Why a single hand tells you almost nothing
A blackjack hand ends in a win, a loss, or a push, and none of those outcomes tells you whether the decision that produced it was sound. A player who hits on 16 against a dealer's 10 and draws a 5 has made a losing decision that happened to work out this time. Expected value, usually shortened to EV, is the tool that separates the quality of a decision from the luck of one result. It answers a different question than "what happened": it asks what would happen on average if the same situation were played out thousands of times under the same rules.
The arithmetic is not complicated. EV is the sum of each possible outcome multiplied by the probability of that outcome. In blackjack the possible outcomes of a single decision are usually win, lose, and push, and the probabilities come from the specific rule set in force: how many decks are in the shoe, whether the dealer stands or hits on soft 17, whether doubling after splitting is allowed, and whether surrender is offered. Change any one of those rules and the probabilities shift, which is why a published basic-strategy chart always states the rule set it was derived from.
The pieces you need before any calculation
Three inputs are required to calculate the EV of a blackjack decision. The first is the probability of each outcome. The second is the amount you stand to win or lose on each outcome. The third is the amount returned on a push, which is the original stake with no gain and no loss. With those three in hand, the calculation is mechanical.
For a standard win you receive your stake back plus an equal amount in profit. For a loss you forfeit the stake. A push returns the stake. A blackjack, meaning an ace and a ten-value card as the first two cards, pays at a different rate than an ordinary win, and that rate is printed on the table layout. The payout for a blackjack is one of the most consequential numbers in the game because it changes the EV of every hand you are dealt, not just the hands that produce a blackjack.
| Outcome | Amount returned on a 10-unit bet | Net gain or loss |
|---|---|---|
| Win | 20 units | +10 units |
| Loss | 0 units | -10 units |
| Push | 10 units | 0 units |
| Blackjack | 25 units at a 3-to-2 payout | +15 units |
The table uses a 10-unit bet purely as an example. The same relationships hold for any stake. The blackjack row shows a 3-to-2 payout because that is the rate most readers will encounter, but the layout at any given table states its own rate, and some tables pay a different rate. That difference is not cosmetic: a lower blackjack payout increases the house edge on every hand, because it reduces the value of the best possible starting hand.
Building an EV calculation from hypothetical numbers
Suppose a particular decision in a particular rule set has a 42% chance of winning, a 49% chance of losing, and a 9% chance of pushing, with no blackjack possible on the decision because the hand is already in progress. Suppose also a 10-unit stake. The EV is the sum of three products: the win probability times the win amount, the loss probability times the loss amount, and the push probability times the push amount. That is 0.42 multiplied by 10, plus 0.49 multiplied by negative 10, plus 0.09 multiplied by 0. The result is 4.2 minus 4.9, which is negative 0.7 units. On average, over many repetitions of this exact decision under these exact rules, the player loses 0.7 units per 10 units staked.
Those probabilities are hypothetical and labelled as such. They are not the probabilities for any specific hand, because the probability of winning, losing, or pushing depends on the player's total, the dealer's upcard, the number of decks, and the rule set. The method, however, is exactly the method used to build every basic-strategy chart in existence. The chart is a lookup table of decisions that have already had their EV calculated under a stated rule set, and the recommended play is the one with the highest EV, which in a negative-expectation game usually means the least negative EV.
Where the house edge lives
The house edge in blackjack is not a fee collected at the door. It is built into the structure of the game, and the largest structural element is the order of play: the player acts first, and if the player busts, the stake is lost even if the dealer subsequently busts as well. That single rule is the reason the dealer does not need to outplay anyone. The player and dealer bust on the same hand often enough that the dealer's position, acting last, is worth a measurable advantage over time.
The second structural element is the payout schedule. A blackjack that pays less than the full value of the hand reduces the player's return on the strongest starting combination. The third is the set of optional decisions: doubling, splitting, and surrendering. Each of those options has its own EV, and a player who uses them correctly reduces the overall expected loss per hand; a player who ignores them or uses them incorrectly increases it. None of these elements is hidden. All of them are visible in the table rules before the first card is dealt.
House edge and RTP are complements. For the same game under the same rule set, they sum to 100%. If a game returns a certain percentage to players over the very long run, the house edge is the remainder. Neither number predicts a session. A player can win for an evening while playing a game with a house edge, and can lose quickly on a game with a relatively low one, because the edge only expresses itself over a large number of hands.
Why short sessions look nothing like the average
Variance is the distance between individual results and the long-run average. Blackjack is a relatively low-variance game compared with many others, but it is nowhere near zero-variance. A session of a few hundred hands is a tiny sample, and tiny samples routinely produce results far from the expected value. A player who wins 60 units in an hour has not disproved the house edge; they have experienced a run of outcomes that sits on the favourable side of the distribution. The same player, playing the same strategy against the same rules, can lose 60 units in the next hour without changing a single decision.
What the player controls is not the variance but the rate at which it is experienced. Decisions per hour set the pace of total exposure. A full table deals fewer hands per hour than a heads-up game. A player who slows down, takes breaks, or plays at a fuller table places fewer bets in the same period, and therefore exposes less total money to the edge. The edge per bet is unchanged; the total expected loss over a session is the edge per bet multiplied by the number of bets, so fewer bets means a smaller total expected loss, all else equal.
What a progression system does and does not do
A progression system changes the size of the next bet based on the previous result. The most familiar version doubles the stake after a loss, on the theory that a win will recover all prior losses plus one unit. The arithmetic works when the win eventually arrives and the table allows the required bet size. The problem is that the win is not guaranteed to arrive before the stake reaches the table maximum or the player's bankroll runs out, and the sequence of bets required after a run of losses grows quickly.
Suppose a player starts at 1 unit and doubles after each loss. After one loss the next bet is 2 units, after two losses it is 4, after three it is 8, after four it is 16, after five it is 32. The total staked across those six bets is 63 units, all to chase a 1-unit profit. The house edge applies to every one of those 63 units. The progression has not removed the edge; it has increased the total amount wagered, which increases the total expected loss, while concentrating the outcomes into a small number of large bets. A progression redistributes results: many small losses and occasional large wins become many small wins and occasional large losses. The expected value of the underlying game is unchanged.
No betting pattern changes the built-in edge of a negative-expectation game. What changes is the shape of the risk: how much is at stake at any moment, how quickly a bankroll can be exhausted, and how much total money passes across the table. A player who understands EV can see a progression for what it is, a rearrangement of variance rather than an improvement in expectation.
Where to find the numbers that apply to you
The probabilities used in a real EV calculation are not something a player computes at the table. They are published in basic-strategy charts and in reference works that state the rule set they assume. The rule set is the starting point: the number of decks, the dealer's standing rule, the doubling and splitting permissions, the surrender option, and the blackjack payout. Every one of those rules is printed on the table layout or available from the dealer, and every one of them changes the probabilities.
The payout for a blackjack is printed on the felt. The minimum and maximum bets are printed on a sign at the table. The number of decks is visible in the shoe or stated in the game information panel for an online game. For an online blackjack game, the game information panel also states the theoretical return, usually as RTP, and that figure is the long-run average for that game under that rule set. It is not a per-session promise, and it assumes the player makes the decisions the game's strategy model assumes.
For a player who wants to calculate EV from scratch, the method is the one described above: list the outcomes, assign probabilities from a published source that matches the rule set, multiply each probability by its payoff, and sum. The result is the average result per decision. For a player who wants to know the EV of a specific hand against a specific dealer upcard, a published strategy chart has already done the arithmetic, and the chart's recommendation is the highest-EV play under the stated rules.
FAQ
Does EV tell me what will happen on my next hand?
No. EV is an average over many repetitions of the same decision under the same rules. Any single hand can end in a win, a loss, or a push, and the result of one hand carries no information about the quality of the decision that produced it.
Can I calculate EV for a whole blackjack session?
You can estimate the expected loss for a session by multiplying the house edge per bet by the total amount wagered, but the actual result will differ because of variance. The house edge is a long-run average, and a session of a few hundred hands is not the long run.
Why do different blackjack tables have different edges?
Because the rule set differs. The number of decks, the dealer's standing rule, the blackjack payout, and the availability of doubling, splitting, and surrender all change the probabilities of each outcome, and therefore change the EV of every decision and the overall house edge.
Gamble responsibly. If you or someone you know is struggling with gambling, contact professional support services such as the National Problem Gambling Helpline at the helpline printed in the footer below. Help is available 24/7, and all calls are confidential.
Play responsibly
Gambling should be treated as paid entertainment, never as a way to earn income or recover losses. 1-800-MY-RESET (1-800-697-3738)
18+ or 21+ depending on where you are; follow the minimum age that applies to you.
This article is general information about how these mechanics work. It is not legal advice and not a recommendation to gamble or to use any particular operator. Availability and legality differ by jurisdiction — check the rules that apply where you are.