RTP is a long-run average — here's what that actually means
No — 96% return to player does not mean you get $96 back from a $100 deposit in a session. RTP is the share of all money staked that a game returns as winnings on average, measured over an enormous number of rounds. Your next hour is not that average. This guide is the companion to the RTP and house-edge calculator and to turnover vs. bankroll: why total wagered can exceed the cash you hold.
The one-sentence answer
Return to player is a long-run statistical average, not a refund schedule, not a prediction, and not something you are entitled to collect in one sitting. A 96% RTP game still has a 4% house edge: on average, four cents of every dollar staked eventually leave the player side over enough repeats. In a single session you can finish well ahead, lose the whole stake quickly, or land anywhere between — and none of that contradicts the published percentage.
How RTP is defined and tested
RTP is usually expressed as a percentage of total stakes returned as winnings. If the published figure is 96%, the model says that for every 100 units staked across a very large number of rounds, 96 units are returned as wins on average and 4 units are retained on average. That “on average” is doing all the work. It is not a promise about your balance after fifty spins.
Independent test labs evaluate many real-money-style games by simulating or auditing outcomes over millions of rounds and comparing the measured return to the figure the game is configured to deliver. The number you see in a help screen is that certified or configured long-run target — not a forecast for the next time you press spin. This site does not reproduce lab certificates or name testing houses; it uses the percentage you type as an input to plain arithmetic. The sourcing contract is on the methodology page.
Why the same title can show different RTP figures
Game suppliers sometimes ship one title with more than one return setting. An operator chooses which setting runs in its lobby. That is why a generic “96% slot” mention elsewhere can disagree with the percentage inside the game you actually opened. Before you rely on any calculator output, open the information panel for the exact version in front of you and copy the RTP printed there.
Mixing versions within one session breaks the single-RTP assumption every calculator here uses. So does switching between a high-RTP base game and a bonus feature with a different weighting unless you know both figures. When in doubt, use the conservative figure — the lower RTP — for a cost estimate, not the highest number you saw in marketing copy.
Why your session is not the average: sample size
The law of large numbers says averages converge only after enormous repetition. Your session is a small sample drawn from a wide distribution. The table below is illustrative: it describes direction, not measured variance for any named game. No row is a prediction.
| Rough volume of stakes | How tightly results cluster around RTP |
|---|---|
| About 100 rounds | Very wide. Finishing far ahead or busting out quickly are both ordinary. |
| About 1,000 rounds | Still wide for volatile games. The average cost may not feel “fair” yet. |
| Millions of rounds | The measured return converges toward the published RTP. |
That gap is why volatility matters as a separate idea: two games can share the same RTP and still feel nothing alike in a short session. Read volatility: why two 96% games feel completely different. For price-per-unit math at a fixed edge, read house edge in plain numbers: the price of play. Recompute expected loss for your own stake and round count in the RTP and house-edge calculator.
What RTP is still useful for
RTP is a comparison tool for average cost, not a winning strategy. If one game is configured at 94% and another at 97% with the same stake pattern, the 97% line has a lower average cost per unit staked — three cents less retained per dollar, in that simplified comparison — but both still carry a house edge. RTP does not rank casinos, predict bonuses, or tell you which title to play for profit.
House edge
100% − RTP
Turnover
stake × number of rounds
Every stake counts, including re-staked wins.
Theoretical expected loss
turnover × house edge
Long-run average only.
Worked example (hypothetical)
Different numbers from the calculator page: a 97% RTP game, $2 per round, 250 rounds.
- House edge
- 100% − 97% = 3%
- Turnover
- $2 × 250 = $500
- Theoretical expected loss
- $500 × 3% = $15
Fifteen dollars is the average cost of placing $500 of stakes at a 3% edge. You might lose $80 in those 250 rounds, break even, or finish ahead. The $15 figure does not predict which. If $500 of turnover sounds large relative to your cash, read turnover vs. bankroll: why you can wager 3,000 from a 100 budget for the bookkeeping split, then recompute with your own numbers.
Check the figure before you calculate
- Open the game information or help screen for the version you would actually play.
- Copy the RTP percentage shown there — not a lobby banner from memory.
- Enter that percentage, your stake, and a realistic round count in the RTP and house-edge calculator.
- Compare theoretical expected loss to the cash you can afford to lose entirely — not to turnover itself.
Expected loss is a statistical model, not a session forecast. It is not gambling, legal, tax, or financial advice.
A safer-play note
A favourable RTP does not make a session “safe” or justify chasing losses. Set money and time limits before any stake, treat any budget as spending you can lose in full, and stop when a limit is hit. Independent help resources are on the responsible gambling page. 18+ only.
Frequently asked questions
If RTP is 96%, will I lose exactly 4% of my deposit?
No. RTP describes return on turnover — the sum of stakes — not on a single deposit. Re-staked wins add to turnover, and any one session can finish far above or below the long-run average.
Is a higher-RTP game a way to win?
No. A higher RTP only lowers the average cost per unit staked over enormous volume. It does not remove the house edge, predict a session, or guarantee you keep more cash.
Why does my friend’s session look nothing like the RTP?
Because RTP is measured over millions of rounds. A few hundred or thousand stakes is a tiny sample. Large wins and fast losses are both normal short-run outcomes around the same average.
Can two copies of the same game have different RTP?
Yes. The same title can ship with more than one configured return setting. The figure on a review site or lobby card may not match the version in your session — check the in-game information screen.
Where should I type the RTP for your calculator?
Use the percentage printed for the exact game and version you would play — usually in the help screen, paytable, or game information panel — not a generic marketing figure.
Sources
Every figure on this page is a definition or a worked arithmetic example with its formula shown — no external factual claims are made, so no external sources are cited.
No affiliate links: this page has no paid operator placements or commission links. Read the full affiliate disclosure.