Skip to content
Insights

Maximum Win, Feature Frequency and RTP Contribution Explained

By Smiling Ghost · 4 October 2026 · Slot Mathematics

Most modern slots split their return between the base game and a bonus feature, and cap the biggest possible win. How that split is designed decides how a game feels far more than the headline RTP. This article walks through one illustrative game, “Game D”, and shows what feature frequency, RTP contribution and a maximum win mean in numbers.

About the numbers. Game D is a small model built for this article, not one of our games. Exact values come from its paytable; session results come from 20,000 simulated sessions of 300 spins (Python/NumPy, fixed random seed).

Game D in one table

Each spin either triggers the feature, about once in 199 spins, or pays a base-game result. The base game alone returns 70% of the bet. When the feature triggers, it pays one of these amounts:

Feature pays Share of features Adds to the feature average
10x 40% 4.00x
20x 30% 6.00x
50x 18% 9.00x
150x 9% 13.50x
500x 2.95% 14.75x
5,000x (maximum win) 0.05% 2.50x
Average feature 49.75x

Total RTP is 94.65%: 69.65 points from the base game and 25.00 points from the feature.

RTP contribution: a quarter of the return in one feature

About 26% of everything Game D pays back comes from a feature that a player sees once in 199 spins on average. That has a direct effect on sessions:

Spins played Chance of not seeing the feature at all
200 36.5%
500 8.1%
1,000 0.6%

In our 300-spin simulation, 21.9% of sessions never triggered the feature. Their median return was 70.0%, the same as the base game’s RTP. Sessions that did trigger it had a median return of 86.8%. A player who never sees the feature is, in effect, playing a 70% game.

Maximum win: what “1 in N” means

Game D’s maximum win, 5,000x, happens once in 398,000 spins on average and adds only 1.26 points to the RTP. Probabilities like this are easy to misread. If the chance per spin is 1 in N, the chance of seeing it at least once is:

  • in N/10 spins (39,800): 9.5%;
  • in N spins (398,000): 63.2%, not 100%;
  • in 2N spins (796,000): 86.5%.

The general formula is 1 − (1 − p)k for k spins at probability p per spin.

Platforms set rules for this. Stake Engine’s math requirements ask that the maximum win stated in the rules matches the real one and is realistically obtainable, “typically more frequent than 1 in 10,000,000”. A cap you will almost never see is not a feature, it is a marketing number.

Pricing a bonus buy

If players can buy the feature directly, the price follows from the same numbers. To give the buy the same RTP as the base game, divide the average feature win by the target RTP: 49.75x ÷ 0.9465 ≈ 52.6x the bet. Some platforms require the RTP of every mode to stay close together. On Stake Engine, all modes of a game must be within 0.5% of each other.

Takeaways

  • Ask how much of the RTP sits in the feature. The bigger that share, the more a session depends on whether the feature shows up.
  • Read “1 in N” as an average, not a promise. Even after N spins, there is a 37% chance of not having seen it.
  • A maximum win is meaningful only together with its probability.

Related: Slot RTP Explained · Slot Volatility vs Hit Frequency · How we build and check our maths.

Want to work with us?

Distribution, media or a collaboration idea: tell us what you have in mind.