Slot volatility and RTP simulator

Slot volatility and RTP simulator

Return to player (RTP) is the long-run share of wagers a machine pays back. Volatility is how that return is distributed: many small wins or rare large ones. Every machine here has the same RTP, and the Chill and Degen modes change only the shape of the payouts.

Spin on Chill, then switch to Degen and watch the session RTP line. It wanders far more on Degen before it settles toward the theoretical line. The table below has the exact figures for this machine.

Open this experiment in the lab ↑

Analysis

All numbers below were computed by the same simulator and models that run in the lab above, when this page was built.

Volatility at a fixed RTP

Exact figures for Classic Fruit, computed from the reel weights and paytable. Chill and Degen use the same 94.0% return to player; only the distribution of wins changes.

ModeRTPHit rateStd dev per spinTop line paysBonus chance
Chill94.0%19.2%×7.2×5010.9%
Degen94.0%16.6%×11.7×3720.9%
RTP
RTP = Σ paid ÷ Σ wagered
Std dev per spin
σ = √(E[pay²] − E[pay]²)

Questions

Does higher volatility mean a lower RTP?

Not here. Both modes are normalized to the same RTP. Higher volatility lowers the hit rate and raises the standard deviation, so results swing further from the RTP line.

Why does my session RTP differ from the theoretical one?

A few hundred spins are a small sample. Session RTP converges only slowly, and more slowly on high volatility, which is exactly what the chart shows.

Are the slot results random?

Live spins use the browser's cryptographic random generator. The figures on this page are computed exactly from the reel weights.

Can I win money?

No. Virtual coins only, no cash value, no prizes, 18+.

More experiments

Free-to-play entertainment and math education with virtual coins only. No real money, no deposits, no withdrawals, no prizes. 18+. Simulated results do not predict real outcomes.

18+ | Virtual Coins Only. No real money, no deposits, no cash-out, no prizes. Simulated results do not predict real outcomes.