Contents
RTP and volatility answer different questions. RTP shows the expected payout relative to the amount staked. It is not the probability of winning and not the chance of finishing a session in profit. Volatility describes the spread of results around the average, but it does not fully define the payout distribution, the exact hit frequency, or any sequence of wins and losses.
In short, RTP is about the average value, while volatility is about the spread around that average. So a 97% RTP slot does not have to be "better" than a 95% slot in one specific session, and 96% RTP by itself does not tell you how a run of spins will end.
TL;DR — what matters most
| Metric | What it shows | What it does not show |
|---|---|---|
| RTP | The theoretical share of turnover returned to the player | The outcome of your specific session |
| Volatility | The spread of results | Exact odds of a big win |
| Variance | The average squared deviation, in squared stake units | Whether a slot is "good" |
| Standard deviation | The spread, in stake units | Jackpot probability |
RTP: return on stakes, not profit probability
RTP is the theoretical return to player as a percentage of turnover. If RTP is written as a percentage, expected loss is:
expected loss = turnover × (1 - RTP / 100)
It is important to separate stake, turnover, and deposit. Turnover is the total of all stakes during a session, including winnings that are staked again. Deposit is the money added to your balance. RTP is calculated from turnover.
If you play 1,000 spins at a stake of 10, turnover is 10,000. At 96% RTP, expected loss is 400. That is not a prediction for one session. It is the average expectation for that turnover.
| RTP | Expected loss at 10,000 units wagered |
|---|---|
| 94% | 600 |
| 95% | 500 |
| 96% | 400 |
| 97% | 300 |
| 99% | 100 |
Two comparisons are especially useful here:
- 95% vs 97%: expected loss is 40% lower with the same turnover.
- 96% vs 99%: expected loss is 75% lower with the same turnover.
That is not a promise of profit and not a promise that you will "win it back" in a single session. It is only a difference in expected value.
Turnover is the key metric
RTP is calculated from the sum of stakes, not from the deposit. Different turnover changes expected loss, but it does not determine the actual result of one session. Two players can get very different outcomes at the same RTP, even under the same rules.
Is 96% RTP a good or bad return level?
On its own, 96% RTP is neither good nor bad. It is simply a theoretical return level. To judge it properly, you need to look at both turnover and result spread, not hunt for one universally "best" percentage.
A 96% RTP slot has lower expected loss than a 95% RTP slot at the same turnover, but that does not mean a short session will necessarily feel better. Over a small number of spins, randomness can easily overwhelm a gap of 1 to 3 percentage points.
Actual RTP and why a small sample is not enough
Actual RTP is calculated as total payouts / total stakes × 100%. For example, if total payouts are 900 and total stakes are 1,000, the actual RTP on that sample is 90%. But that alone does not prove the game's theoretical RTP changed. On small samples, actual results can deviate sharply from expectation.
The right sample size depends on payout spread and on the level of precision you want. There is no universal spin count after which actual RTP must match theoretical RTP. For terminology and measurement approach, you can check the regulator materials on key terms relating to live return to player performance monitoring of games and what volume of play should be achieved before measuring the actual RTP.
Volatility, variance, and standard deviation
Volatility is a broad description of how widely results can swing. If you want a fuller terminology breakdown, see our separate guide on slot volatility. Variance is the average squared deviation, and standard deviation is the square root of variance. If returns are measured in stake units, variance is in squared stake units, while standard deviation is in stake units.
Marketing labels such as "low," "medium," or "high" volatility are not a single strict scale shared by every game. Also, volatility does not fully determine the shape of the distribution, and one number alone cannot recover the exact frequency of payouts.
Two toy models for one spin with the same RTP
These are not real slots and not measurements of a real game. They are invented teaching examples. Assume a unit stake of 1 per spin, and payouts include the returned stake.
Model A
- payout 1.2 with probability 80%
- payout 0 with probability 20%
- mean payout: 0.96
- RTP: 96%
- standard deviation: 0.48
Model B
- payout 12 with probability 8%
- payout 0 with probability 92%
- mean payout: 0.96
- RTP: 96%
- standard deviation: 3.2555
The point is simple: the same RTP does not mean the same payout distribution. Model A is much smoother. This is a single-spin distribution at unit stake, not a statement about how results must unfold over time.
| Model | Full payout distribution | Mean payout | RTP | Standard deviation |
|---|---|---|---|---|
| A | 1.2 with probability 0.8; 0 with probability 0.2 | 0.96 | 96% | 0.48 |
| B | 12 with probability 0.08; 0 with probability 0.92 | 0.96 | 96% | 3.2555 |
Low volatility guarantees nothing
Even a smoother game does not promise frequent wins and does not protect you from drawdowns. It only means a narrower spread, not protection from losses.
A historical example of RTP settings and volatility rating
In the public copy of Play'n GO's "Game Rules and Settings — Book of Dead," GRS0595 dated September 14, 2022, page 4 lists RTP settings of 96.21%, 94.25%, 91.25%, 87.25%, and 84.18%, plus a volatility rating of "10 VERY HIGH." This is a historical example, not proof of the current version at any casino. It also matters that this provider scale is not 10% and not a standard deviation equal to 10.
Book of Dead: RTP settings and volatility rating in Play'n GO document GRS0595, page 4
Excerpt from page 4 of the document dated September 14, 2022. Public PDF copy. Historical example only, not confirmation of any casino's current settings.
95% and 97%: which matters more?
If you compare RTP alone, 97% is mathematically better than 95% at the same turnover. But that still does not answer how one specific session will go.
What really changes:
- at the same turnover, expected loss is lower at 97%;
- the difference appears in expected value, not in any guaranteed outcome;
- in a short run, results can still differ sharply because of randomness.
So the phrase "97% is better than 95%" is true only as a comparison of theoretical expectation at the same turnover.
96% and 99%: how big is the gap at 10,000 units wagered?
Here the gap is much more noticeable: at 10,000 units of turnover, 99% RTP gives an expected loss of 100, while 96% gives 400. So the expected-loss difference is 300, or 75% relative to 96%. Even here, 99% is not a "99% chance to win." It is only the theoretical return on turnover.
But do not jump to the wrong conclusion. A higher RTP does not mean a session is more likely to finish in profit. It only reduces average expected loss at the same turnover.
What matters more in a short session?
In a short session, both metrics matter. RTP sets the expected value at a fixed turnover, while volatility shows the spread around that average. There is no universal rule that one is always more important.
That does not mean you should keep playing until RTP somehow "kicks in." There is no such rule. In both short and long sessions, actual results can deviate a lot from the average. If you want a wider comparison with other game types, our article on crash-game volatility vs slots is a useful companion.
Can volatility tell you the jackpot chance?
No. Volatility and max win do not determine jackpot probability.
Why not:
- volatility describes the overall spread of payouts;
- max win shows the upper ceiling;
- jackpot frequency is a separate characteristic and cannot be derived directly from those two alone.
A game can have a very high max win and still make the path to it extremely rare. Another may look less extreme and still offer no easy route to a big result.
Max win is not jackpot probability
Max win shows the payout ceiling, not the probability of reaching it. To estimate probability, you need the game rules and its mathematical model.
Where to check a specific slot's RTP
You need the rules for the exact game version you have open. If you want a reference point, you can use the slot RTP database, but it is still a database, not proof of the settings currently used by a specific operator.
How to use the RTP calculator
In our RTP calculator, choose the mode by number of spins, enter 1,000 spins, a stake of 10, then test 96% RTP and 99% RTP separately. Turnover is 10,000 in both cases, and expected losses are 400 and 100.
That is a mathematical expectation calculation, not a prediction of a session result.
RTP calculator: 1,000 spins at 10, 96% RTP, expected loss 400
RTP calculator: the same turnover at 99% RTP, expected loss 100
Screenshots of the local ToolsGambling calculator interface. They illustrate expected-value calculations, not gambling results.
What not to mix into one conclusion
A few mistakes show up again and again:
- RTP is not profit probability. It is a turnover-based metric, not a session-success rate.
- Volatility is not hit frequency. Win frequency may relate to spread, but it is not reducible to it.
- A high max win does not mean a high jackpot chance. Ceiling and probability are different things.
- You cannot transfer one version's RTP to another without checking. Configurations can differ.
Bottom line
Slot RTP and slot volatility are two different characteristics.
- RTP describes theoretical return on turnover and expected loss.
- Volatility, variance, and standard deviation help describe result spread.
If you want the short answer on what matters more, it depends on the task:
- for a short session, you need to understand both RTP and volatility;
- for comparing games at the same turnover, RTP is more useful;
- for judging jackpot odds, neither RTP nor max win is enough on its own.
The best way to compare slots is to use both metrics together, not one attractive percentage on its own.
Frequently Asked Questions
By itself, neither. It is a theoretical return on turnover. With the same turnover, 96% means an expected loss of 4% of the total amount staked. In a short session, that is not enough to predict your actual result.
RTP shows what share of turnover a slot returns on average over the long run. It is not the probability of finishing a session in profit and not the chance of winning on a single spin.
If turnover is the same, 97% means an expected loss that is 40% lower than 95%. But that is not a guarantee of a better session and not a promise of a win.
With the same 10,000-unit turnover, 99% gives an expected loss 75% lower than 96%. But 99% is not a 99% chance to win, and a short session can still differ sharply from theory.
Both matter. RTP sets the expected value for a given turnover, while volatility describes how widely results can swing around that average.
No guarantee. Marketing volatility labels are not a single certified scale, and low volatility does not promise frequent wins or protection from losses.
No. Max win shows the payout ceiling, not the probability of reaching it. Max win size and volatility do not, by themselves, determine jackpot odds.
Check the rules for the exact version of the game you have open, because RTP and settings can differ. A calculator can help with a quick estimate, but it shows expectation, not a guarantee.
Evgeniy Volkov
Verified ExpertFullstack developer with a background in mathematics. I build the calculators and game-style tools on ToolsGambling with Pixi.js and modern web tech, and every result uses transparent probability formulas you can verify yourself.