Crash games
House edge and RTP in crash games
The crash game house edge is the share of every stake the game keeps on average over the long run; RTP (return to player) is the share it pays back. The two always add up to 100%, and neither says what will happen in a single session.
Key takeaways
- Published figures differ: bustabit states a 1% house edge, Spribe lists Aviator at 97% RTP, and Pragmatic Play’s own documents give two figures for Spaceman, 95.50% in the Game Help and 95% on its product page (as checked on 2 October 2026).
- In a UK Gambling Commission worked example for licensees (a game with volatility 5.6), the acceptable gap between actual and theoretical RTP is still ±4.91% after 50,000 games and ±1.10% after 1,000,000.
- The long-run average cost of play is the total staked multiplied by the house edge: 200 rounds at 1 unit come to 2 units on average at a 1% edge and 10 units at a 5% edge, with wide variation around either.
House edge and RTP: two views of one number
house edge = 100% − RTP
An RTP of 97%, for example, and a house edge of 3% are the same fact stated from two sides. Over a very large number of rounds, about 97 of every 100 units staked come back as payouts and about 3 stay with the house.
Which figure gets quoted is a matter of convention. Spribe and Pragmatic Play quote an RTP for their games; bustabit states its house edge.
Neither is a fee deducted from each round. A crash bet returns either a multiple of the stake or nothing, so the edge exists only as an average across many rounds.
Theoretical RTP versus actual RTP
The figure in a game’s help text is the theoretical RTP: the return implied by the game’s math. Actual RTP is measured after the fact. The UK Gambling Commission’s guidance for licensees defines it as wins divided by turnover: total paid out over total staked.
The two converge only with volume. The Commission’s consumer guide, written about gaming machines, says that RTP is an average achieved over a significant number of game plays, not each time the game is played.
The licensee guidance works through an example game with a volatility figure of 5.6 and shows the acceptable gap between actual and theoretical RTP narrowing as plays accumulate:
| Games played | Acceptable deviation |
|---|---|
| 50,000 | ±4.91% |
| 400,000 | ±1.74% |
| 1,000,000 | ±1.10% |
Note
These tolerances belong to the Commission’s example game, not to Spaceman or any other crash game. What they show is the scale: after 50,000 plays, the example’s measured RTP could still sit almost five percentage points either side of the theoretical figure. One person’s session is far smaller than that.
Published figures for well-known crash games
The third column of the table is our arithmetic from the formula above.
| Game | Published figure | Implied counterpart | Stated by |
|---|---|---|---|
| bustabit | House edge 1% | RTP 99% | bustabit’s published documentation |
| Aviator | RTP 97% | House edge 3% | Spribe’s Aviator page |
| Spaceman | Theoretical RTP 95.50% | House edge 4.50% | Spaceman Game Help, version 1.7 |
| Spaceman | RTP 95% | House edge 5% | Pragmatic Play product page, as checked on 2 October 2026 |
| JetX | Not published by the provider | — | SmartSoft Gaming’s JetX page lists no RTP |
Spaceman appears twice because two Pragmatic Play documents give two figures, and the provider does not explain the difference. The page about Spaceman’s RTP figures sets out what each source says, including a third figure, 96.50%, that appears only on review sites with no primary source.
For JetX, SmartSoft publishes no RTP, so no RTP, house edge or expected value can be verified from provider-published information.
A lower house edge means a lower average cost per unit staked, and that is all it means. Maximum multipliers, features and fairness models are set side by side in the comparison of Spaceman with other crash games.
Where the edge sits and what it does not depend on
In the standard model of a crash game, the edge is not shaved off each cashout. It comes from the share of rounds that end at the floor before any target can pay, and the chance of reaching a target is the RTP divided by the target, so every target returns the same on average. The house edge therefore does not depend on when a player cashes out.
Both points are worked through under the constant-RTP model. For Spaceman it is a model, not a distribution published by Pragmatic Play.
What a crash game house edge costs: a worked example
The long-run average cost of play is a three-part product:
average cost = stake per round × number of rounds × house edge
Take 1 unit per round for 200 rounds. The total staked is 200 units, and the average cost depends only on the edge:
| House edge | Figure taken from | Total staked | Average cost |
|---|---|---|---|
| 1% | bustabit | 200 units | 2 units |
| 3% | Aviator, 97% RTP | 200 units | 6 units |
| 4.50% | Spaceman, 95.50% RTP in the Game Help | 200 units | 9 units |
| 5% | Spaceman, 95% RTP on the provider’s product page | 200 units | 10 units |
Every row is a loss on average. The table is arithmetic, not a ranking of games and not a recommendation to play any of them.
The cost follows the total staked, not the starting balance. Winnings that are staked again count again. A balance of 20 units can generate 200 units of stakes, and the edge applies to all 200.
It is an average, not a forecast. Over 200 rounds the actual result is spread widely around it. At a 2.00x target under the model at an assumed RTP of 95.5%, the standard deviation over 200 one-unit rounds is about 14 units, against an average cost of 9 (our calculation). The size of those swings, and how they depend on the target, is the subject of the page about volatility in crash games.
What RTP cannot tell you
- How volatile the game is. Two games with the same RTP can swing very differently, and in a crash game the swing depends on the target chosen.
- How often a bet wins. That also depends on the target. A high RTP is compatible with losing most rounds.
- Whether the game is implemented as stated. RTP is a design figure. Checking that a game behaves according to its rules is the job of testing and regulation, described under RNG testing and certification.
- Which figure applies where a game is offered. Under the UK Gambling Commission’s technical standards, information on the likelihood of winning, such as RTP or house edge, must be available to customers of British-licensed operators. Elsewhere, the in-game help remains the place to look.
Sources
- UK Gambling Commission — Return to player: how much gaming machines payout
- UK Gambling Commission — How to calculate return to player (RTP)
- UK Gambling Commission — RTS 3: Rules, game descriptions and the likelihood of winning
- Pragmatic Play — Spaceman product page (as checked on 2 October 2026)
- Spribe — Aviator game page
- SmartSoft Gaming — JetX game page
- Pragmatic Play — Spaceman Game Help, version 1.7 (change log to 10 July 2024), as distributed to licensed operators (cited by name, not linked)
- bustabit — published documentation (cited by name, not linked)