RTP is the most quoted statistic in online casino content and the most consistently misunderstood. Plenty of pages will tell you to “always play high-RTP slots” as though it were a strategy. It isn’t, and the reason why is worth understanding properly.
Here’s what the number actually measures, what it can’t tell you, and the one situation where it genuinely matters.
18+. No RTP figure makes gambling profitable. Support: BeGambleAware.org | 0808 8020 133.
The definition
Return to Player is the percentage of all money staked on a game that the game is designed to return as winnings over its full theoretical lifetime.
A slot with 96% RTP returns £96 for every £100 staked, on average, across an enormous number of spins. The remaining 4% is the house edge — the operator’s mathematical margin.
Two words in that definition carry all the weight: average and enormous.
The figure is a design specification derived from the game’s mathematics — its symbol distribution, paytable and feature frequencies. It’s verified by independent testing labs against millions of simulated rounds. It is not a forecast of your session, your week or your year.
Why 96% doesn’t mean you keep £96
Three reasons, and each one matters.
1. It’s calculated over an unimaginable number of spins
RTP converges over hundreds of millions of rounds. That’s the scale across which a game’s actual return approaches its theoretical figure — across every player on every site running that title, over years.
Your thousand-spin session is statistically invisible at that scale. You could return 20% or 400% of your stake and neither would be remotely unusual.
2. Turnover isn’t the same as your deposit
This is the point most players miss, and it’s the important one.
RTP applies to total amount staked, not to your deposit. Every time you win, you’re usually staking that money again.
Take £100 deposited, played at £1 a spin on a 96% game:
- After 100 spins you’ve staked £100 and have roughly £96 left
- Play it again: £96 staked, roughly £92 left
- Again: roughly £88
Each cycle of turnover shaves off about 4%. Play long enough and the balance trends toward zero — not because the game “turned”, but because the edge applies to every single spin. What RTP predicts is not how much of your deposit you keep, but how slowly it erodes.
3. Variance dominates everything short-term
Two games can share the same 96% RTP and behave completely differently. One pays small amounts frequently; the other pays almost nothing for long stretches then occasionally returns a very large multiple. The long-run average is identical. The experience, and the range of realistic outcomes, is not.
Volatility matters more than RTP
Volatility (sometimes called variance) describes how a game distributes its returns.
| Volatility | Behaviour | Realistic session |
|---|---|---|
| Low | Frequent small wins | Balance drifts slowly, few big swings |
| Medium | Balanced | Mixed |
| High | Rare, large wins | Long dry spells; occasional big hits; deposits can vanish quickly |
For anyone playing with a fixed budget over a limited number of spins, volatility affects the outcome far more than a percentage point of RTP.
A 96.5% high-volatility slot will empty a £50 budget faster and more often than a 95.5% low-volatility one, despite the “better” RTP — because most sessions on a high-volatility game are the dry spells that fund the rare large hits. The 96.5% is real, but the money that makes it real is largely concentrated in outcomes you almost certainly won’t experience.
If you want longer play for your money, low volatility does more for you than high RTP.
Typical RTP by game
| Game | Typical RTP |
|---|---|
| Blackjack (basic strategy) | 99%+ |
| Video poker (optimal play) | 97–99.5% |
| Baccarat (banker bet) | ~98.9% |
| European roulette | 97.3% |
| Craps (pass line) | ~98.6% |
| Online slots | 92–97% |
| American roulette | 94.7% |
| Progressive jackpot slots | 88–94% base game |
| Keno / scratchcards | 75–92% |
Two things stand out.
Table games generally return more than slots. Blackjack played with correct basic strategy has one of the lowest edges available — but “with correct basic strategy” is doing serious work in that sentence. Played by feel, blackjack’s real return is considerably worse than the theoretical figure. Slots require no decisions, so their stated RTP is what you get.
Progressive jackpots have low base-game RTP because a portion of every stake diverts to the jackpot pool. The total RTP including the jackpot may look respectable, but that portion is only realised by whoever eventually wins it. For everyone else — which is essentially everyone — the effective return is the base figure.
American roulette’s extra zero costs about 2.6 percentage points against the European wheel. If both are available, there is no reason to play the American version. Ever.
The RTP setting most players never check
Here’s the detail that actually matters, and it’s not the one the “play high-RTP slots” advice points at.
Many slot studios ship the same game in multiple RTP configurations — a title might exist at 96.5%, 94.5% and 92% versions. Operators choose which to licence. The game looks identical: same artwork, same features, same name in the lobby.
The difference between 96.5% and 92% is substantial. It roughly doubles the house edge, and it’s invisible unless you look.
How to check. Open the game’s information or paytable panel, usually behind an “i” or menu icon, and find the stated RTP. It should be there — UK technical standards require the figure to be available to players. If you can’t find it, that’s worth noting about the site.
This is where RTP awareness has practical value: not in chasing a fractionally higher figure across games, but in noticing when a casino is stocking a materially degraded version of a game available elsewhere at full setting. We check this in our casino testing.
Worth watching over the next couple of years: Remote Gaming Duty rose sharply to 40% in April 2026, the largest single increase in the tax’s history. Operator costs feed through somewhere, and lower-RTP game configurations are one of the available levers. Checking the number in the info panel is likely to become more useful, not less.
What RTP does not do
It doesn’t predict your session. Variance dominates any realistic number of spins.
It doesn’t change during play. RTP is fixed by the game’s mathematics. A slot doesn’t tighten because you’re winning, or loosen because you’ve been losing. Certified RNGs produce independent results and operators can’t adjust a game’s return in response to individuals.
It doesn’t create “due” payouts. Each spin is independent. A game that hasn’t paid in 500 spins has exactly the same probability on spin 501 as it did on spin 1. Believing otherwise — the gambler’s fallacy — is the single most expensive misconception in gambling, because it turns a losing run into a reason to keep going.
It doesn’t make anything profitable. Even at 99%, the expected outcome of sustained play is loss. There is no RTP figure at which a casino game becomes a way to make money, and the entire business model depends on that being true.
It isn’t a strategy. Choosing a 96.5% game over a 96% one changes your expected loss by about 50p per £100 staked. That’s real, and it’s also nothing compared with how long you play and how much you stake.
What actually affects how much you lose
In descending order of impact:
- How much you stake in total — turnover is the multiplier on everything
- How long you play — more spins means the average asserts itself
- Volatility — determines the shape and range of outcomes
- RTP — a small effect, applied to the total above
- Game choice within a category — usually marginal
Stake size and session length are entirely within your control. RTP is a rounding error next to them.
This is why deposit limits and session limits do more for your bankroll than any amount of game selection. UK operators must offer both, and increases to a deposit limit are subject to a cooling-off period specifically so that a decision made mid-session can’t take effect mid-session.
Frequently asked questions
What’s a good RTP for a slot? 96% and above is standard for mainstream titles. Below 94% is poor, and below 92% is worth avoiding outright. But volatility and your own stake and session limits affect your results far more than this figure.
Can casinos change a game’s RTP? Not for individual players, and not during play. Operators can choose which configuration of a multi-RTP game to stock, and that applies to everyone on the site. The version in use should be shown in the game info panel.
Does a higher RTP mean I’ll win more often? No. RTP describes total return, not win frequency. A high-RTP, high-volatility game can pay far less often than a lower-RTP, low-volatility one — it just pays more when it does.
Where do I find a game’s RTP? In the game’s info, help or paytable panel. UK technical standards require it to be available. If a site hides it, treat that as a mark against the site.
Do jackpot slots have bad RTP? The base game usually does, because stakes divert to the jackpot pool. The headline combined figure includes money that only one eventual winner receives. Judge the base game figure, because that’s the one you’ll experience.
Is a 99% blackjack RTP achievable? Only with correct basic strategy on every hand, and only on rule sets that support it. Deviate and the real return drops well below the quoted figure. The number describes optimal play, not typical play.