The Mathematics of Poker (Probability and EV)

Poker looks like a card game and turns out to be a probability engine. The mathematics of poker — the hand-frequency tables, the pot-odds equation, the expected-value calculation that drives every long-term winner — is what separates the game’s lucky-night players from the people who pay rent with it. The math isn’t hidden. It’s just easier to feel the cards than to count them, and feel loses to count over enough hands.

The hand frequency table

The 5-card draw probability table is the bedrock of poker math. From a 52-card deck, there are 2,598,960 possible 5-card hands (the combination C(52,5)). Each hand category has a frequency that’s a hard mathematical fact:

• Royal Flush: 4 hands — 0.000154%
• Straight Flush (non-royal): 36 — 0.00139%
• Four of a Kind: 624 — 0.024%
• Full House: 3,744 — 0.144%
• Flush: 5,108 — 0.197%
• Straight: 10,200 — 0.392%
• Three of a Kind: 54,912 — 2.11%
• Two Pair: 123,552 — 4.75%
• One Pair: 1,098,240 — 42.3%
• High Card: 1,302,540 — 50.1%

These are well-documented on the Wikipedia entry for poker probability. In Texas Hold’em, the post-flop math shifts because you have hole cards and community cards interacting, but the underlying combinatorics are the same.

Pot odds

Pot odds is the ratio of the price you’re being asked to pay versus the size of the pot. If the pot is $100 and the bet to call is $20, the pot is offering you 5-to-1 odds: you risk 20 to win 120 (the pot plus the bet).

The pot-odds calculation: divide your call cost by the total pot after your call. $20 ÷ $120 = 16.7%. That’s the breakeven percentage — if your hand wins more than 16.7% of the time, the call is profitable.

The skill is knowing your hand’s equity (the percentage of the time you win) against the opponent’s range. If you have a flush draw with one card to come (roughly 19.6% to hit), calling 16.7% pot odds is profitable. If you have a gutshot straight draw (roughly 8.5% to hit), calling at 16.7% pot odds is losing money.

Expected value

Expected value (EV) is the average outcome of a decision if you made it many times under identical conditions. EV = (probability of winning × amount won) − (probability of losing × amount lost).

Example: you have a 30% chance to win a $100 pot for a $20 call. EV = (0.30 × $100) − (0.70 × $20) = $30 − $14 = +$16. The decision is positive-EV by $16 — over many identical situations, you’d average $16 profit per call.

Positive-EV plays are correct decisions even when they lose. A 99% favorite who loses once didn’t make a bad decision; they hit the 1% variance side of a correct play.

The Fundamental Theorem of Poker

David Sklansky’s Fundamental Theorem of Poker says: “Every time you play a hand differently from the way you would have played it if you could see all your opponents’ cards, they gain; and every time you play your hand the same way you would have played it if you could see all their cards, they lose.”

The theorem reframes poker as a decision-quality problem. You can’t see your opponent’s cards, but you can estimate the probability distribution of their possible holdings (their “range”) and play the decision that would be correct against the average of that range. When you’re right about the range, your decision is the same as a full-information decision, and the opponent loses long-term even on hands where you happen to lose the pot.

Implied odds

Pot odds account only for the current pot. Implied odds account for the money you can win on later streets if you hit your hand. A flush draw on the turn might offer poor immediate pot odds but excellent implied odds if your opponent is likely to pay a big bet on the river when you complete.

The implied-odds adjustment: estimate the additional bet size you’d extract on later streets if you hit, multiply by your hit probability, and add it to the win column of your EV calculation. The result is your “true” expected value with future betting included.

The Independent Chip Model (ICM) in tournaments

Tournament poker introduces a layer cash poker doesn’t have: ICM. The Independent Chip Model converts chip counts into real-money expectations based on remaining payouts. A chip late in a tournament can be worth less than a chip mid-tournament because the payout jumps from one finishing position to the next are non-linear.

ICM affects all-in decisions on the bubble (the point just before payouts start). A play that would be positive chip-EV can be negative ICM-EV because losing the tournament costs you more real money than winning it gains. Tournament regulars memorize ICM tables for common bubble situations.

Variance vs. skill

Variance is the statistical spread of results around your expected value. Poker has high variance, which means even a winning player can lose for months and a losing player can win for months. The math is unavoidable.

The way to think about it: skill is the size of your edge in EV per hand. Variance is the noise around that edge. Over thousands of hands, the edge accumulates faster than the noise; over a few hundred hands, the noise can completely mask the edge. Players who quit after a bad month are quitting on variance, not on skill.

Professional players track their win rate in “big blinds per 100 hands” because that smooths variance over a large enough sample. A 5 bb/100 win rate is solid; a 10 bb/100 win rate is elite at most stakes.

Range vs. hand thinking

The shift from amateur to intermediate poker is the shift from thinking about “the hand my opponent has” to “the range of hands my opponent could have.” A pre-flop raiser opens a defined range; a continuation bettor on the flop continues with a sub-range of that opening range.

The math: your hand has a fixed equity against each specific opponent hand. Against a range of opponent hands, your equity is the weighted average of each matchup. Solvers (programs like PioSolver and GTO+) calculate exact game-theory optimal frequencies for raising, betting, and calling at every node of the decision tree.

Game-theory optimal play

GTO (Game-Theory Optimal) poker is the unexploitable strategy — the equilibrium where neither player can improve their result by changing their strategy unilaterally. It’s the Nash equilibrium of poker.

Pure GTO is unbeatable on average but doesn’t maximize against a specific exploitable opponent. The contrast is GTO vs. exploitative play: GTO is the floor; exploitative is the ceiling against weaker opponents. Professional players use GTO as the baseline and deviate to exploit reads.

Why the math wins long-term

Over millions of hands, the player who makes positive-EV decisions wins. The math takes a long sample to show through variance. Every other element (psychology, tells, table image) is a layer on top of EV.

Frequently asked questions

What’s the probability of getting a Royal Flush in poker?

0.000154% — or 1 in 649,740 hands. There are exactly 4 Royal Flushes possible (one per suit) out of 2,598,960 possible 5-card hands. In Texas Hold’em with 7 cards (2 hole + 5 community), the probability rises to roughly 1 in 30,940.

What are pot odds in poker?

Pot odds is the ratio of your call cost versus the total pot after your call. $20 to call into a $100 pot gives you pot odds of $20:$120, or 16.7%. If your hand wins more often than 16.7%, the call is profitable long-term.

What does positive EV mean in poker?

Positive expected value (EV) means a decision averages a profit if repeated many times under identical conditions. A play that’s +$16 EV averages $16 profit per occurrence, even when individual hands lose. Winning poker is built from positive-EV decisions, not from winning individual pots.

What is the Fundamental Theorem of Poker?

David Sklansky’s theorem says you win every time you play your hand the same way you would have if you could see all opponents’ cards, and lose every time you play it differently. The math of poker reduces to: estimate the opponent’s range correctly and play as if you knew their hand.

How much does luck matter in poker?

Short-term, almost everything. Long-term, almost nothing. Variance dominates samples under a few thousand hands and approaches zero over hundreds of thousands. Professional players track win rates in big blinds per 100 hands precisely because that’s the only sample size where skill consistently shows through.

The bottom line

The mathematics of poker is hand frequencies, pot odds, expected value, and the Fundamental Theorem. Variance is real and short runs lie. Long-term winners aren’t lucky — they’re correct. For a different probability-pure mental break, the Chrome Dino game at the top of this page has exactly one outcome per jump: success or game-over, no implied odds to calculate.