What the house edge measures
The house edge is the average amount a game is expected to take from each bet, written as a share of the amount wagered. A 1 percent edge means that, across a very large number of bets, the casino expects to keep about 1 unit for every 100 units wagered. It describes a long-run average. It does not predict what will happen in one hand, one evening or one trip.
Seen from the player’s side, the same number is a negative expectation: the expected value of an average bet is below zero. Games with a house edge lose money for players over time, and nothing about betting size, timing or table choice turns that around.
A coin game as an illustration
Suppose a game pays 0.90 for every 1 you stake when a fair coin lands heads, and takes your stake when it lands tails. The expected result of one 1-unit bet is 0.5 x 0.90 minus 0.5 x 1, which is 0.45 minus 0.50, or minus 0.05. The edge is 5 percent.
Over 1,000 such bets, 1,000 units are wagered and the expected loss is about 50 units. Any actual session can finish above or below that. Short runs are noisy, and plenty of players finish ahead on a given night. What the law of large numbers says is that the average result per bet tends to settle near the expected figure as the number of bets grows. It does not say that a loss is “due” to be followed by a win.
Where blackjack’s edge comes from
In blackjack the player tries to finish closer to 21 than the dealer without going over. The main source of the edge is the order of play. The player acts first, so a player who goes over 21 loses immediately, even in the cases where the dealer would also have gone over afterwards. The dealer, by contrast, follows fixed rules, typically drawing on low totals and standing on higher ones, with no decisions to make.
Against that, the player has real advantages. A two-card 21 (a natural) usually pays 3 to 2, and the player can choose to double, split pairs or, at some tables, surrender. These features pull the edge down to a level that is low compared with many casino games. They do not remove it.
Blackjack also differs from a coin flip because cards are dealt without replacement. A standard deck has 52 cards, 16 of which count as ten. As cards leave the deck, the odds for the next card shift. That is the mathematical reason card counting is possible in principle, but it is hard to do, casinos are entitled to ask players to stop, and it carries no promise of profit.
How table rules change the edge
The same game can carry quite different edges depending on the rules posted at the table. The usual differences are:
- Payout for a natural. A 3 to 2 payout returns 15 on a 10 bet; a 6 to 5 payout returns 12. A natural arrives in roughly 4.7 to 4.8 percent of hands, and the gap of 0.3 per unit staked on those hands adds up to roughly 1.4 percentage points of edge. That is a large change for what looks like a minor wording difference.
- Number of decks. More decks generally favour the casino slightly, because some useful combinations become a little less likely to appear early.
- Dealer on soft 17. A rule requiring the dealer to draw on a soft 17 usually adds a fraction of a percentage point to the edge.
- Doubling, splitting and surrender rules. Limits on doubling after a split, on re-splitting, or the absence of surrender each add a little.
Under favourable rules and correct play, the edge is often described as being below 1 percent. Under poor rules it can be well above that. Always read the posted rules rather than assuming they match another table.
How player decisions change the edge
Basic strategy is a table of choices, one for each combination of the player’s hand and the dealer’s visible card, worked out by calculating the expected result of each option. It is not a secret technique. A typical entry is to stand on a hard 12 when the dealer shows a 6, because a 6 is a weak card and the dealer will often go over.
Decisions matter because mistakes raise the edge. Playing on hunches, avoiding hits “to be safe”, or following a table’s folklore can add noticeably to what the casino already holds. Even flawless play, though, leaves the edge above zero under ordinary rules. Good decisions shrink the expected loss; they do not remove it.
Side bets show the opposite case. Take insurance, which is offered when the dealer shows an ace and pays 2 to 1 if the dealer has a ten underneath. Ignoring the player’s own cards, 16 of the 51 unseen cards are tens, so the chance is 16/51, about 31.4 percent. The expected value per unit is 2 x 16/51 minus 35/51, which equals minus 3/51, or about minus 5.9 percent. This is far worse than the main game, which is why basic strategy normally declines it.
What the house edge does and does not tell you
The edge is a rate, not a forecast. Two players with identical strategy can finish a night far apart, because results swing widely around the average; this gap between average cost and day-to-day swings is the difference between volatility and return. A lucky streak does not mean the edge has gone away, and a losing streak does not mean the next hand is more likely to win.
It also tells you nothing about how to recover losses. Raising bets after a loss does not change the expected value of any hand, and chasing losses is one of the quickest ways to turn a small expected cost into a large real one.
Gambling can be harmful, and outcomes are always uncertain. Treat blackjack as paid entertainment, not income, and decide how much you can afford to lose before you start, in line with the habits of responsible play. If gambling is causing you or someone close to you difficulty, a local support service can help, and speaking to one early is better than waiting.