Scratch-card odds describe probabilities, not a timetable for wins. A statement such as 1 in 4 does not promise a prize on every fourth attempt. Understanding that distinction is more useful than looking for patterns in past results.
This educational guide explains odds notation, prize tiers and return to player. It does not recommend games, operators or methods of choosing tickets.
What does 1 in 4 mean?
If the stated chance of receiving any prize is 1 in 4, its percentage equivalent is 1 divided by 4, or 25%. A hypothetical 1 in 3.5 probability is approximately 28.6%. These examples explain the notation; they are not verified odds for a particular product.
An average is not a guarantee for a small group of attempts. Under a model with independent trials and a constant 25% prize probability, the chance of four consecutive non-prize outcomes is 0.75 to the power of four, approximately 31.6%. Even in that simplified example, four attempts without a prize would not be unusual.
The assumptions matter. A calculation based on independent trials is not automatically suitable for a finite pool of tickets or every instant-win product.
Overall odds and top-prize odds are different
Overall prize odds combine every outcome counted as a prize. Top-prize odds refer to one specified prize tier. A headline probability for any prize therefore says very little about the likelihood of the largest advertised amount.
Prize tables also describe payment amounts. Two hypothetical games can have the same chance of a prize but very different distributions of small and large payments. The chance of receiving something is not the same as the chance of recovering the amount spent.
A £2 payment after a £5 purchase is a £3 net loss, even if the interface celebrates it. A return equal to the purchase price breaks even on that purchase; it does not recover earlier losses. These distinctions matter when interpreting words such as win, payout and profit.
Return to player and expected loss
RTP expresses a theoretical aggregate return relative to aggregate stakes. Consider a hypothetical product with 90% RTP: over its mathematical model, £100 of stakes corresponds to £90 of expected returns and £10 of expected loss. This is an illustration, not a claim about typical scratch-card RTP.
Expected loss is an average across possible outcomes. It is not a maximum loss. Someone can lose their full spending amount, and actual results over a short period can be far from the average.
RTP is also different from prize frequency. A 90% RTP does not mean nine out of ten purchases receive prizes. Probability and prize size jointly determine expected return. A high advertised maximum prize cannot establish either of those quantities on its own.
Online games and printed tickets
A scratch animation describes the presentation, not necessarily the mechanism used to allocate an outcome. Printed tickets can belong to a finite production run. Digital instant-win products can use different mechanisms, including random-number-based systems or finite prize pools. The rules of the particular product determine which model applies.
It is therefore unsafe to assume that every online scratch game has an unlimited prize supply, that every outcome is independent, or that the result is always fixed at the same point in the animation. Those claims need product-specific evidence. The gesture used to reveal a symbol is not evidence of control over the underlying result.
What rules and disclosures can establish
The Gambling Commission’s RTS 3 standard covers accessible explanations of rules, prize allocation and relevant chances of winning for products within its scope. Its RTS 7 standard addresses acceptable randomness. These are Great Britain regulatory standards, not a universal promise about all sites worldwide.
Useful distinctions include whether a quoted figure concerns any prize or a particular tier; whether payments include the original stake; and whether a feature carries an additional charge. Missing information should not be replaced with assumptions based on the theme or purchase price.
A run of losses does not establish that a prize is due. More purchases create more spending and more exposure to loss; they do not create a guaranteed route to profit.
Self-exclusion and support
Self-exclusion is intended to help people stop gambling. A game or software brand is not an alternative to those protections. GAMSTOP offers online self-exclusion, and GamCare provides information and support. If gambling is causing harm, stopping and seeking support is more useful than looking for different games.
Reviewed 21 October 2026. Educational content based on a Surfer draft. Numerical examples are illustrative calculations, not current product offers.