Five ways the numbers get misread
The maths of games is not complicated, but a few intuitions run against it. These are the misreadings that turn a fair description of a game into a false expectation about tonight.
1. “It hasn’t paid, so it’s due”
This is the gambler's fallacy: the belief that independent rounds owe you a result because of what came before. Rounds with no memory cannot owe anything. A game that has not paid for a long time is not more likely to pay next round; its odds are exactly as published, unchanged by history.
The same error in reverse is expecting a game that just paid to be “used up”. Neither state exists. Independence means the past is not information about the next round.
2. Treating RTP as a session guarantee
RTP is a long-run average over an enormous number of rounds. Applying it to a single session — expecting that a 96% return will “show up” over an evening — is a category error. In a short window, the spread swamps the average, and the session can end far from the published figure in either direction.
Worse, the reading tempts the idea that stakes must be added to “realise” the RTP. There is nothing to realise in a session; the figure is a property of the design, not a balance waiting to be claimed.
3. Confusing hit frequency with profit
A game that returns something on most rounds can still retain the same edge as one that returns rarely. Frequent small returns feel generous and are not the same as profit: the amounts returned can sit below the stakes, and the long-run average can be identical to the spiky game next to it.
Hit frequency describes the rhythm of returns, not their total. Reading it as a measure of how much comes back is a common and costly slip.
4. “Hot” and “cold” games
With independent rounds, a game has no temperature. A run of wins does not make a game hot, and a run of losses does not make it cold. Streaks feel meaningful because human attention is drawn to order, but the underlying odds do not change to produce them.
The practical harm of the temperature story is that it licenses larger stakes on a game believed to be running well — increasing variance and exposure on the basis of a pattern that carries no information.
5. Reading a session as skill
Over a session, results are dominated by variance, and variance is not controllable. A good outcome is not proof of a good method, and a bad one is not proof of a mistake. Crediting skill for a lucky session leads players to press an advantage that was never theirs, and to keep stakes high on the strength of an outcome the maths does not support.
Poker and similar games, where decisions genuinely matter against other players' choices, are a different question. In fixed-odds games, the only decisions a player controls are the stake, the volatility and when to stop.
This page carries an affiliate link to gamdom.com/r/csgo2026. If you open an account through it we may earn a commission. It costs you nothing extra, it does not change what we write, and no operator pays for a position here. 18+ only. Gambling involves risk and can cause serious financial harm — including debt, damaged relationships and mental-health problems. Game maths is descriptive, not predictive: a favourable-looking RTP states how a game is built over millions of rounds, never what will happen to your balance, which can be lost entirely in a single session regardless of the number. Free, confidential support exists in most countries through national gambling-harm helplines. Never stake money you cannot afford to lose, never borrow to gamble, and never stake more to recover a loss.