Poker
Poker is a family of gambling card games built on a single skeleton: each player receives private cards, players bet in turn into a common pot, and the pot goes either to the best hand shown at showdown or to the last player left after everyone else folds.
11 frameworks, from 1820 to today · Card Games · Games
- Draw Poker 1820-1900
- Lowball and Split-Pot Variants 1900-Present
- Stud Poker 1900-1980
- Game-Theoretic Optimal Play 1944-Present
- Community Card Poker 1970-Present
- Mixed Games 1970-Present
- Tournament and Cash-Game Strategy 1970-Present
- Psychological and Opponent-Reading School 1984-Present
- Expected-Value Theory 1987-Present
- Exploitative Play 1990-Present
- Computer Poker and Solver Research 2015-Present
Overview
Poker is a family of gambling card games built on a single skeleton: each player receives private cards, players bet in turn into a common pot, and the pot goes either to the best hand shown at showdown or to the last player left after everyone else folds1. What makes the game historically interesting is the tension at its core. Chance deals the cards, but money is won by betting under uncertainty against opponents who are themselves deciding what to do with incomplete information. Every major development in poker's history—new game forms, new betting structures, new theories of how to play—is a response to that tension. This is the story of the games players made, the styles of play they developed, and the ideas that turned poker from table lore into a solved-in-part mathematical contest.
Origins and the Rise of Draw Poker
Poker's ancestry runs through several European gambling games2. The sixteenth-century Spanish game primero and the French poque (itself related to the German Pochspiel) all involved bluffing, betting, and ranking combinations of cards. English settlers brought related games such as brag to North America, and by the early nineteenth century a game called poker—reportedly played in New Orleans around 1820—was spreading up the Mississippi on riverboats and into frontier saloons. The earliest form was a twenty-card deck game, dealt to four players.
The game that took hold, though, was draw poker: each player receives five cards, may discard some and draw replacements, then bets13. Draw poker's innovation was the draw itself—a moment where a player could improve a hand while opponents watched and inferred. This made bluffing and reading central rather than incidental. By the mid-nineteenth century, draw poker with a standard fifty-two-card deck was the dominant American form, and it traveled with soldiers, miners, and gamblers through the Civil War and westward expansion. Its culture produced the enduring figure of the professional gambler and a body of table lore about tells, nerve, and reading opponents—knowledge passed on orally rather than written down.
Stud Poker and the Spread of Variants
Stud poker changed the information structure14. In five-card stud, each player gets one card face down and one face up, with further cards dealt face up one at a time and betting rounds between them. Because some cards are visible to everyone, stud rewards memory and inference: a player can track which cards have appeared and estimate what opponents hold. Seven-card stud, which became widespread in the twentieth century, added more betting rounds and more visible information, and it became the signature game of American casinos and of the World Series of Poker in its early decades.
Alongside these, a cluster of variants developed that changed how pots are won. Lowball games invert hand rankings so the lowest hand wins; split-pot games divide the pot between the best high hand and the best low hand, so a player may win half the pot with a hand that would lose everything in a high-only game15. These variants forced players to think about equity—the share of the pot a hand can expect to claim—rather than simply "best hand wins." Mixed games, in which a table rotates through several variants on a schedule, became a fixture of high-stakes professional play, demanding breadth rather than specialization67.
Community Cards and the Modern Game Family
The most consequential structural change came with community card poker18. In Texas hold 'em, each player receives two private cards and shares five community cards dealt face up in the middle, with betting rounds before and after9. In Omaha, players receive four private cards and must use exactly two with three community cards10. Community cards multiply the ways hands can connect and make the board itself a shared object of inference: everyone sees the same five cards, so the question becomes what each opponent's private cards might be given that board. Hold 'em's rise—from a curiosity in Texas to the main event of the World Series of Poker by the 1970s and the dominant game of the poker boom after 2003—made it the game around which modern strategy was built.
The Psychological School and the Fundamental Theorem
For most of poker's history, strategy was transmitted as craft: play tight, bet strong hands, and read your opponents. The first systematic theorist was David Sklansky, whose The Theory of Poker (1987) framed play around expected value—the average result of a decision over many repetitions—and stated what he called the Fundamental Theorem of Poker: you gain when you play as you would if you could see your opponent's cards, and lose when you don't. This gave players a way to evaluate any decision, not just a hand.
Running alongside this was the psychological and opponent-reading school, which held that the decisive skill is modeling this opponent: identifying tells, spotting betting patterns, and exploiting predictable players. Its classic statement is Mike Caro's Caro's Book of Tells (1984), a catalogue of physical and behavioral cues. The two schools coexisted and often competed: the expected-value approach claimed to be universal and teachable, while the reading school claimed that real money comes from opponents' mistakes, which no formula captures.
Tournaments, Cash Games, and Exploitative Play
Poker's betting structures shaped its strategy as much as its card rules. Cash game play—buying in for a fixed amount and cashing out at any time—rewards long-run expected value and bankroll management. Tournament poker, formalized with the World Series of Poker in 1970, changed the calculus: chips have no cash value, blinds rise, and survival matters, so correct play differs sharply from cash games1112. The distinction between tournament and cash strategy became a basic division of the field.
The dominant practical style that emerged, especially in the professional era, was exploitative play: deviate from any "correct" baseline to attack the specific weaknesses of specific opponents. Exploitation is the natural heir of the reading school, but it now uses expected-value reasoning to decide how to deviate. Its logic is simple: if an opponent folds too often, bluff more; if they call too often, value-bet thinner. The style's weakness is that it depends on accurate reads, and against strong or unknown opponents it can be punished.
Game Theory and Optimal Play
The answer to that weakness came from mathematics. Poker is a game of incomplete information, and game theory offers the concept of an equilibrium: a strategy that cannot be exploited, because it makes the opponent indifferent between their options13. In the 1940s, John von Neumann and Oskar Morgenstern's Theory of Games and Economic Behavior (1944) analyzed a simplified poker model and showed that bluffing could be part of an optimal strategy—a result that surprised readers who thought bluffing was mere deception. Later work extended this: the "fundamental theorem of poker" has a game-theoretic counterpart in the idea that a balanced strategy randomizes between betting and checking so opponents cannot profit from knowing your tendencies.
Game-theoretic optimal (GTO) play entered practical poker through the work of players and theorists in the 2000s and 2010s, and it became a training standard once solvers—software that computes equilibrium strategies for specific situations—became widely available. The practical meaning is concrete: in a given spot, a solver may say to bet a certain size with a certain frequency and to check the rest, mixing hands so that no opponent can exploit the pattern. GTO play is not always the most profitable against weak opponents—exploitation usually earns more—but it is the baseline against which deviations are measured.
Computers and the Solving of Poker
The most dramatic turning point was the arrival of superhuman poker AI. In 2015, a program called Libratus, developed at Carnegie Mellon, defeated a group of top professional players at heads-up no-limit Texas hold 'em over a long match. In 2017, Libratus beat four elite specialists in a 120,000-hand match. Its successor, Pluribus, in 2019, defeated multiple professionals in six-player no-limit hold 'em—a harder problem because multi-player games are not two-player zero-sum in the same clean way. These programs used counterfactual regret minimization, an algorithm that iteratively adjusts strategies to reduce regret and converges toward equilibrium play114.
The consequence for the game was profound. Strategies that humans had believed optimal were shown to be exploitable; the AI's play was strange to professionals precisely because it balanced in ways humans had not imagined. Computer poker research turned equilibrium play from a theoretical ideal into a practical reference, and it accelerated the solver culture that now dominates serious study.
The Present
Today poker is played in three overlapping registers. Cash games and tournaments remain the two great structures, with hold 'em dominant but Omaha, stud, lowball, and mixed games retaining devoted followings6157. Strategy is taught through a synthesis: expected-value reasoning as the frame, exploitative play as the profit engine, and GTO and solver study as the baseline and training method. The open questions are real: how to play multi-way pots well, how to exploit opponents who are themselves solver-trained, and how much of the game's remaining edge lies in psychology rather than mathematics. The history above is the record of how a riverboat gambling game became a subject with its own theory—and then a subject whose theory outgrew its human practitioners.
References
- Poker (introduction, Gameplay, Variants, Artificial intelligence), Wikipedia
- History of poker, Wikipedia
- Draw poker, Wikipedia
- Stud poker, Wikipedia
- High-low split, Wikipedia
- List of poker variants (Mixed poker games), Wikipedia
- HORSE (poker), Wikipedia
- Community card poker, Wikipedia
- Texas hold 'em, Wikipedia
- Omaha hold 'em, Wikipedia
- Poker tournament, Wikipedia
- World Series of Poker (introduction, History), Wikipedia
- Nash equilibrium, Wikipedia
- Libratus (Background), Wikipedia
- Seven-card stud, Wikipedia
The frameworks of Poker
Each framework is a way of working the field has used: when it held, and the question it answers.
- 1820-1900
- 1900-Present
Lowball and Split-Pot Variants
How do low and split pots change strategy?
- 1900-1980
- 1944-Present
- 1970-Present
- 1970-Present
- 1970-Present
Tournament and Cash-Game Strategy
How should tournaments and cash games be played?
- 1984-Present
Psychological and Opponent-Reading School
How do you read an opponent's tells?
- 1987-Present
- 1990-Present
- 2015-Present
How the ideas connect
Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.
Computer Poker and Solver Research grew out of Game-Theoretic Optimal Play
Computer poker programs like Libratus and Pluribus used counterfactual regret minimization to compute equilibrium strategies, turning game-theoretic optimal play from a theoretical ideal into a practical reference. The solvers that followed grew directly out of this line of research.
Carried by Libratus; Pluribus.
Lowball and Split-Pot Variants grew out of Draw Poker
Lowball and split-pot games were built on the draw-poker skeleton, keeping the draw and betting structure but inverting or dividing the pot. The draw itself, which made bluffing and reading central, carried over into these variants.
Mixed Games grew out of Lowball and Split-Pot Variants
Mixed games rotate through several variants, including lowball and split-pot games, on a schedule. The breadth demanded by mixed games grew directly out of the cluster of variant games that had developed earlier.
Community Card Poker grew out of Stud Poker
Community card games like Texas hold 'em combined private cards with shared board cards, extending the stud principle of visible information to a shared object of inference. The board became a common set of face-up cards that all players read, a direct development from stud's visible-card structure.
Exploitative Play rival of Game-Theoretic Optimal Play
Game-theoretic optimal play and exploitative play competed as answers to how one should play: GTO aims to be unexploitable, while exploitation aims to maximize profit against specific opponents. The two schools coexisted and often competed in practical poker strategy.
Exploitative Play grew out of Psychological and Opponent-Reading School
Exploitative play is the natural heir of the psychological and opponent-reading school, using expected-value reasoning to decide how to deviate against specific weaknesses. It kept the reading school's focus on modeling this opponent but added a mathematical framework for choosing deviations.
Exploitative Play influenced by Expected-Value Theory
Exploitative play uses expected-value reasoning to decide how to deviate from a baseline, so the expected-value framework provided the tool for evaluating exploitative adjustments. The style's logic of bluffing more against frequent folders is an application of expected-value calculation.
Carried by David Sklansky; The Theory of Poker (1987).
Beyond the field
Where Poker's ideas came from in other fields, and what they led to there: the strongest 12 of 34 links.
Game-Theoretic Optimal Play led to Solver-Based GTO Study in Poker Strategy
Solver-based GTO study is a direct application of von Neumann and Morgenstern's game-theoretic optimal play to poker, made practical by computers. The solvers compute Nash-equilibrium strategies for poker subgames, so the later practice is a computational branch of the earlier theory.
Carried by John von Neumann; Oskar Morgenstern; Theory of Games and Economic Behavior (1944).
Psychological and Opponent-Reading School influenced by Tell in Poker Strategy
The concept of the tell, popularized by books such as Caro's Book of Poker Tells (1978), became the central object of the psychological school, which built systematic opponent-reading methods around physical and behavioral cues.
Carried by Mike Caro; Caro's Book of Poker Tells (1978).
Expected-Value Theory influenced by Expected Value Theory in Choice Under Uncertainty
Poker's expected-value theory applies the classical expected-value calculus developed from Pascal and Fermat through Huygens and Bernoulli to pot odds, drawing odds and bet sizing.
Carried by Blaise Pascal; Christiaan Huygens.
Expected-Value Theory grew out of Expected Value and Fundamental Theorem School in Poker Strategy
The expected-value framework in modern poker grew directly out of the earlier expected-value and Fundamental Theorem school, which had already framed correct play as maximizing expectation given opponents' cards.
Carried by David Sklansky; The Theory of Poker (1987).
Exploitative Play grew out of Expected Value and Fundamental Theorem School in Poker Strategy
Exploitative play is a direct outgrowth of the expected-value and Fundamental Theorem school, which taught that deviations are correct whenever they exploit opponents' mistakes.
Carried by David Sklansky.
Computer Poker and Solver Research led to Solver-Based GTO Study in Poker Strategy
Solver-based GTO study in poker grew directly out of computer poker research, as equilibrium-finding algorithms were packaged into commercial solvers that players use to study play.
Computer Poker and Solver Research influenced by Game-Theoretic Equilibrium Play in Poker Strategy
Computer poker research rests on game-theoretic equilibrium play, using equilibrium computation and counterfactual regret minimization to solve poker games.
Carried by John von Neumann.
Draw Poker influenced Gambling Lore and Oral Tradition in Poker Strategy
Draw poker, introduced in the United States in the early nineteenth century, supplied the game whose odds, hands and table customs the oral gambling lore of the period described. The lore of professional gamblers and riverboat players grew around this specific game rather than being a separate tradition.
Stud Poker led to Community Card Poker in Poker Variants
Community card poker grew out of stud poker by taking the idea of shared exposed cards further: instead of each player's upcards, a common board is dealt for all to use. The earliest community games were played as stud variants with shared cards.
Game-Theoretic Optimal Play influenced Expected Value and Fundamental Theorem School in Poker Strategy
Game-theoretic optimal play provided the equilibrium and mixed-strategy analysis that the expected-value school in poker built on when computing whether a call or bluff is profitable. The later school treats equilibrium play as a benchmark within its expected-value calculations.
Mixed Games grew out of High-low split in Poker Variants
High-low split games are among the variants regularly rotated in mixed-game formats, so the mixed-game framework grew out of the existing split-pot repertoire. The format combines established variants rather than introducing a new mechanic.
Mixed Games grew out of Stud Poker in Poker Variants
Mixed games assemble existing poker variants into rotating formats, and stud games are among the standard components of those rotations. The format grew out of the established variant repertoire, of which stud was a major part.

Poker is a planet in Nousfarer
Its city has a tower for each of these 11 frameworks. Land, walk in, and climb one concept at a time.
Land on PokerRelated fields
Fields its ideas are linked to
- Poker Strategy 20 links
- Poker Variants 8 links
- Behavioral Decision Theory 1 link
- Trick Taking Games 1 link
- Game Playing Ai 1 link
- Insurance Theory 1 link
- Choice Under Uncertainty 1 link
- Draughts 1 link