In casino terms, RTP (return to player) is the percentage of all money wagered on a game that it pays back to players over the long run. Whatever it does not pay back is the house edge, so a 96% RTP slot keeps an average of 4% of everything staked on it.
What you need: a calculator, access to the game’s info or help screen, and the operator’s license jurisdiction.
Step 1: Learn the RTP Meaning Casino Games Use, Return to Player as a Long-Run Percentage
Return to player (RTP) is the percentage of all money wagered on a game that it pays back to players over time. A 96% RTP returns an average of $96 per $100 wagered. The remaining 4% is the house edge.
The key word is “average.” The figure on a game’s help screen is theoretical RTP. It describes the expected return across millions of rounds, the full cycle of outcomes the game’s math model can produce. It says nothing reliable about a single session of a few hundred spins.
That gap explains why one player’s results can sit far above or below the published number. A player who hits a large bonus feature early can finish a session at 300% of what they staked. Another can lose their whole budget without a single meaningful win. Both outcomes fit a 96% game, because the RTP only settles toward its stated value as the number of rounds grows. Short samples are dominated by variance, not by the long-run percentage.
For a player, RTP tells you the price of play over time. For an operator, it is the margin the game is built to deliver across the whole player base. Neither side should read it as a prediction for the next spin. If you want to see how players apply it in practice, our guide on how RTP and volatility shape slot bet sizing covers that side. This article stays with the platform view.
One rule runs through every step that follows:
House edge = 100% − RTP.
If you meet RTP alongside related terms such as hold, GGR or hit frequency, the gambling glossary of casino and regulatory terms defines each one.
Step 2: Calculate RTP and House Edge From Total Wagers and Total Payouts
RTP equals total payouts divided by total wagers, multiplied by 100.
RTP = (total payouts ÷ total wagers) × 100
A worked example: a game takes $1,000,000 in wagers and pays out $962,000.
- RTP = (962,000 ÷ 1,000,000) × 100 = 96.2%
- House edge = 100% − 96.2% = 3.8%
- Gross win for the operator = $38,000
That calculation works on your own transaction data. It tells you what a game actually returned over a period. It is not how a certified figure is set. Test labs such as GLI, and regulators such as the UK Gambling Commission, work from the game’s math model instead. The lab takes the paytable, the reel strips or symbol weights, and the probability of every winning combination and feature trigger. It multiplies each payout by its probability and sums the results. The output is the theoretical RTP, which is then confirmed by large-scale simulation. A short play sample would be far too noisy to certify anything.
The table below converts common RTP levels into what they cost per $1,000 wagered:
| RTP | House edge | Expected loss per $1,000 wagered |
|---|---|---|
| 99% | 1% | $10.00 |
| 96.7% | 3.3% | $33.00 |
| 96% | 4% | $40.00 |
| 95% | 5% | $50.00 |
| 92.85% | 7.15% | $71.50 |
Note that $1,000 wagered is total turnover, not a deposit. A player who deposits $100 and recycles wins can easily turn over $1,000 in one session.
The math also has a limit. Expected loss is an average across all players and all rounds. Volatility decides how far any one session moves away from it. Two games at 96% can feel completely different. One pays small amounts often. The other pays rarely but large, and swings much further from the $40-per-$1,000 line.
Step 3: Find a Game’s Real RTP, Not Just the Advertised Default
The RTP a player gets is the version the operator switched on, which may sit below the studio’s advertised default.
Many providers ship one title in four or more certified RTP versions. Each version is a separate math configuration, tested and approved in its own right. The operator picks which build goes live, usually through the remote game server (RGS) integration settings. The art, features and name stay the same. The paytable or weighting underneath changes, and so does the return.
To check the figure that applies to you:
- Open the game’s info or help screen. The stated RTP there should match the build in play, not the studio’s marketing default.
- Compare it with the provider’s published default. If the help screen shows a lower number than the provider’s product sheet, the operator has selected a lower certified version.
- Confirm the license. The jurisdiction decides whether the operator must display the figure at all and how accurate it must be.
Display rules differ by market. In 2026, UKGC-licensed casinos must display accurate RTP figures for every game, so a UK player should always find the live figure in the help screen. In the US, the situation is different. A claim circulated that every slot must show RTP from January 1, 2026. No state rule backs it, so do not assume an American game will disclose its version.
For operators, version selection is a configuration decision with commercial and compliance weight. It should be recorded per market, matched to the certificate for that jurisdiction, and reflected on the help screen the moment it changes.
Step 4: Compare RTP Across Online Slots, Table Games and Land-Based Machines
Online slots usually return about 95% to 96.7%, blackjack played with basic strategy returns 99% or more, and land-based slots return less.
The benchmarks break down as follows:
- Major studio slot defaults: published figures cluster between 95.0% and 96.7%.
- Push Gaming: an analysis of 394 of its slots found an average of 96.58%, among the highest consistent returns of any studio.
- Legal US online slots: the average returns about 96%.
- Nevada land-based slots: a 2026 analysis from the UNLV Center for Gaming Research found an average hold of 7.15% in 2025, roughly 92.85% RTP statewide.
- New Jersey machines: state law sets a minimum return of 83%.
Keep in mind that the published defaults are only the starting point. Step 3 still applies, so the build on a given site may return less than its studio’s headline figure.
Land-based machines sit lower mostly because of cost. Floor space, hardware, staff and maintenance all come out of the hold, so physical casinos run tighter returns than online operators with lower overhead. A legal minimum such as New Jersey’s 83% is a floor, not a typical setting, but it shows how wide the permitted range can be.
Table games work differently from slots. A slot’s RTP is fixed by its math model: the player’s choices do not change the probabilities. In blackjack, the RTP depends on decisions. The 99%+ figure assumes optimal basic strategy on every hand. A player who stands on the wrong totals or splits badly plays at a lower return than the rules allow. House rules matter too, including the number of decks, dealer action on soft 17 and the blackjack payout ratio. If you are building or buying a table product, our guide to live blackjack software covers how those rule sets are configured.
Step 5: Check How 2026 Taxes and Regulation Are Lowering Casino RTP
In 2026, several UK operators lowered slot RTP settings to offset the 40% remote gaming duty, which was 15% when it was first set in 2014.
The mechanism is the version system from Step 3. A higher duty on gross gambling yield cuts the margin an operator keeps from each pound of player loss. One lever available is to switch a title from a higher certified version to a lower one. Nothing about the game looks different to the player. The help screen shows a new figure, and the long-run return drops. That is why a familiar title can pay back less in 2026 than it did when a player first tried it.
The Gambling Commission’s January to March 2026 data gives some market context. Online slot sessions lasting more than an hour fell 12% year on year to 8.9 million. Gross gambling yield per session slipped from £4.01 to £3.82. These figures describe shifts in player behavior and yield during the same period. They do not show that RTP changes caused them, and many factors affect session length and spend.
Run this recheck list whenever you look at a game, whether you play it or operate it:
- The version in use. Read the live RTP on the help screen and compare it with the provider default.
- The license’s display rule. Know whether the jurisdiction requires accurate RTP disclosure, as the UK does, or leaves it to the operator.
- Any tax change in the market. A new or higher duty is a strong signal that operators may revisit their version selection.
RTP is a long-run percentage, but the number attached to a game is not permanent. It reflects a math model, a configuration choice and the tax regime around it.
