The number people look for usually does not exist
If you're trying to find a single percentage that says how much casinos must pay back, the first thing to know is that payout is not one universal number. It depends on the specific game, the specific bet within that game, and the rule set attached to that bet.
That is why broad claims like "a casino pays out X%" are not very helpful on their own. A slot game has its own Return to Player, or RTP. A roulette table's expected return depends on which wheel is used. A blackjack table's expected return depends on the exact rules in force and on whether the player follows a basic strategy chart built for that rule set. On a craps table, different bet types sitting a few inches apart can carry materially different house edges.
So the practical answer is not a universal requirement. It is a method: identify the exact game and bet, find the game information panel or posted rules, and read the RTP or house-edge information for that exact rule set if it is published. Then treat that figure as a long-run average for the bet, not as a promise about your next hour of play.
RTP and house edge describe the same built-in maths
RTP is a theoretical percentage of total wagered money a game returns to players over a very large number of spins or hands. House edge is the operator's mathematical advantage on that same game or bet, expressed as a percentage of each wager. For the same bet under the same rule set, RTP and house edge sum to 100%.
That relationship matters because it tells you where the money goes in expectation. If, as an example, a game returned R% under one rule set, then its house edge on that same wager would be 100% minus R%. If a different version of the game changed the rules, the numbers could change as well.
This is also why you should be careful with casino-wide marketing language. Even if a venue publishes a broad average across many games, you do not wager against an average. You wager on one machine, one table, one hand, one spin, one bet type at a time.
| Term | What it means | What it does not mean |
|---|---|---|
| RTP | Theoretical long-run percentage returned across a very large sample of wagers | A guarantee for your session, day, or account |
| House edge | The expected percentage of each wager retained by the operator over time | A prediction of what you will lose on the next bet |
| Rule set | The exact game conditions that determine the maths | Something you can ignore when comparing games |
How the edge is built into each bet
The edge is not a fee added at checkout. It is built into the payoff structure of the bet itself. A bet pays less than true odds would require, or the game includes outcomes that work against the player often enough to create a long-run advantage for the house.
Roulette is a clean example. European wheels have 37 pockets with a single zero. American wheels have 38 pockets with both a single zero and a double zero. That extra pocket increases the house edge on the American layout because it adds an additional outcome that is not covered by the standard even-money and inside-bet payoff structure.
Baccarat shows a different kind of design. The player, banker, and tie bets are all placed on the same game, but they do not carry the same edge. The tie bet has the highest house edge of the three. So even within one dealing procedure, the bet choice matters.
Craps works the same way. The table is one game, but the bets are not interchangeable. Some are relatively less costly in expected loss terms than others, while some carry much higher edge. Saying "the craps edge" without naming the actual bet is too vague to be useful.
Blackjack is another case where the edge is conditional. The objective is to get closer to 21 than the dealer without exceeding it, but the expected return depends on the rule set in use. Published basic-strategy charts are derived for specific conditions such as deck count, dealer stand rules, and doubling and splitting rules. Use the wrong chart for the wrong rules and the expected loss rate changes.
Short sessions can look nothing like the long run
This is where many payout misunderstandings begin. RTP is a long-run statistical average. Variance, also called volatility, describes how outcomes are distributed over time. High volatility means rarer but larger payouts; low volatility means more frequent but smaller ones. Volatility does not change a game's RTP.
In a short session, variance can drown out expectation. A player can finish ahead, finish behind by much more than the long-run edge would suggest, or hover around break-even for a while. None of those outcomes proves the posted RTP wrong. It only shows that the sample is small.
Suppose, as an example, a hypothetical game returned R% over a very large sample. If a player made a small number of wagers, their actual result could land well above or below that average because outcomes arrive unevenly. Over more wagers, the observed return may move closer to the theoretical figure, but there is no point at which the short term becomes a guarantee of the long term.
Jackpots make this especially visible. A fixed jackpot pays a set amount, while a progressive jackpot grows from a share of qualifying wagers and resets after a win. Those large but infrequent outcomes can produce long stretches where ordinary sessions look far worse than the game's long-run average, followed by rare hits that move the aggregate return back toward its theoretical level. Some progressives also require a maximum-stake bet to qualify, so the posted headline prize may not apply to every stake level.
Speed of play changes your total exposure
Even when the edge per bet stays the same, the number of bets you place in an hour changes how much money you expose to that edge. More decisions per hour means more total amount wagered, which means the expected loss accumulates faster.
Use labelled example arithmetic to see it clearly. Suppose a player makes B-unit bets, and suppose they place N decisions in an hour. Total amount wagered for that hour is B × N units. If the house edge on that specific bet under that specific rule set is E%, then the expected loss for that hour is:
Expected loss = B × N × E%
As an example, suppose the bet size is 1 unit, the game pace is 100 decisions in an hour, and the house edge on that exact wager is 2%. The total staked is 100 units, and the expected loss is 2 units. If the pace doubles to 200 decisions while the stake and edge stay the same, total staked becomes 200 units and expected loss becomes 4 units.
That does not predict what will happen in that particular hour. Variance still rules the short run. But it does show why game speed matters. A slower game with pauses between decisions can reduce total exposure compared with a fast game or rapid auto-play behaviour, even if the edge per individual wager is identical.
Progression systems do not remove the edge
Many systems sound persuasive because they change the shape of wins and losses. Doubling after losses, pressing after wins, or switching stake sizes in a pattern can create different short-session experiences. What they do not do is alter the underlying expected value of a negative-expectation game.
The reason is simple. The house edge is attached to each qualifying wager. If each bet carries the same negative expectation in percentage terms, changing the order or size pattern of those bets redistributes variance but does not remove the built-in disadvantage.
Suppose, as an example, a player uses a progression with bets of 1 unit, then 2 units, then 4 units after losses. If the edge on each of those wagers is E%, the expected loss across the sequence is still E% of the total amount staked in the sequence. The system may produce many small winning sessions and occasional larger losing sessions, or the reverse, depending on the pattern and the table limits. But the percentage edge remains attached to the money put at risk.
This is also where table limits and bankroll size matter structurally. A progression can require larger and larger stakes after a bad run, and every real table or account has a ceiling somewhere. Once the next required stake cannot be placed, the claimed recovery logic breaks. No betting pattern changes the built-in edge of a negative-expectation game, and progression systems redistribute outcomes rather than remove the edge.
Where to find the number that applies to your game
If you want a payout figure that means something, look at the level where the maths is defined. For slots, check the game's information panel for RTP and payline or ways-to-win details. For table games, check the posted rules for the exact variant and bet options. For blackjack, compare the house rules with the basic-strategy chart intended for that rule set. For roulette, note whether the wheel is European or American. For baccarat and craps, identify the precise bet, not just the table name.
If promotions are involved, separate game return from bonus conversion. A no-deposit bonus or free spins offer is commonly subject to wagering requirements, game restrictions, expiry windows, and maximum-cashout caps. Example arithmetic only: a 30x requirement on a 50-unit bonus means 1,500 units must be wagered, and game weighting can change how much each wager counts. That is a bonus-terms question, not the same thing as the underlying RTP of the game you are playing.
And if a game uses an RNG, understand what testing means. Independent laboratories are commonly used to certify RNG behaviour, which speaks to game mechanics. It does not certify an operator's financial conduct. In some crypto-based games, a provably fair scheme can let a player verify that an individual round was not altered after the bet, but that verifies round integrity only; it is not an audit of everything else.
FAQ
Do casinos have to pay out a fixed percentage by law?
There is no single fixed percentage you can apply across all casinos and all games. The meaningful figure, where one is published, is usually attached to a specific game or bet and the rule set that governs it.
If you want the relevant number, check the game's own help or information screen, the table placard, or the posted rules for that exact variant.
Is a high RTP game guaranteed to pay me more tonight?
No. RTP is a long-run theoretical average over a very large number of wagers, not a promise about one session. A high-volatility game can go a long time without producing the outcomes that make up that average.
For a short session, variance usually matters more than the headline RTP figure.
Can I beat the payout percentage by changing my bet pattern?
Changing your bet pattern changes variance and the path your bankroll takes, but it does not change the built-in edge on a negative-expectation wager. If each bet still carries the same house edge, staking more after losses or wins does not remove that edge.
What you can control is which bets you choose, the rules you accept, how fast you play, and how much total money you expose to the game over time.
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.