What matters
Boxing bouts are scheduled from 4 to 12 rounds, and when a bout goes the distance the result can hinge on how three judges see a handful of close rounds. Reading scorecards and officials is the foundation of any boxing analysis.
Important data
| Input | Why it matters |
|---|---|
| Official records | Every card on official record |
| Judges’ scorecards | Round-by-round scoring by each judge |
| Officials | Judges and referees with their histories |
| Titles and rankings | Sanctioning-body context |
| Scheduled rounds | Sets how long a finish has to happen |
Common metrics
| Metric | What it measures | How to read it |
|---|---|---|
| 10-point must | Round scoring system | 10–9 close, 10–8 dominant |
| Split / majority decisions | Judge disagreement | Signals close fights and scoring variance |
| Scheduled rounds | 4–12 | Distance probability context |
Model concepts
The Boxing methodology states that the PropBetEdge bout model is untrained, so no probability, fair line or pick is published. That is a deliberate no-call at the product level — the right choice until a model can pass validation.
Everything in the Model Literacy path still applies when a model arrives: out-of-sample Brier and log loss against baselines, calibration, and publication gates.
Live intelligence
A boxing bout's state is the round, the scheduled distance and the scorecards so far. Unlike most sports, the running scorecards are usually not public during the fight, which adds uncertainty for anyone reading it live.
PropBetEdge Boxing tools
Worked example
- Scorecards: 115–113, 113–115, 116–112.
- Two judges favour Fighter A, one favours Fighter B: a split decision for A.
- 115–113 means A won 7 rounds to 5 on that card; one round flipping changes it to 114–114.