Crash games
Cashout timing in crash games
Cashout timing in a crash game comes down to one choice: the target multiplier. That choice sets how often a bet pays and how much it pays, but under the standard model it does not change the long-run return.
Key takeaways
- A 2019 paper in Scientific Reports by Wang and Pleimling describes crash as a fixed-odds betting game in which the player selects the odds.
- At an assumed RTP (return to player) of 95.5%, a 1.50x target is reached in about 63.7% of modeled rounds and a 10.00x target in 9.55%, and both return 0.955 per unit staked on average (our calculation, not published Spaceman data).
- Under the constant-RTP model, once a round has reached a multiplier a, the chance that it goes on to reach b is a ÷ b: from any point, doubling is a 50% chance.
Cashout timing is a choice of target multiplier
A crash round looks like a test of nerve: the multiplier climbs and each player decides when to leave. But in the games whose rules are documented, the crash point is fixed before the multiplier starts moving, as how crash games work explains. The only open question is which multiplier a player will accept, and a bet pays only if the round is still running when the multiplier gets there.
Wang and Pleimling, in a 2019 paper in Scientific Reports, describe crash in the same terms: a fixed-odds betting game in which the player selects the odds. Aiming for 2.00x is one set of odds; aiming for 10.00x is another. “Timing” is the everyday name for that selection.
The trade-off: how often versus how much
Under the reach-probability model, the chance that a round reaches a multiplier x is roughly RTP ÷ x. The table applies that rule to a stake of 1 unit at an assumed RTP of 95.5%.
| Target | Modeled chance it is reached | Profit when reached | Average return per unit |
|---|---|---|---|
| 1.50x | 63.7% | +0.50 | 0.955 |
| 2.00x | 47.75% | +1.00 | 0.955 |
| 3.00x | 31.8% | +2.00 | 0.955 |
| 5.00x | 19.1% | +4.00 | 0.955 |
| 10.00x | 9.55% | +9.00 | 0.955 |
| 20.00x | 4.78% | +19.00 | 0.955 |
| 50.00x | 1.91% | +49.00 | 0.955 |
| 100.00x | 0.955% | +99.00 | 0.955 |
Note
These are model figures, not measured Spaceman frequencies: Pragmatic Play does not publish Spaceman’s multiplier distribution. The assumed 95.5% is the theoretical RTP in Spaceman’s Game Help (sources differ).
A missed round costs the full 1 unit stake whatever the target was. Moving down the table, wins become rarer and larger in step, and the last column never moves.
Why no target changes the long-run return
Over 100 rounds at 1 unit each, the model’s averages of 63.7 wins of 1.50 units and 9.55 wins of 10.00 units both come to about 95.5 units back: an average loss of 4.5 units, which is the house edge at the assumed RTP and is identical at every target.
That average only shows over a very large number of rounds. Across 100 rounds the actual result can sit a long way from −4.5 units, and how far depends heavily on the target chosen.
Why watching the curve adds no information
Because the crash point is already fixed, nothing visible during the round says where it will stop: not the speed of the climb, not how long it has lasted, not what other players do. A crash round is a shared event, but each player bets against the house, not against the others.
The model makes this precise. If the chance of reaching x is RTP ÷ x, then once the multiplier has reached a, the chance that it goes on to reach b is the second reach chance divided by the first. The RTP cancels out (our calculation):
chance of reaching b, once the round has reached a = a ÷ b
- At 1.50x, the chance of going on to 3.00x is 50%.
- At 2.00x, the chance of going on to 4.00x is 50%.
- At 5.00x, the chance of going on to 10.00x is 50%.
From any point, doubling is a coin flip in the model, and the amount at risk has grown to match: 2.00 units available now against a 50% chance of 4.00. On average, staying in neither gains nor loses; it only widens the spread of results. The house edge was taken earlier, in the rounds that end before any target is reached.
So a multiplier that has “got this far” has no momentum, and a slow start is not a warning. Recent results do not help either: past rounds do not influence future ones, and reading patterns into them is one of the common crash game myths.
Manual and pre-set cashouts
A target can be executed by clicking during the round or by setting the multiplier in advance as an auto cashout. Neither route changes the expected return under the model. A pre-set target takes reaction time and connection delay out of the execution; a manual click keeps the decision open until the last moment.
How each route settles depends on the game. In Spaceman, according to the Game Help, a manual cashout is settled at the multiplier in force when the game server processes the request, while an auto cashout pays at exactly its preset value; the Spaceman cashout page covers settlement and the auto cashout page covers the settings.
What the choice of target does change
What differs between targets is the shape of the results. A low target produces frequent small wins and shorter losing runs; a high target produces longer stretches of lost stakes with an occasional large win. The standard deviations and losing-streak figures are on the volatility and risk page.
A partial exit, such as the 50% cashout in Spaceman, splits one bet across two targets and lands between them in spread. Named plans — fixed targets, ladders, doubling after losses — are examined on the Spaceman strategy page.
Keep in mind
Timing changes variance, not expected return. With a negative expected return at every target, the long-run cost of play grows with the total amount staked, however the cashouts are timed.
Sources
- Wang and Pleimling, Online Gambling of Pure Chance: Wager Distribution, Risk Attitude, and Anomalous Diffusion, Scientific Reports 9, 14712 (2019)
- Pragmatic Play — Spaceman Game Help, version 1.7 (change log to 10 July 2024), as distributed to licensed operators (cited by name, not linked)