Skip to main content
Slots guide

Volatility vs variance: are they the same thing?

Close to it. Variance measures how widely a slot's results spread around its average, and volatility is usually expressed as standard deviation, the square root of variance.

· · 4 min read

Add as a preferred source on Google
Lucky Tiger News title card: Volatility vs variance: are they the same thing?

For practical purposes you can treat them as one idea measured in two ways. Variance is the mathematical measure of how far a game's results spread around its average return. Volatility is the more familiar name, and the Gambling Commission's guidance on monitoring online games says a game's volatility is most commonly represented by its standard deviation, which is the square root of its variance. So a game with higher variance also has higher volatility. The difference is mostly about units and readability, not about two separate properties of the game.

What is variance?

Every spin has a range of possible results, each with its own probability. The game's average return per spin sits somewhere in the middle of that range. Variance takes the distance between each possible result and that average, squares it, and weights it by how likely that result is. Add those together and you have the variance.

The Commission describes variance as depending on the game's cycle and prize frequency. In plain terms, how often prizes land and how big they are relative to each other shapes how spread out the results will be.

Where does volatility come in?

Squaring the distances has a side effect: variance ends up in squared units. If you measure results in pounds, variance is in squared pounds, which has no everyday meaning. Taking the square root brings the figure back into pounds, the same unit as the stake. That square root is the standard deviation, and it is the figure the Commission's guidance uses to represent volatility.

The guidance also gives a sense of what the number describes. A highly volatile game has a larger tolerance and may be made up of prizes in the very large but rare category. A low volatility game is more predictable and is mostly made up of prizes in the small and often category.

A worked example

Here is a hypothetical example with two simple games, each costing £1 a spin and each with a stated RTP of 96%. Real slots have many more possible outcomes, but the method is the same.

Hypothetical gameResults of a £1 spinAverage return per spinVarianceStandard deviation
Game A4 in 5 spins return £1.20; 1 in 5 return nothing96p0.230448p
Game B1 in 5 spins return £4.80; 4 in 5 return nothing96p3.6864£1.92

For Game A, a £1.20 return is 24p above the 96p average and a nil return is 96p below it. Squared, those distances are 0.0576 and 0.9216. Weighting them by their chances gives 0.04608 plus 0.18432, which is a variance of 0.2304. Its square root is 0.48, or 48p.

For Game B, a £4.80 return is £3.84 above the average and a nil return is 96p below it. Squared, those are 14.7456 and 0.9216. Weighted, that is 2.94912 plus 0.73728, a variance of 3.6864. Its square root is 1.92, or £1.92.

Notice the scale. Game B's standard deviation is four times Game A's, but its variance is sixteen times as large, because variance squares the gap. That is why the square root version is the easier one to read: it tells you, in pounds, roughly how far a typical spin lands from the average.

What neither figure tells you

Both hypothetical games have the same average return of 96p per £1 staked. Over a very large number of spins, each would be expected to keep about 4p of every £1 staked. Variance and volatility don't change that margin. They describe how unevenly the results arrive along the way, not what the game costs on average.

Results are also random. The Commission requires online game outcomes to be acceptably random, and it does not permit a game to adapt its probabilities to steer results towards a target. A higher or lower volatility figure doesn't make any particular spin more or less likely to win, and it says nothing about what will happen in your next session. Over a short run, a game like Game B will usually swing much further from its average, in either direction, than a game like Game A.

One scope point. Some gaming machines in venues use a compensation system that varies the chance or size of prizes to keep returns closer to target, which gives a less volatile experience. That applies to machines in venues, not to online games, where the remote standards rule out that kind of adaptive behaviour.

How spread shrinks with more play

Volatility is also how the Commission's guidance sets the tolerance used when operators check that an online game is paying out as designed. Its example takes a game with a standard deviation of 5.6. After 50,000 games, the acceptable range is about 4.91 percentage points either side of the theoretical RTP. After 200,000 games, it is about 2.45 points. Four times as much play halves the range, which is the square root at work again.

That narrowing happens across the game's total play, often from many players. It is a test of whether the game is working as designed, not a promise that any one person's results will settle near the RTP.

Will a game tell you its volatility?

The Commission says designers calculate both the theoretical RTP and the volatility, and that these figures are reviewed during the required external testing. Before you play, the Commission requires a game to make available a description of how it works and its RTP, house edge or the probability of winning events. Volatility is not on that list, so whether a game shows a volatility figure or a label such as low or high, and how it defines that label, depends on the game's own information.

The point to keep

Variance and volatility measure the same thing: how widely results scatter around the average. Volatility, as standard deviation, is simply variance brought back into pounds. Neither changes the house margin, which stays the same however the results are spread, and neither predicts what the next spin or session will bring.

Sources: UK Gambling Commission, Live return to player performance monitoring of games of chance; UK Gambling Commission, How to calculate return to player (RTP); UK Gambling Commission, Remote gambling and software technical standards: RTS 3, rules, game descriptions and the likelihood of winning; UK Gambling Commission, Remote gambling and software technical standards: RTS 7, generation of random outcomes; UK Gambling Commission, How can gaming machines meet their %RTP if they are random? (LA Bulletin, August 2020).

How this was written. Drafted with AI assistance from the sources listed above in Harvey Watkins's style and published after automatic checks for accuracy and compliance, without a line-by-line read by an editor. If you spot a mistake, tell us. Our editorial standards.

More Guides