Free · browser-local · one total xG input
Over/Under goals probability calculator
Convert one total expected-goals mean into Over and Under probabilities for every half-goal line from 0.5 to 5.5.
No sign-upNo data uploadSix goal linesNot betting advice
One required input
Enter total expected goals
This tool needs λTotal only. It does not ask you to invent a home-away split when the question is simply total goals.
Free, local and upload-freeThe calculation runs in your browser. No match, account or input is sent to a prediction service.
Six totals lines
Over and Under from 0.5 to 5.5 goals
Each row is a complementary pair: Under x.5 counts zero through x goals; Over x.5 counts x + 1 or more. The two probabilities sum to 100% before display rounding.
| Goal line | Over probability | Over fair odds | Under probability | Under fair odds |
|---|---|---|---|---|
| O/U 0.5 | 92.9% | 1.08 | 7.1% | 14.15 |
| O/U 1.5 | 74.2% | 1.35 | 25.8% | 3.88 |
| O/U 2.5 | 49.4% | 2.02 | 50.6% | 1.98 |
| O/U 3.5 | 27.5% | 3.64 | 72.5% | 1.38 |
| O/U 4.5 | 13.0% | 7.71 | 87.0% | 1.15 |
| O/U 5.5 | 5.3% | 18.94 | 94.7% | 1.06 |
Exact formula for every lineP(Under k + 0.5) = Σn=0…k e−λTotalλTotaln / n!
P(Over k + 0.5) = 1 − P(Under k + 0.5). Fair decimal odds are 1 ÷ probability with no bookmaker margin or adjustment.
Method
How are Over/Under probabilities calculated from total xG?
Treat the full-match goal total N as a Poisson count with mean λTotal. For Under 2.5, add the exact probabilities of N = 0, N = 1 and N = 2. Over 2.5 is the complement: one minus that cumulative probability. The same rule shifts the cutoff for every other half-goal line.
Because half-goal lines cannot tie, each Over and Under pair is exhaustive and mutually exclusive. Its unrounded probabilities therefore add to exactly 100% under the model.
Different input contract
Why this calculator asks for total xG only
A totals question needs the distribution of the match total, not a fabricated home-away split. When independent home and away Poisson means are available, their sum is the total mean. But λTotal may also come from a separately documented totals model; this calculator does not infer how it was produced.
If you need home/away results, correct scores or BTTS, use a two-rate calculator instead. A single total mean cannot recover which team is expected to score.
Scope and limits
What the totals table does—and does not—mean
- The full-match total is a Poisson count with one fixed mean.
- The total xG input is user supplied and is not fetched, estimated or validated here.
- No team identity, game state, red card, extra time or parameter uncertainty is modelled.
- The probabilities are a synthetic consequence of the stated input—not evidence of prediction accuracy.
Fair odds remove margin, not uncertainty.
Reciprocal odds are included to make probability conversion auditable. They are not bookmaker quotes, value signals, profit claims, staking guidance or betting recommendations.
Research trail
A legible baseline still needs out-of-sample testing
- Maher (1982), Modelling association football scores
- Karlis & Ntzoufras (2003), Analysis of sports data by using bivariate Poisson models
These sources explain football Poisson modelling and richer dependence structures. They do not validate a particular total xG input, forecast source or future Football Proof AI model.
Canonical publication record
Abstract
A browser-local cumulative-Poisson calculator that converts one user-supplied total expected-goals mean into complementary Over and Under probabilities and zero-margin fair odds for every half-goal line from 0.5 to 5.5.
- Author and publisher
- Football Proof AI
- Technical report
football-total-goals-poisson/1.0.0- Published
- Last modified
- Release status
- Current release
- Review status
- Editorial technical note; not externally peer reviewed
- Version history
football-total-goals-poisson/1.0.0: Initial public release.