Aviator and Crash Games: The Fixed Edge Behind the Cash-Out Choice

Aviator is a crash game in which a multiplier rises from 1.00x and can stop at any moment. The player sets a cash-out point in advance or presses a button during the round. If the multiplier reaches that point before the round crashes, the stake is returned multiplied by that figure. If the round crashes first, the stake is lost. The format is sometimes listed under slots on operator lobbies, and the term aviator slot is used loosely, but the arithmetic is not slot arithmetic. There are no reels and no paylines. There is a single random number per round, a curve drawn from it, and a cash-out decision made by the player.

The appeal is obvious. The player appears to control the outcome, because the player chooses the number that decides the bet. That appearance is the whole product. The house edge does not sit in the cash-out choice. It sits in the curve that generates the multiplier, and it is the same at 1.01x as it is at 100x.

The payout curve and where the 1% comes from

A crash game publishes a multiplier for each round. The multiplier is generated from a random value, and the mapping from that value to the multiplier is set so that the expected return on any cash-out target is a fixed percentage below the stake.

The clean way to state it: for a cash-out target of m, the probability that the round reaches m is approximately

P(reach m) = (1 - edge) / m

where edge is the house edge built into the curve. If the edge is 1%, then

P(reach m) = 0.99 / m

The expected return on a bet cashed out at m is the payout m multiplied by the probability of reaching m:

E = m × (0.99 / m) = 0.99

That is 99% of the stake, or a 1% house edge, for every m. The m cancels. This is the fact the category’s marketing works hard to obscure: the cash-out point changes the variance, not the mean.

A worked example with numbers a reader can check. Suppose the stake is 10.00 and the cash-out target is 2.00x.

• Probability the round reaches 2.00x: 0.99 / 2.00 = 0.495, or 49.5%.
• If it reaches 2.00x, the return is 10.00 × 2.00 = 20.00, a profit of 10.00.
• If it does not, the return is 0.00, a loss of 10.00.
• Expected return: (0.495 × 20.00) + (0.505 × 0.00) = 9.90.
• Expected loss per bet: 10.00 − 9.90 = 0.10, which is 1% of 10.00.

Now the same stake at 10.00x.

• Probability the round reaches 10.00x: 0.99 / 10.00 = 0.099, or 9.9%.
• If it reaches 10.00x, the return is 100.00, a profit of 90.00.
• If it does not, the return is 0.00, a loss of 10.00.
• Expected return: (0.099 × 100.00) + (0.901 × 0.00) = 9.90.
• Expected loss per bet: 0.10, still 1% of 10.00.

The two bets feel nothing alike. One wins about half the time. The other wins about one time in ten. The expected loss is identical.

What the figure depends on

The 1% figure is not a universal constant. It is a parameter of the specific game version and the specific operator configuration. The table below sets out the conditions that change it.

If a page states a house edge for aviator without naming the version and the operator configuration, the figure is not reproducible. The same applies to any aviator game online. The edge is a setting, not a law of nature.

The cost in money per hour

A percentage of turnover is not a quantity anyone can feel. The following converts the edge into an hourly figure at a stated stake and round rate.

Assume a stake of 1.00 per round and 200 rounds per hour. That is 200.00 of turnover per hour. At a 1% edge, the expected loss is 1% of 200.00, which is 2.00 per hour.

At a stake of 5.00 per round and 200 rounds per hour, turnover is 1,000.00 per hour. Expected loss is 10.00 per hour.

At a stake of 10.00 per round and 200 rounds per hour, turnover is 2,000.00 per hour. Expected loss is 20.00 per hour.

The round rate matters as much as the stake. A crash game with a short auto cash-out can complete a round every few seconds. If a player completes 400 rounds per hour at 1.00 per round, turnover is 400.00 and expected loss is 4.00 per hour. If the player completes 100 rounds per hour at 5.00 per round, turnover is 500.00 and expected loss is 5.00 per hour.

The hourly figure is an expectation, not a schedule. In any given hour the result will be above or below it. The figure describes the average over a very large number of hours, and it is the number that determines whether the activity is affordable at the chosen stake and pace.

What people get wrong about aviator

The most common mistake is the belief that the cash-out choice is a skill that can move the expected return. It is a natural mistake because the player genuinely does make a decision that determines whether the bet wins or loses. In a game of pure chance with a fixed payout curve, controlling the win condition is not the same as controlling the expectation. The player controls the shape of the distribution. The provider controls the mean.

A second mistake is the belief that a round is due to crash early or late because of the previous rounds. Each round is generated independently. A run of low multipliers does not make a high multiplier more likely, and a run of high multipliers does not make a crash more likely. The display of past rounds is a record, not a forecast.

A third mistake concerns provably fair hashing. Crash games commonly publish a server seed hash before the round and the server seed after, so the player can verify that the multiplier was determined by the seed and not altered after the bet was placed. That is a real property and it is worth understanding. It proves the round was not changed after the fact. It proves nothing about the edge. A game can be perfectly provably fair and still return 99% of turnover, or 95%, or 90%. The hash verifies integrity. The edge is a separate parameter, set in the curve, and it is not visible in the hash.

A fourth mistake is treating a crash game as a slot. The term aviator slot appears in some listings because the game is grouped with slots on the lobby. The mechanics are different. A slot has a fixed set of outcomes and a published return to player for the version. A crash game has a continuous multiplier and an edge parameter. The two are not interchangeable, and a comparison between them has to be made on the edge, not on the label.

The aviator bet and the aviator betting game

The aviator bet is a single stake placed on a single round. The aviator betting game is the sequence of those bets. The distinction matters because the hourly cost is a function of the sequence, not the single bet. A player who places one 1.00 bet and stops has an expected loss of 0.01. A player who places 200 such bets in an hour has an expected loss of 2.00. The game is the pace.

Some versions allow two simultaneous bets in the same round, with separate cash-out targets. This doubles the stake per round and therefore doubles the expected loss per round, while leaving the edge unchanged. It is a way to increase the amount at risk, not a way to improve the return.

Auto cash-out is a convenience. It removes the need to press a button at the right moment. It does not change the probability of reaching the target, because the target is the same. A player who sets auto cash-out at 2.00x and a player who manually cashes out at 2.00x have the same expected return.

Spribe aviator and the version question

Aviator is most commonly associated with the provider Spribe. Spribe aviator is the original version in most markets, and the edge parameter is set by the operator within the provider’s configuration. Other providers have released crash games with the same format and different edges. A page that states a single figure for all crash games is stating a figure for one configuration and applying it to all of them.

The honest position is that the edge is typically around 1% in the versions that publish it, and that the exact figure depends on the version and the operator. A reader who wants the figure for a specific game should look for the published return to player or the game rules, and should treat a missing figure as a reason to assume the edge is not in the reader’s favour. The house does not build games with a negative edge for itself.

The cash-out choice is real. The control is real. The expectation is fixed. That is the whole of the arithmetic, and it is the same arithmetic at every multiplier on the curve.