Volatility is the number that determines whether your session goes the way you planned, and it is the one players most often skip. RTP gets the attention because it is a percentage and percentages look decisive. But RTP describes an average across millions of spins that no individual will ever reach, while volatility describes what happens across the few hundred spins you are actually going to play.
Put differently: RTP tells you what the game does over a lifetime. Volatility tells you what it does tonight.
Volatility is a description of distribution shape
Two games can both return 96% and behave nothing alike. One returns it in small amounts, frequently. The other withholds it for long stretches and returns it in rare, large bursts. Same average, entirely different experience, and entirely different bankroll requirement.
Statistically, volatility is the standard deviation of returns per spin. Practically, it is expressed by three related properties: how often any win lands, how large the biggest possible win is, and how much of the total return is concentrated in rare events.
Providers publish a volatility rating alongside RTP — usually as a rating out of five, or as a low/medium/high label. The ratings are not standardised across studios, which limits their comparability, but they are consistent within a studio’s catalogue and correlate reliably with the underlying maths. Platforms displaying both figures in the game preview, Caswino casino online among them, make the pairing checkable before you commit, which matters because the two numbers only mean something together.
Hit frequency and return are independent
This trips people up constantly, so it is worth separating cleanly.
Hit frequency is the percentage of spins returning anything at all. A game with 40% hit frequency pays something on two spins in five.
Return is what those payments add up to relative to what you staked.
A high hit frequency does not imply a good return. Many low-volatility games pay frequently but pay less than the stake on a large share of those hits — a 40% hit frequency where half the hits return 0.4x the bet produces a lot of visible wins and a steadily declining balance.
That is the single most useful thing to understand about low-volatility design: frequent small returns feel like the game is paying, and a substantial proportion of them are net losses presented as wins. Counting spins that return something tells you very little; counting spins that return more than the stake tells you considerably more.
The tiers compared
| Volatility | Hit frequency | Typical max win | Feature trigger | Bankroll for 500 spins | Session character |
|---|---|---|---|---|---|
| Low | 35–45% | 500–1,000x | 1 in 80–120 | ~150x base bet | Steady, slow drift |
| Medium | 25–35% | 2,000–5,000x | 1 in 130–200 | ~300x base bet | Balanced swings |
| High | 20–28% | 10,000–25,000x | 1 in 200–300 | ~600x base bet | Long flat stretches |
| Very high | 15–22% | 25,000–50,000x+ | 1 in 250–400 | ~1,000x base bet | Extended droughts, rare spikes |
The bankroll column is the practical output of the whole table. It is an approximation of what you need in order to reach the game’s characteristic behaviour rather than running out before it appears — and on very high volatility titles it is roughly seven times what a low-volatility game requires at the same stake.
Playing a 40,000x-cap game at a stake sized for a low-volatility session is the most common mismatch in slots. The maths is not unfavourable; the bankroll simply cannot survive to the point where the game does what it was built to do.
How long dry spells actually run
The arithmetic here is worth seeing, because “the game isn’t paying” usually describes something entirely ordinary.
On a game with a 20% hit frequency, the chance of twenty consecutive spins returning nothing is 0.8²⁰, or about 1.2%. That sounds rare — until you consider that a 600-spin session contains hundreds of overlapping twenty-spin windows, which makes at least one such run very likely rather than exceptional.
Feature droughts run longer. If a bonus triggers once in 200 spins on average, the probability of not seeing it in 600 spins is roughly 5% — meaning one session in twenty on a high-volatility game ends without the feature appearing at all, despite three times the average wait.
Neither of these indicates anything about the game. They are what the stated hit frequency looks like when it is actually played, and expecting otherwise is the same sample-size error that makes short demo runs uninformative.
Matching volatility to bankroll
The working formula is simple:
Base stake = session budget ÷ bankroll multiplier for that volatility tier
A $200 budget on a very high volatility game: $200 ÷ 1,000 = $0.20 per spin. The same budget on a low-volatility title: $200 ÷ 150 = around $1.30.
That difference surprises people, and it is the correct direction. Higher volatility requires a smaller stake, not a larger one, because survival to the rare event is the entire requirement. Players routinely do the opposite — raising stakes on high-variance games because the potential payout is larger — which guarantees the balance ends before the distribution has a chance to express itself.
If the resulting stake is below the game’s minimum, that is a clear signal to choose a different game rather than to play the one you wanted at an unsuitable stake.
Where volatility ratings mislead
Four limitations worth knowing:
- No cross-studio standard exists. One provider’s “high” is another’s “medium”. Ratings are comparable within a catalogue, not between catalogues.
- A single rating hides a bimodal distribution. Some games are low-volatility in the base game and extremely high-volatility in the feature. The average rating describes neither accurately.
- Max win caps distort the top end. A game with a 50,000x ceiling concentrates a meaningful share of its return in outcomes almost nobody will see, which raises measured volatility without changing the practical experience for most sessions.
- Feature buys change the profile entirely. Buying into the bonus removes the base game from the equation, converting a medium-volatility title into a very high-volatility one at a hundred times the stake per action.
The general approach that follows: read the volatility rating, note the max win cap and the published feature trigger rate, and size your stake from the bankroll multiplier rather than from what feels normal. Those three numbers together predict how a session will feel far better than RTP does, and all three are printed in the game info panel before you spend anything.
A closing note that belongs in any honest piece on this subject: volatility describes how a negative expected return arrives, not whether it is negative, and no volatility tier is more favourable than another over time. Set a budget you are entirely comfortable losing, size your stake to survive the session you planned, and stop when you reach the figure you set. Gambling is for adults 18+ only, and free confidential support services are available in most countries.