Return to player, usually shortened to RTP, is a long-run measure. It does not promise what an individual will receive during a session. A percentage on a game information screen is therefore not a personal repayment guarantee.
A long-run average is not a session forecast
The Gambling Commission explains that RTP is measured over substantial numbers of plays. For a random game, previous wins or losses do not change the probability of the next outcome. A losing sequence does not, by itself, establish that a win is due. This statement concerns random games; the Commission separately describes compensated gaming machines.
House edge describes the casino’s expected advantage under the game’s rules and relevant assumptions. It is not a reliable prediction of one person’s next result. Source: RTP and house edge explained.
Observed results can differ from theoretical RTP
Actual RTP is calculated by dividing prizes paid by total stakes over the measured sample. A theoretical RTP is a design expectation. The Commission explains that volatility and sample size affect how widely observed results can differ from that expectation. A short run below the theoretical figure does not alone prove a malfunction; a short run above it does not establish a repeatable advantage. Source: calculating and assessing RTP.
What fairness does and does not mean
Fairness testing concerns whether games operate as required. It does not mean every participant wins or avoids losses. For operators under its oversight, the Commission describes checks on game performance after launch as well as testing requirements. Source: online game fairness.
Treat claims such as “a win is overdue” or “this percentage guarantees your money back” as claims requiring evidence, not conclusions supported by RTP. Neither an isolated win nor an isolated loss establishes the underlying probability distribution.