What Is RTP in Casino Games?

RTP is not measured by watching players. It is calculated from the game's own design before anyone plays it.

For a slot, the developer knows exactly which symbols sit on each reel strip, how often each combination can land, and what the paytable pays for it. Multiply the probability of every possible outcome by what that outcome returns, add them all together, divide by the stake, and you have the RTP. It is arithmetic over the complete set of possibilities, not an observation.

For table games the same principle applies to cards or a wheel. A roulette wheel's return follows directly from how many pockets it has and what the payouts are; a blackjack variant's return follows from its rule set — and, unlike a slot, from whether the player uses correct strategy, since player decisions are part of the maths.

Independent testing laboratories then verify the figure. They review the source mathematics and run simulations over sample sizes far larger than any real player could produce, confirming that observed returns converge on the stated number and that the random number generator is producing genuinely unpredictable results. That is why a provider's published RTP is the figure to trust, and why claims about RTP from anywhere else are worth very little.

House edge is simply RTP viewed from the operator's side of the table. If a game returns 96% of stakes to players over the long run, it keeps 4%, and that 4% is the house edge.

The right-hand column is the one worth internalising, because it is measured per amount wagered, not per amount deposited. A player who deposits $100 and cycles it through the game several times has wagered far more than $100, and the theoretical cost scales with the total wagered, not the deposit. Our guide to the casino house edge covers that distinction in more depth.

Take a hypothetical slot with 96% RTP played at $1 per spin. Here is what the theory says, and what it does not say.

Now the important caveat. In the first row, almost nobody finishes at exactly −$4. Over 100 spins on a typical slot, real results are spread across a wide range: many players are down most of their stake, some are roughly even, and a few are up substantially because a bonus round landed. The −$4 is the centre of a very wide distribution, not a typical outcome.

As the sample grows, that spread narrows relative to the amount wagered. By a million spins the observed return is close to 96% with high confidence. This is the whole idea of a long-run average, and it is why the figure is genuinely useful for comparing games and almost useless for predicting an evening.

Two forces are at work in any session: the expected return, which pulls slowly towards the RTP, and randomness, which dominates everything short.

The width of that randomness is called volatility, and it is a completely separate property from RTP. Two games can both return 96% and feel nothing alike — one paying small amounts constantly, the other paying almost nothing for hundreds of spins and then a large multiple of your stake. Same average, entirely different experience and entirely different bankroll requirements. Slot volatility explains how to read that dimension.

This is also why "a game is due" is never true. Each round is generated independently, and the game holds no memory of what came before. A slot that has paid nothing for 300 spins is not owed anything, and one that just paid a large win is not depleted. The RTP is not a quota being tracked and settled; it emerges from the odds of each independent round.

RTP is a comparison tool. Used as one, it is genuinely valuable.

• Compare within a category. A percentage point between two similar slots is a real difference in theoretical cost per amount wagered, and it compounds over many rounds.
• Read it alongside volatility. A high-RTP, high-volatility game can still empty a small bankroll quickly. The two numbers answer different questions.
• Check the paytable, not a listicle. Some games ship with more than one RTP configuration, and optional side bets or bonus-buy features often carry a different return from the base game.
• Remember what it cannot do. No RTP figure, however high, makes a game a positive expectation. Every number below 100% is a cost over time.
• Ignore it entirely for one session. Over an hour, volatility swamps a two-point RTP difference completely.

• Myth: a 96% RTP means I get $96 back from every $100 I deposit. Reality: it applies to the total amount wagered, which is usually far more than the deposit because winnings get re-staked.
• Myth: a game with a high RTP pays out more often. Reality: that is hit frequency, a volatility characteristic. A high-RTP game can pay rarely and largely.
• Myth: RTP changes based on your stake or the time of day. Reality: it is fixed in the game's mathematics and does not respond to how or when you play.
• Myth: a game that has not paid out is "due" and its RTP will force a win. Reality: rounds are independent. There is no balancing mechanism.
• Myth: RTP tells you whether a casino is fair. Reality: fairness comes from certified game mathematics and a verified RNG. RTP describes the game's design, not the operator's conduct.

• What is casino house edge — the same number from the operator's side
• What is slot volatility — how far results swing around the average
• What is an RNG — the mechanism that makes each round independent
• Casino glossary — RTP, hit frequency, variance and other terms defined

Lunaris describes RTP the same way this guide does: the theoretical percentage of all wagers a game is expected to return over a very large number of rounds, with no guarantee about any individual session, because every round is random. Each game's RTP is set by the studio that built it, and you will find it in the game's own information panel — the paytable or rules screen inside the game — rather than on the lobby tile.

RTP is also worth understanding before you claim a bonus, because the two interact. Lunaris bonuses carry a 20x wagering requirement calculated on your deposit plus your bonus amount. On a $100 deposit with a $100 bonus, that is $200 × 20 = $4,000 in wagers. Run that $4,000 through games returning 96% on average and the theoretical cost of completing it is around 4% of $4,000 — about $160 in expectation, spread across thousands of rounds and with wide variation in either direction around that figure. That is not a prediction of your result; it is what the mathematics says on average, and knowing it makes the offer easier to evaluate honestly.

Game type matters here too, because Lunaris uses a contribution table: non-progressive slots contribute 100% towards wagering, video poker 20%, blackjack RNG variants and roulette 10% each, and live games 1%. A game with a higher RTP but a low contribution percentage takes far longer to clear a requirement. You can browse the full game library or go straight to the slots collection, where each title's information panel carries its own figures.