CORTENOver 2.5 means at least three goals in total in normal time; 0, 1 or 2 goals is Under 2.5. The implied probability of a price is 1 / odds; the Over and Under probabilities add up to more than 100%, and the excess is the margin. A historical “over percentage” only means something together with its sample size.
Over 2.5 means the two teams score three or more goals in total; Under 2.5 means 0, 1 or 2 goals. Because of the half-goal line there is no push (refund): every match finishes either over or under. The same logic applies to lines such as 1.5 and 3.5.
Examples with scores
2-1 (3 goals) is over, 1-1 (2 goals) is under and 0-3 (3 goals) is over. It does not matter which team scores; only the total counts. Extra time and penalties are normally excluded; regular time is what counts.
From odds to probability
A synthetic example: Over 2.5 at 1.85, Under 2.5 at 1.95. The raw probabilities are 1/1.85 ≈ 54.05% and 1/1.95 ≈ 51.28%, about 105.34% in total. The excess (5.34%) is the margin. Divide each by the total and you get normalised probabilities of Over 51.32% and Under 48.68%.
In other words, once the margin is removed, a price of 1.85 expresses something close to a coin flip. You can do the same calculation for three-outcome markets in the probability and margin calculator.
From expected goals to the over probability
In statistics, the total number of goals is often approximated with a Poisson distribution. If the expected total goals per match is λ, the probability of three or more goals is 1 − e−λ(1 + λ + λ²/2). For λ = 2.6 that comes to about 48.16%. This is a simple model: it ignores each team’s scoring tendencies, changes in play with the score and draw effects. To work with separate expected goals for each team, use the Poisson score calculator.
Reading “60% of matches in this league go over”
A historical over percentage only means something together with the number of matches behind it. If 24 of 40 matches went over, the rate is 60%, but the 95% Wilson interval is roughly 44.60%–73.65%: the same data is also consistent with a balanced league. With 240 overs in 400 matches the interval would be much narrower. Try your own numbers in the sample interval calculator.
You also need to compare the historical percentage with the probability implied by those matches’ prices: seeing a high over rate in matches where the over was already short-priced is not surprising. The interesting part is the gap between what the price expected and what happened.
What to watch for
The narrower the odds range and the shorter the period you pick, the smaller the sample and the more likely you are to see high percentages by chance. Trying one line with many different filters and keeping the best also makes the result look better than it is. The past is no guarantee of the future; see methodology and data limits.
Cortenscore members can review Over/Under 2.5 price pairs in past matches next to the sample size and the probability implied by the price.
Create account Membership and pricing (Turkish) ↗
Published 28 October 2026 · Last updated: 28 October 2026 · Responsible: Cortenscore editorial team · Spotted an error? Report a correction. This page is for statistics education; it is not a prediction or betting advice. Example odds are synthetic.