House advantage is the casino’s expected average gain from a wager, expressed as a percentage of the amount wagered. In most casino discussions, it is another name for the house edge.
A 2% house advantage does not mean the casino wins 2% of every player’s buy-in, takes 2% from each individual bet, or finishes every session ahead. It means that, if the same wager is repeated under the same rules often enough, the casino’s average result is expected to approach 2 units for every 100 units put into action.
That distinction—average expectation versus actual session result—is the whole definition.
The percentage comes from probabilities and payouts
Every casino wager has possible outcomes. Each outcome has:
- a probability of occurring; and
- a net financial result for the player.
The player’s expected value per wager is:
Player EV = Σ (Probability of outcome × Player’s net result for that outcome)
House advantage = −Player EV ÷ Initial wager × 100%
Here:
- Σ means add the contribution from every possible outcome;
- player EV is the player’s average mathematical result per wager;
- net result excludes the returned stake and counts only profit or loss; and
- initial wager is the amount used as the denominator.
The minus sign converts a negative player expectation into a positive casino advantage. If a $10 wager has a player EV of −$0.30, the house advantage is 3%.
This calculation assumes the stated rules, payout table, and player decisions. Change any of those and the percentage may change.
A complete American roulette example
Consider a $1 straight-up bet on one number on an American roulette wheel. The wheel has 38 pockets: 1 through 36, 0, and 00. A winning straight-up bet normally pays 35 to 1.
The player has two mutually exclusive results:
| Result | Probability | Net player result | Contribution to EV |
|---|---|---|---|
| Chosen number wins | 1/38 | +$35 | +$35/38 |
| Any other number wins | 37/38 | −$1 | −$37/38 |
So:
Player EV = (1/38 × $35) + (37/38 × −$1)
Player EV = −$2/38 = −$0.052631… per $1 wagered
House advantage = 5.2631…%
The two extra green pockets create the gap. True odds for choosing one of 38 pockets would require a 37-to-1 profit, but the casino pays 35 to 1. The missing two units become the mathematical advantage.
A player can still win $35 on the next spin. The 5.26% figure describes the average value of the wager, not what the wheel must do tonight.
House advantage, RTP, and expected loss
These terms are connected, but they answer different questions.
| Measure | What it expresses | Typical question |
|---|---|---|
| House advantage | Casino’s expected gain as a percentage of action | “How expensive is this wager mathematically?” |
| RTP | Player’s expected return as a percentage of action | “How much is returned on average?” |
| Expected loss | House advantage converted into money | “What is the average cost of this amount of play?” |
| Hold percentage | Actual or theoretical win divided by a specified base | “What did the casino retain relative to drop, handle, or another denominator?” |
For a complete fixed-odds game using the same wagering denominator:
House advantage = 100% − RTP
A theoretical RTP of 97% therefore corresponds to a 3% house advantage. That relationship should not be applied carelessly when the figures use different denominators, when pushes or partial stakes are handled differently, or when a strategy-dependent game assumes different player decisions.
The UK Gambling Commission describes house edge as the percentage a casino expects to keep on average from each hand or spin under normal patterns of play. Its house-edge and return-to-player guide also stresses that these are average measures rather than promises about an individual session.
Turning the percentage into money
The standard expected-loss calculation is:
Expected loss = Total action × House advantage
Suppose a player makes 200 wagers of $10 on a game with a 2% house advantage.
- Total action = 200 × $10 = $2,000
- Expected loss = $2,000 × 0.02 = $40
The $40 is not a bill that arrives at the end of the session. It is the average result across a very large number of comparable sessions. One player may finish $400 ahead while another loses $500. The expectation becomes useful for comparing games and estimating long-run cost, not predicting a single outcome.
The denominator also matters. House advantage normally applies to money wagered, not the cash a player brought to the casino. A person can buy in for $200 and recycle the same chips through $2,000 of action. Applying a 2% edge to the $200 buy-in would understate the expected cost by a factor of ten.
What the percentage does not tell you
House advantage alone does not describe the experience of playing the game.
It does not measure volatility
Two wagers can have the same 3% house advantage and behave very differently. One may produce frequent small wins and losses. Another may lose almost every time but occasionally pay a large prize. Their average cost can match while their short-term swings do not.
It does not predict time to loss
Game speed changes how quickly action accumulates. A 1% wager made 30 times an hour can cost less in expectation than a 0.5% wager made 200 times an hour at the same stake.
Expected hourly loss = Average wager × Decisions per hour × House advantage
This is an estimate. Pauses, pushes, changing bets, imperfect strategy, side bets, and different game speeds can move the real figure.
It does not equal casino hold
A table can have a positive house advantage yet lose money on a shift. Actual hold reflects the particular results that occurred and usually uses a different denominator, such as drop. The mathematical edge is stable under fixed rules; actual hold is noisy.
It does not prove unfair dealing
A disclosed game can operate randomly and follow its approved rules while still favoring the casino. The advantage comes from the relationship between probability and payout. Cheating would be a separate issue involving manipulation or failure to follow the rules.
Strategy-dependent games need an assumption
Roulette’s edge is fixed for a given wheel and rule set because the player’s number choice does not change the probabilities. Blackjack and video poker are different. Their expected return depends partly on decisions.
A published blackjack house advantage might assume correct basic strategy. A player who repeatedly makes weaker decisions creates a larger effective disadvantage. Likewise, a video poker RTP normally assumes a specified paytable and accurate play.
Whenever a percentage is quoted, ask:
- Which rules and paytable were used?
- What player strategy was assumed?
- Is the percentage based on the initial wager or another measure?
- Does it include optional side bets?
Without those details, a precise-looking number may not describe the game in front of you.
Why casinos care about a small advantage
A small edge becomes commercially meaningful when multiplied by large action. Casinos combine the game’s mathematical advantage with betting volume, game speed, staffing, limits, credit exposure, and operating cost.
A low-edge baccarat table can still generate substantial theoretical win if it handles large wagers. A high-edge novelty bet may produce little revenue if few players take it. House advantage is the price built into the wager; volume determines how much of that price is put to work.
This is also why a low edge should not be confused with low total cost. Repeated play can make a mathematically inexpensive wager costly in dollars. The practical comparison is not only “Which percentage is lower?” but also “How much will I wager, and how quickly?”
The useful way to read the number
Treat house advantage as a cost rate for wagering, not a forecast of your final balance.
It is useful for:
- comparing two versions of the same game;
- identifying expensive side bets;
- estimating expected loss from planned action;
- separating mathematical value from a large advertised payout; and
- understanding why short-term wins do not overturn long-term game design.
For the player, lower is generally better when the comparison uses the same assumptions. For the casino, the advantage is the foundation of expected revenue, but it never removes short-term risk.
Continue with expected value for the probability calculation, expected loss for the money calculation, and hold percentage for the operational measure that is most often confused with house advantage.