CORTENOdds and probability
Decimal odds — The price published for an outcome, e.g. 2.10, shown in decimal format.
Implied probability — The raw probability a price corresponds to: 1 / odds. About 47.62% for 2.10. Convert ↗
Margin (overround) — The part of the total implied probability across all outcomes of a market that goes above 100%. More ↗
Margin-free probability — Probabilities that add up to 100% after the margin is removed, using basic normalisation or the power method. Not a model prediction. More ↗
Opening odds — The first price recorded for a match. When it was recorded depends on the source. More ↗
Closing odds — The last recorded price. Whether it was recorded before or after kick-off needs to be checked separately. More ↗
Odds movement — The price difference between two records; it does not show every change in between. More ↗
Markets
1X2 — Full-time result: home win (1), draw (X), away win (2). More ↗
HT/FT — The half-time and full-time results expressed together; there are nine possible outcomes (e.g. X/1). More ↗
Double Chance — A selection covering two of the three 1X2 results: 1X, 12, X2. More ↗
BTTS Yes / No — Both teams score at least once (Yes), or at least one fails to score (No). More ↗
Over / Under 2.5 — Three or more goals in the match (Over) or at most two (Under). More ↗
Comeback — The team behind at half-time wins the match: 2/1 or 1/2 in HT/FT. More ↗
Statistics
Sample — The number of matches a rate is calculated from. The same percentage does not carry the same uncertainty in 5 and 500 matches. More ↗
Wilson interval — A range showing the sampling uncertainty of an observed rate; for 24 out of 40 the 95% interval is roughly 44.60%–73.65%. More ↗
Expected goals (λ) — The average number of goals a model expects per match or per team. Calculate ↗
Poisson distribution — A distribution used to approximate counts such as goals; it relies on simplifying assumptions. Poisson calculator ↗
Multiple testing — Trying many rules or ranges at once, which raises the chance of finding one that looks good by luck. More ↗
Cherry-picking — Showing only the examples that worked and hiding the ones that didn’t; it makes a result look better than it is. More ↗
Pre-registration — Writing down and fixing a hypothesis and its test method before looking at the results. More ↗
Last updated: 28 October 2026. This page is for statistics education; it is not a prediction or betting advice.