Auto cash-out in crash games: why every target has the same expected value

What does the auto cash-out setting do?

Spribe's in-game rules for Aviator place Auto Cash Out on the "Auto" tab of the bet panel. Once you switch it on, the game cashes your bet out by itself when the multiplier reaches the number you entered. You press nothing during the round.

One rule in the same text matters for the sums. If the round ends exactly at your chosen multiplier, the bet is lost. Spribe's own example sets auto cash-out at 1.5x, the plane leaves at 1.5x, and the stake is gone.

So the crash point has to be above your number for the bet to pay. Manual cash-out follows the same payout rule, and the guide on what cash out means in a crash game covers the button itself. This page is about the number you type into the auto box.

Why does every target have the same expected value?

Spribe states that Aviator's return to player (RTP) is 97%. Spribe does not publish how crash points are distributed, so the arithmetic here uses the standard crash model. In that model a round reaches multiplier m with chance r/m, where r is the RTP, as shown in the guide to the probability of reaching any multiplier.

Expected value is what a bet pays when it wins, times the chance that it wins. With auto cash-out at m, a winning bet pays m times the stake, and it wins with chance r/m. Multiply the two and the m cancels: m x (r/m) = r.

Illustrative example: you stake KES 100 with auto cash-out at 2x. The bet pays KES 200 in 48.5% of rounds, so on average it returns 200 x 0.485 = KES 97. At 10x it pays KES 1,000 in 9.7% of rounds, and 1,000 x 0.097 is KES 97 again.

Read the table across each row. The chance falls as the target rises, the payout rises by the same factor, and the average return stays at KES 97 every time. Only the last column moves, and that column is the real effect of your choice.

What does a low target change, if not the expected value?

A low target changes variance, which is how far a session can land from the average. At 1.10x you win most rounds, but each win adds only KES 10 to a KES 100 bet. One crash below 1.10x costs KES 100, the profit of ten wins.

A high target works the other way round. At 10x most bets lose the full stake, and the rare win is large. The spread of results is wide, yet its centre is the same 3% loss.

Illustrative example: the chance of 20 rounds in a row all clearing 1.10x is 0.8818 to the power 20, about 8.1%. In roughly 92 of every 100 such stretches, at least one KES 100 loss turns up. A streak of small wins feels steady, but over 20 rounds the crash that wipes it out arrives more than nine times in ten.

The chances come from the binomial distribution, worked out for each row. Every row has the same average result, a loss of KES 300, which is 3% of KES 10,000. What changes is the shape: low targets give many small wins and a steady drain, high targets give more heavy losses and a few big gains.

One row surprises people. The 1.10x target finishes 100 rounds ahead less often than 2x does, because each loss there needs ten wins to repair.

Can a system of changing targets come out ahead?

No fixed target escapes the 3% edge, so no mix of targets escapes it either, because expected values add up. Spribe's rules let you place two bets in one round through a second bet panel. Each bet still returns 97% of its own stake on average, and the pair returns 97% of the total.

Illustrative example: KES 50 at 1.5x and KES 50 at 10x in the same round. The first returns 0.97 x 50 = KES 48.50 on average, the second also KES 48.50, so the KES 100 total comes back as KES 97 on average.

Waiting for a run of low rounds before raising your target does not help. Spribe's rules say each round's result is generated from that round's server seed and the client seeds of its first three bettors, so past rounds carry no information about the next one. Raising the stake after a loss only makes the swings bigger, as the guide on why no method wins over the long run sets out with numbers.

What the 97% figure does and does not tell you

RTP describes the average over a very large number of bets. Over 100 rounds the second table already shows results spread widely around the average loss, and over one evening anything in that spread can happen. The 97% RTP figure explained in shillings covers the long-run idea in more depth.

In the game's rules text, the "Return to Player" line is a template that the game fills in with a number. Read the figure in the rules panel of the game you actually open. If it is lower than 97%, the average loss in every row above rises to match.

To test a target of your own, the crash multiplier calculator gives the chance of reaching any cash-out. Whatever number you try, the chance and the payout move against each other.

What can auto settings do for you?

Auto settings cannot change the edge, but they can hold you to limits you chose in advance. Spribe's Auto Play panel offers stop conditions such as "Stop if cash decreases by" and "Stop if single win exceeds". A stop on losses caps what one session can cost.

Kenyan rules add tools on the operator's side. Regulation 84(2) of the Gambling Control (Conduct of Gambling Operations) Regulations, 2026 requires a licensee to offer deposit limits on a daily, weekly and monthly basis, loss, session and expenditure limits, and reality-check pop-ups. Section 71(1) of the Gambling Control Act, 2025 sets the minimum online bet at twenty shillings.

Limits are the one setting that changes what a session can cost. The guide to limits and self-exclusion for crash games in Kenya lists the options and where to get help.