at the same time, and each party's best choice depends on what the others do.
The method does not tell you what people should do. It tells you what is likely to actually happen once everyone reacts to everyone else.
Identify everyone whose decision genuinely affects the outcome you're trying to predict.
For each player, write down the realistic choices actually available to them, not just the ideal one.
For each combination, ask: holding everyone else fixed, could any single player do better by switching alone?
The combination where no one can unilaterally improve is an equilibrium. Check whether more than one exists.
Open a use case to see what the model changes about how you'd think through it.
Before cutting price, ask what your competitor's best response is. If they will simply match you, you haven't gained share — you've both moved to a lower-revenue equilibrium, as in the coffee-shop example above.
Consider not just what you want, but what response is stable given your manager's real constraints and your own alternatives. A request with no credible alternative behind it rarely shifts the equilibrium.
Feature parity often reflects a stable equilibrium where no competitor can drop the feature without losing customers to the others — not necessarily proof that users need it.
Check whether the deadlock is genuinely stable, or whether one side would move if the other credibly changed its own strategy or walked-away alternative first.
Treat the workaround as a rational, stable response to the current incentives, not a one-off exception. Changing the behavior usually means changing the payoffs, not repeating the rule.
Once enough people adopt a standard, switching alone is costly for any single user, so the group can settle into a stable but not necessarily best format — a coordination equilibrium, not a technical verdict.
This is an illustrative example designed to teach the method, using round numbers.
Expand for the distinction between what Nash actually proved and how the idea is popularly told.
In a 1950 note and a fuller 1951 paper, mathematician John Nash proved that every finite game — a finite number of players, each with a finite number of strategies — has at least one equilibrium point. Reaching that guarantee sometimes requires "mixed strategies," where a player randomizes between choices rather than picking one outright.
No. It is only guaranteed to be stable, not efficient. Both the classic Prisoner's Dilemma and the coffee-shop example above show equilibria that leave every player worse off than a different, non-equilibrium outcome would have.
The 2001 film dramatized Nash's life and introduced the concept to a mass audience. Its famous bar-room explanation is a simplified, and not fully accurate, illustration of the underlying mathematics rather than a textbook definition.
Yes. Many games have multiple Nash equilibria, and a meaningful part of game theory is studying which one a group of players is likely to actually settle into — a coordination problem in its own right.
An equilibrium describes what is stable, not what is desirable. Confusing the two turns a prediction into a justification.
A one-shot game and a repeated relationship can have different stable outcomes. Future consequences can sustain cooperation a single round would not.
A naive best response to today's move can miss that the other side is adjusting to you at the same time.
Real situations often involve incomplete knowledge of the other side's payoffs or constraints, which changes the analysis considerably.
Multiple stable points can exist. Missing one can mean predicting the wrong outcome, or missing a better one worth coordinating toward.
"It's the equilibrium" is not a reason to stop looking for ways to change the rules, payoffs, or players — not just play within them.
the other party's incentives are unknown or shifting quickly. A static equilibrium analysis assumes payoffs you can actually estimate.
the relationship is long-term and cooperation can be sustained by future consequences, not just this round's payoffs.
the stable outcome of unregulated self-interest would cause real harm. The model predicts what tends to happen, not what should happen.
This browser tool structures your thinking. It does not send your information anywhere or perform AI analysis.
Save as Skill converts the model into portable Markdown instructions that teach an AI when and how to apply Nash Equilibrium, what questions to ask, what evidence to seek, what mistakes to avoid and how to structure its analysis.
Save the skills you find useful and organize them into your own reusable library. The Markdown is intentionally model-neutral, so it can be used wherever an AI system accepts uploaded files, project knowledge, custom instructions or reusable text context. Platform capabilities differ, so this is a portable skill file rather than a claim of one-click installation.
The cleanest version assumes so. Real analysis often has to work with estimates or ranges of the other side's payoffs, which is why the model is a starting hypothesis to test, not a guaranteed forecast.
A dominant strategy is a choice that is best regardless of what others do. When every player has one, the combination of dominant strategies is a Nash equilibrium, as in the coffee-shop example. Not every equilibrium involves dominant strategies, though.
Both are interested in predicting stable outcomes among self-interested parties: economists study markets and competitors, negotiators study counterparts and alternatives. The underlying logic of "best response given the other side's move" is shared.