Loading field map
Tournament strategy in poker is the body of theory and practice concerned with multi-table tournaments, events in which many players buy in for a fixed stake and are eliminated until one player holds all the chips. Its central problem is that the prize pool is distributed to a small fraction of the field, with payouts rising steeply toward the top. This payout structure, combined with the escalating cost of staying alive, makes tournament play fundamentally different from cash-game play, where chips can be converted directly to money at any moment.
The field studies how a player should adapt decisions to the tournament's unique pressures: the fact that a lost chip costs more than a won chip is worth, the ever-shrinking field, the rising blinds, and the presence of short stacks who are effectively playing a different game from deep stacks. Tournament strategy asks not merely "what is the best play with this hand?" but "what is the best play given my stack, the blinds, the number of players remaining, the payout ladder, and the tendencies of my opponents?"
In a cash game, a chip is a chip; losing $100 and winning $100 are symmetric in value. In a tournament, they are not. If a player has 100,000 chips and the average stack is 50,000, losing half the stack is a disaster far greater than the benefit of doubling it. This asymmetry is the foundation of tournament strategy.
The standard way to formalize this is the Independent Chip Model (ICM), which estimates the dollar value of a player's chip stack given the remaining players' stacks and the payout structure. ICM assumes that each player's chance of finishing in each position is proportional to their share of the total chips in play, adjusted for the fact that when a player is eliminated, their chips are redistributed among the survivors. The model then computes the expected payout for each stack. The key insight is that the value of a chip decreases as a player's stack grows: the first chip is worth more than the millionth chip, because the first chip is what gives the player a chance to survive, while the millionth chip merely adds to an already large stack.
ICM is not a perfect description of tournament reality—it ignores skill differences, position, and the dynamics of play—but it is the standard tool for understanding the endgame of a tournament, when only a few players remain and the payout jumps between finishing positions are large. In this phase, a player with a large stack should often fold strong hands to a short stack's all-in, because eliminating the short stack is worth more to the large stack than winning the pot. Conversely, the short stack should push with a wide range of hands, because the alternative—being blinded away—is worse than risking elimination.
Tournament strategy is usually organized by phase, because the correct approach changes dramatically as the tournament progresses.
Early phase. With deep stacks relative to the blinds, play resembles a cash game, but with one important difference: the blinds will rise on a fixed schedule, so a player cannot simply wait for good hands. The early phase rewards patient, solid play, but a player who plays too passively will find their stack eroding as the blinds increase. The main strategic questions are how to build a stack without risking elimination, and how to exploit opponents who are playing too tightly or too loosely.
Middle phase. As the blinds rise and stacks shrink relative to them, the pressure to act increases. This is where the concept of "fold equity" becomes central. Fold equity is the value of the chance that an opponent will fold to a bet or raise. A player with a short stack can push all-in with a weak hand if the chance that everyone folds is high enough, because the blinds and antes alone may be worth more than the risk of being called. The middle phase is characterized by frequent all-in confrontations, and the skill lies in choosing the right spots to apply pressure and the right hands to call with.
Bubble phase. The "bubble" is the point just before the last few players who will receive prize money are eliminated. On the bubble, the difference between finishing in the money and finishing just outside it is enormous, and this distorts play. Players who are comfortably in the money often play extremely tightly, hoping to outlast the short stacks. Short stacks, meanwhile, are desperate to survive. The bubble is a time of extreme fold equity: a player with a large stack can raise relentlessly, because opponents are reluctant to risk their tournament life. The bubble also creates opportunities for "bubble dynamics," where players collude informally by checking down hands to eliminate a short stack, or where a large stack uses their chip lead to bully the table.
Final table and pay jumps. When the tournament reaches its final table, the payout jumps between positions become enormous—often the difference between first and second place is larger than the difference between second and tenth. This is where ICM becomes most important. Players must weigh the value of moving up the payout ladder against the chance of winning the tournament. A common error is to play for first place when the ICM value of a safer strategy is higher. The final table also introduces the "deal-making" possibility, where remaining players agree to redistribute the prize pool, but this is a negotiation, not a strategic decision within the game itself.
Tournament strategy has developed through several overlapping traditions, each responding to the limitations of the previous one.
The classical school: tight-aggressive play. The earliest systematic tournament advice, popularized in the 1980s and 1990s, emphasized playing few hands but playing them aggressively. The reasoning was that most tournament players made too many loose calls and that patience would be rewarded as weaker players eliminated themselves. This approach worked well in fields of predominantly weak players, but it had a serious flaw: as the blinds rose, a player who waited for premium hands would be blinded away. The classical school's emphasis on avoiding risk also ignored the value of accumulating chips, which is essential for a deep run.
The mathematical revolution: pot odds and expected value. In the late 1990s and 2000s, tournament strategy became more quantitative. Players began to think in terms of expected value (EV), calculating whether a call or raise was profitable in the long run given the pot odds and the range of hands an opponent might hold. This approach was a major advance because it provided a framework for decisions that the classical school treated as matters of intuition or nerve. However, early EV calculations often ignored the tournament-specific issue of chip value; a play that was +EV in chips could be −EV in dollars because of the risk of elimination. The mathematical school needed to be combined with ICM to become fully tournament-aware.
The aggression and pressure school: the squeeze and the resteal. A distinct tradition, associated with the rise of online poker in the 2000s, emphasized the power of aggression itself. The insight was that most players fold too often to raises, so a player who raises frequently can profit even with weak hands. This school developed the "squeeze play" (a raise after another player has raised and been called, designed to force both opponents to fold) and the "resteal" (a re-raise all-in against a frequent raiser, designed to win the pot without a showdown). This approach is not opposed to the mathematical school; rather, it is a recognition that fold equity is a form of value that can be quantified. Its limitation is that it depends on opponents who fold too much; against players who call or re-raise frequently, the aggression school's plays become expensive.
The game-theoretic approach: equilibrium and balance. The most recent major development, driven by the availability of powerful solvers, is the application of game theory. A solver can compute a Nash equilibrium for simplified tournament situations, showing the optimal frequency with which to bet, call, or fold with each hand. The key concept is "balance": a player's strategy should be such that an opponent cannot exploit it. For example, if a player only raises with strong hands, an observant opponent can fold to their raises and call with a wider range; a balanced strategy raises with a mix of strong and weak hands so that the opponent's decision is difficult. The game-theoretic approach has produced many counterintuitive insights, such as the value of "polarized" ranges (betting with very strong or very weak hands, but not medium-strength ones) and the importance of "blockers" (holding a card that reduces the chance an opponent has a strong hand). Its limitation is that solvers can only handle simplified situations, and real tournaments involve many players, changing stacks, and human opponents who do not play equilibrium. The game-theoretic school is best understood as a tool for finding principles, not a complete strategy in itself.
The exploitative approach: adjusting to opponents. In practice, most successful tournament players combine game-theoretic principles with exploitative adjustments. The exploitative approach asks: "Given what I know about this specific opponent, what play maximizes my expected value?" If an opponent folds too often to three-bets, a player should three-bet more; if an opponent calls too loosely, a player should tighten up and value-bet more. The game-theoretic approach provides a baseline—the strategy a player would use against a perfect opponent—and the exploitative approach deviates from that baseline to profit from specific weaknesses. The danger of the exploitative approach is that it can be turned against the player: an opponent who notices the adjustment can exploit it in turn. The two approaches are therefore complementary, not rival.
A persistent theme in tournament strategy is the role of luck. Even a perfect player will lose most tournaments they enter, because the variance is enormous. A player can make the correct decision and still be eliminated when an opponent hits a lucky card. This has two strategic consequences. First, it means that tournament results are a poor measure of skill in the short run; a large sample of tournaments is needed to distinguish a skilled player from a lucky one. Second, it means that tournament strategy must account for the "risk of ruin": a player who takes a slightly negative expected value gamble to double up early may be making a mistake, even if the gamble is close to break-even, because the chance of being eliminated outweighs the benefit of a larger stack. The concept of "survival bias"—the tendency to overvalue aggressive plays that happen to work—is a constant warning in tournament strategy.
Contemporary tournament strategy is a synthesis of these approaches. The game-theoretic school has provided a rigorous foundation, but its insights are filtered through the exploitative lens of real-world play. The mathematical school's EV calculations are now standard, but they are always adjusted for ICM and for the specific dynamics of the table. The classical school's emphasis on patience survives in the early phase, while the aggression school's pressure tactics dominate the middle and bubble phases.
The most important recent development is the widespread availability of training software and solvers, which has raised the general level of play. This has made the game more difficult: the loose, passive opponents of the classical era are rare, and players must be more precise in their decisions. At the same time, the solver revolution has not made tournament strategy a solved problem. Real tournaments involve too many variables—multiple opponents, changing stack sizes, human psychology, and the physical and mental fatigue of long sessions—for a complete solution to exist. The field remains a practical art, grounded in mathematics but requiring judgment, adaptability, and a clear understanding of the tournament's unique structure.