Glossary

1X2

The three possible results of a match: 1 = home win, X = draw, 2 = away win. Settled on 90 minutes plus stoppage time.

Over/under (totals)

Whether the total number of goals is above or below a line, usually 2.5. Over 2.5 means three or more goals.

Both teams to score (BTTS)

Whether each team scores at least one goal.

Odds (decimal)

The bookmaker's price. Decimal odds of 2.50 return 2.50 per 1 staked if the bet wins. 1 ÷ odds gives the implied probability (here 40%).

Margin

The bookmaker's built-in edge: implied probabilities of all outcomes add up to more than 100%. We remove it by scaling them back to 100%; the result is the "market" probability on our pages.

CLV (closing line value)

Compares the price when we published with the final price at kick-off. If we rated an outcome above the market at odds of 2.20 and it closed at 2.00, CLV is +10%: the market moved towards our view. Over many matches, positive CLV is the best evidence of real insight, better than short-term results.

xG (expected goals)

The number of goals a team would be expected to score given its chances. Our goals model produces its own expected goals for each team from season scoring records.

Calibration

Whether probabilities mean what they say: of all outcomes we gave 30%, about 30% should happen. Shown as a chart on the track record page.

Brier score

The average squared difference between predicted probabilities and what happened (1 if it happened, 0 if not). Lower is better. For 1X2 we sum the three outcomes, so it ranges from 0 (perfect) to 2 (completely wrong); for yes/no markets it ranges from 0 to 1. A forecaster who always says 33/33/33 scores about 0.667 on 1X2.

Poisson model

A standard way to model football scores: each team's goals follow a Poisson distribution around its expected goals. Combining both teams gives the probability of every scoreline, and from that 1X2, totals and BTTS.