poker MATH APPLIED PROBABILITY INSTITUTE
APPLIED PROBABILITY INSTITUTE // QUANTITATIVE GLOSSARY

Poker Math Glossary

Authoritative mathematical definitions, closed-form formulas, and worked examples for 35 foundational concepts in quantitative poker analysis.

Pot Odds & Equity

(6)

Pot Odds

#
Pot Odds = Call / (Pot + Call)

The ratio of the current size of the pot to the cost of a contemplated call. It represents the mathematical threshold of equity required to make calling a zero expected value (break-even) decision.

Worked Case Example: If the pot is $100 and you face a $50 bet, the pot is now $150. Your call is $50, so your pot odds are $50 / ($150 + $50) = 25%. You need 25% equity to call profitably.

Implied Odds

#

The extension of pot odds considering future betting streets. It accounts for the expected additional chips a player can extract if their drawing hand improves, altering the required break-even equity of the immediate decision.

Worked Case Example: Facing a bet with only 18% equity on a flush draw but getting 25% direct pot odds, a call is profitable if you can extract sufficient additional chips on the river when you hit the flush.

Reverse Implied Odds

#

The mathematical risk of losing additional chips on future streets when a player makes their hand, but it is second-best to an opponent's stronger completed hand. It reduces the effective profitability of drawing.

Worked Case Example: Calling with a low flush draw on a paired board carries high reverse implied odds, as completing the flush risks losing a large pot to a full house.

Equity

#
Equity = (Probability of Winning * Pot) + (Probability of Tying * 0.5 * Pot)

The proportional mathematical expected share of the pot a given hand or range holds, based on the probability of winning or splitting the pot at showdown across all possible runouts.

Worked Case Example: Preflop, pocket Aces (AA) have approximately 81% equity against any two random cards in Texas Hold'em.

Outs

#

The unseen cards remaining in the deck that will improve a drawing hand to a likely winning hand. The Rule of 2 and 4 uses outs to estimate equity.

Worked Case Example: An open-ended straight draw with a flush draw typically possesses 15 outs (8 for the straight, 9 for the flush, minus 2 intersecting cards).

Equity Realization

#
EQR = Realized Equity / Raw Equity

A multiplicative factor applied to raw equity, reflecting the degree to which a hand will under-realize or over-realize its mathematical value due to positional disadvantage, playability, and post-flop fold equity.

Worked Case Example: Out of position, suited connectors may realize 90% of their raw equity, while strong top-pair hands in position may realize 110%.

Expected Value & Decisions

(5)

Expected Value (EV)

#
EV = (Win% * Win$) - (Lose% * Lose$)

The probability-weighted average outcome of a given decision in the long run. Optimal poker decisions mandate selecting lines that strictly maximize EV.

Worked Case Example: A call with a 20% chance to win $100 and an 80% chance to lose $10 has an EV of (0.20 * $100) - (0.80 * $10) = $20 - $8 = +$12.

+EV (Positive Expected Value)

#

A decision framework indicating a mathematical surplus. Over an infinite sample size, taking a +EV action yields theoretical accumulation of chips or capital.

Worked Case Example: Calling an all-in with pocket Aces preflop is a heavily +EV play, even though short-term variance guarantees occasional losses.

-EV (Negative Expected Value)

#

A mathematical deficit in a decision framework. Such actions systematically erode capital and deviate from strategic equilibrium.

Worked Case Example: Chasing a gutshot straight draw facing a large overbet is invariably -EV without immense implied odds.

Fold Equity

#

The additive equity derived from the probability that an aggressive action will induce opponent folds. It transforms the overall expected value equation of bluffing or semi-bluffing.

Worked Case Example: A flush draw might have 35% raw equity, but shoving adds fold equity (e.g., opponent folds 20% of the time), turning a drawing hand into a profitable +EV bet.

Break-Even Percentage

#
BE% = Risk / (Risk + Reward)

The minimum frequency a bluff must succeed, or a call must win, to yield an Expected Value of strictly zero. It governs bluffing thresholds and MDF frameworks.

Worked Case Example: A $50 pot-sized bluff to win a $50 pot requires a break-even success rate of $50 / ($50 + $50) = 50%. If the opponent folds more than 50%, the bluff is profitable.

Kelly, Bankroll & Variance

(9)

Kelly Criterion

#
f* = (bp - q) / b

A mathematical formula used for bet sizing that dictates optimal capital allocation to maximize the long-term compound growth rate of a bankroll while avoiding theoretical ruin.

Worked Case Example: If a player has a 2% true edge with odds representing even-money (1-to-1 payout), full Kelly suggests wagering exactly 2% of the bankroll.

Risk of Ruin

#
RoR = e^(-2 * EV * Bankroll / Variance)

The exact stochastic probability that a poker player will exhaust their entire bankroll before reaching theoretical geometric growth, modeled continuously over infinite trials.

Worked Case Example: A professional with a $10,000 bankroll playing games with high standard deviation might calculate a 5% Risk of Ruin over the coming year.

Variance

#
Var = E[(X - μ)^2]

The statistical measurement of dispersion mapping how far individual short-term results deviate from the long-term mathematical Expected Value (μ). It defines the intrinsic volatility of poker.

Worked Case Example: High variance in multi-table tournaments requires a dramatically larger bankroll compared to low variance cash games to absorb short-term swings.

Standard Deviation

#
σ = √Variance

The square root of variance. It is utilized to construct confidence intervals around win rates, predicting the probable dispersion of bankroll swings over a given sample of hands.

Worked Case Example: A win rate of 5 bb/100 with a standard deviation of 100 bb/100 means 68% of 100-hand samples will fall between -95 and +105 bb.

Bankroll

#

The segregated capital exclusively allocated for playing poker. Proper bankroll management dictates sizing bets mathematically based on expected variance and win rate to avoid ruin.

Worked Case Example: A professional might maintain a $50,000 bankroll to play $2/$5 No-Limit Hold'em, representing a conservative 10,000 big blinds.

Buy-In

#

The capital required to enter a game or tournament. In cash games, it defines the stack depth which mathematically alters optimal play (e.g., implied odds change with effective stacks).

Worked Case Example: The standard maximum buy-in for a cash game is usually 100 big blinds, standardizing SPR calculations on the flop.

ROI (Return on Investment)

#
ROI = ((Return - Investment) / Investment) * 100

A percentage profitability metric used predominantly for tournaments. It measures the gross mathematical yield generated relative to the total entry fees invested.

Worked Case Example: Spending $1,000 in tournament buy-ins and cashing out $1,200 yields an ROI of ($1200 - $1000) / $1000 = 20%.

bb/100

#

The standard empirical unit of cash game win rate, measuring the number of big blinds expected to be won every 100 hands played. Used to project hourly rates and assess true edge.

Worked Case Example: A solid professional in modern online games might sustain a win rate of 4 bb/100 over a sample of 250,000 hands.

Downswing

#

A statistically inevitable prolonged period of negative variance where a player loses capital, despite making theoretically +EV decisions. Modeled via standard deviation and volume.

Worked Case Example: Even a 5 bb/100 winner can mathematically experience a 20 buy-in downswing over a 50,000-hand sample due to normal variance.

Combinatorics & Ranges

(7)

Combinatorics (Combos)

#
C(n,k) = n! / (k!(n-k)!)

The mathematical enumeration of specific card holdings possible in a given situation. There are 1,326 total starting hand combos in Texas Hold'em (e.g., 6 of each pocket pair, 4 of each suited hand).

Worked Case Example: Before the flop, there are 6 distinct ways to be dealt pocket Aces (AsAh, AsAd, AsAc, AhAd, AhAc, AdAc).

Blockers

#

Cards held in a player's own hand or on the board that mathematically eliminate specific combinations from an opponent's presumed range, altering the probabilities of their holdings.

Worked Case Example: Holding the Ace of spades on a three-spade board drastically reduces the combinations of nut flushes in the opponent's range to zero.

Range

#

The comprehensive set of all possible combinations a player might hold in a specific node of the game tree. Strategic analysis relies on range vs. range equity distributions rather than precise hand reading.

Worked Case Example: An Under-the-Gun (UTG) open raising range might consist of the top 15% of all hands, heavily weighted towards pairs and big broadway cards.

Board Texture

#

The macroscopic classification of community cards regarding how they intersect with predefined player ranges. Textures dictate range advantage, polarity, and mathematically optimal bet sizing.

Worked Case Example: A 'wet' texture like 9♠ 8♠ 7♣ heavily interacts with a preflop caller's range of suited connectors, generating massive combinatorial overlap.

Runout

#

The sequential manifestation of turn and river cards. It determines how initial equity distributions shift mathematically over the remaining game tree.

Worked Case Example: A blank runout like 2♦ 2♣ preserves flop equity distributions, while flush-completing runouts aggressively polarize ranges.

Polarized Range

#

A constructed distribution of holdings comprising exclusively very strong value hands or very weak hands (bluffs), eschewing medium-strength marginal holdings.

Worked Case Example: A river overbet range is strictly polarized: it represents either the absolute nuts or complete air to exert maximum ICM and fold equity pressure.

Linear Range

#

A range structure composed from the absolute top combinatorial holdings downward, including strong hands, medium hands, and some draws, without a disconnected bottom tier.

Worked Case Example: A standard UTG preflop opening range is linear, featuring AA at the top all the way down to suited connectors without skipping broadways.

Game Theory & ICM

(8)

GTO (Game Theory Optimal)

#

A mathematically unexploitable baseline strategy governed by Nash Equilibriums. In GTO, bluff-to-value ratios are perfectly balanced, rendering an opponent's counter-strategy indifferent.

Worked Case Example: A GTO solver calculates that an optimal sizing on the river must include precisely 33% bluffs to render a pot-sized bluff-catcher neutral in EV.

Nash Equilibrium

#

A game state where no player can unilaterally increase their expected value by deviating from their current strategy, assuming opponents play optimally.

Worked Case Example: Push/Fold charts for short-stacked tournament play represent calculated Nash Equilibriums where deviations lower absolute tournament equity.

ICM (Independent Chip Model)

#

A mathematical algorithm converting tournament chips into real dollar equity based on payout structures and stack sizes. It dictates that chips lost are always worth more than chips gained.

Worked Case Example: On the bubble of a tournament, ICM dictates folding high-equity hands like AK against big stacks due to the catastrophic financial risk of elimination.

MDF (Minimum Defense Frequency)

#
MDF = Pot / (Pot + Bet Size)

The mathematical threshold of how often a player must call or raise facing a bet to prevent an opponent from profitably bluffing with any two cards.

Worked Case Example: Facing a half-pot bet, the MDF dictates defending exactly 67% of your range, preventing the bettor from auto-profiting on arbitrary bluffs.

Exploitation

#

Strategic deviation from GTO baselines specifically designed to maximize EV against known leaks in an opponent's sub-optimal frequencies.

Worked Case Example: If an opponent folds 80% to 3-bets (exceeding MDF parameters), the exploitative adjustment is to aggressively 3-bet with an expanded, linear bluffing range.

Balance

#

Constructing ranges where strong hands and mathematically appropriate bluffs are paired across all bet sizings and game nodes to remain unexploitable.

Worked Case Example: A balanced river jam range constructs its combinatorics to hold approximately 2 value hands for every 1 bluff, dictating exact indifference for the caller.

Effective Stack

#

The smallest chip stack among active players in a hand. It mathematically limits maximum potential loss or gain and governs all implied odds and SPR metrics.

Worked Case Example: If Player A has $1,000 and Player B has $300, the effective stack is strictly $300.

SPR (Stack-to-Pot Ratio)

#
SPR = Effective Stack / Pot Size

A geometric ratio calculated on the flop that dictates commitment thresholds. Low SPRs automatically commit top-pair holdings, while high SPRs require multi-street nuts.

Worked Case Example: In a $10 pot with $40 effective stacks remaining, the SPR is 4.0, indicating strong but non-nut hands might struggle by the river.
18+ RISK NOTICE