Casixno
← Glossary

Glossary

What is RTP (return to player) in iGaming?

RTP is the share of wagers a game is designed to pay back to players over its full cycle. How it is set, certified and read, and what it means for a site owner.

Updated

What is RTP? The long-run return to players, not a promise for one session.

Definition

RTP, return to player, is the percentage of all money wagered on a game that the game is designed to return to players as winnings over a very large number of rounds. It is a property of the game's mathematical model, set by the studio that builds the game and certified by an independent test laboratory before the game is released. (source: UK Gambling Commission remote technical standards, iTech Labs)

RTP is theoretical. It describes the long run across millions of rounds; any single session, day or even month can sit far above or below it. The part the game keeps, 100% minus RTP, is the house edge. (source: UK Gambling Commission remote technical standards)

Example

A slot is certified at 96.0% RTP. If players wager 1,000,000 USDT on it over many millions of spins, the model returns about 960,000 USDT in winnings and about 40,000 USDT stays with the site as gross gaming revenue, before any costs. In one evening the same game can pay a 50,000 USDT jackpot and show a negative result; the 96% only appears when the sample is large enough. (source: UK Gambling Commission remote technical standards)

Formula

  • RTP (%) = total returned to players ÷ total wagered × 100, measured over the game's full cycle (theoretical RTP). (source: UK Gambling Commission remote technical standards)

  • House edge (%) = 100 − RTP. A 96% RTP game has a 4% house edge. (source: UK Gambling Commission remote technical standards)

  • Expected GGR = wagers × (1 − RTP). Worked example (illustrative): 250,000 USDT wagered × (1 − 0.96) = 10,000 USDT expected GGR.

  • Actual RTP (a period) = winnings paid ÷ wagers in that period; it converges toward theoretical RTP as volume grows.

Comparison

  • Typical theoretical RTP by game type (illustrative ranges, studio-specific): online slots 94–97% · European roulette 97.3% · American roulette 94.7% · blackjack with basic strategy about 99% · baccarat banker bet about 98.9% · crash and instant games 96–99%, set by the provider. (source: iGaming Times glossary, GLI testing and certification)

  • RTP vs volatility: two games can share 96% RTP while one pays small wins often and the other pays rarely but big. Volatility is a separate property and is not in the RTP figure. (source: iGaming Times glossary)

  • RTP vs hold: hold is what the site keeps in a period (GGR ÷ wagers); over time hold approaches 1 − RTP, blended across the games played.

Impact on the owner

  • You do not set RTP. The studio sets it and the lab certifies it; where a studio offers several certified versions of a game (for example 96%, 94% and 92%), the operator or its aggregator chooses which version is live in a market. (source: iTech Labs)

  • A higher RTP means less expected GGR per wager but often longer play and more wagering; a lower RTP raises hold per wager and can shorten sessions. The right mix is a lobby decision, not a per-game one.

  • RTP is public. Players compare it, and regulated markets require it to be shown on the game info page; choosing low-RTP versions is visible to the audience you want to keep. (source: iTech Labs)

  • Jackpots and high volatility create months where actual RTP is far above theory; a revenue model built on expected GGR needs a buffer for them.

How it works in practice

The studio builds the game with a random number generator and a pay table that together produce the target RTP; a test laboratory (for example GLI, iTech Labs or BMM Testlabs) verifies the RNG and the RTP by simulation and certifies the result. The certified RTP is printed in the game's information screen. In the back office an operator sees actual RTP per game, provider and period, and uses it to spot anomalies: a game running far from its certified figure over a large sample is a configuration or integration problem to raise with the aggregator or studio, not a feature. (source: iTech Labs, GLI testing and certification)

FAQ

  • That over the game's full cycle it is designed to return 96 of every 100 units wagered; it is not a promise for any session.
  • Yes on expectation; it also tends to keep players playing longer, which is why many operators prefer it for retention.
  • Not freely. It can only select among versions the studio has certified, where offered. (source: iTech Labs)
  • Variance. Short periods and jackpots make actual RTP swing; theory applies to very large samples.
  • They are the two sides of one number: house edge = 100 − RTP.