Football statistics glossary

The terms you'll find on Scout Specialist and on statistics websites, explained in two lines: from 1X2 to xG, from BTTS to regression to the mean.

1X2
The final result of the match: 1 home win, X draw, 2 away win. Scout Specialist estimates the probability of each of the three outcomes.
Brier score
Measures the accuracy of probabilistic forecasts: the lower it is, the better the model estimated the probabilities. It's one of the numbers we'll publish on the Method page.
BTTS (Both Teams To Score)
Shows whether both teams score. BTTS yes: both score at least one goal. BTTS no: at least one fails to score. BTTS guide.
Calibration
How closely a model's probabilities match reality: if a model is well calibrated, events estimated at 60% happen about 6 times out of 10.
Sample
The number of matches a statistic is calculated on. With a small sample (few matchdays) percentages change a lot from one match to the next.
Cards (count)
In Scout Specialist's statistics a yellow counts 1 and a red counts 2, to give more weight to sendings-off. Cards guide.
Clean sheet
A match in which the team doesn't concede. The clean-sheet percentage measures defensive solidity.
Expected corners
The estimate of a match's total corners, obtained by combining the corners won and conceded by both teams at home and away.
Expected goals (xG)
The probability that a shot becomes a goal, based on position and type of shot. Added up, they tell you how many goals a team "deserved" to score. xG guide.
Home advantage
The advantage of playing at home. That's why statistics should be read with home and away kept separate. Read more.
Form
The results of the last few matches (usually 5), shown as W (win), D (draw), L (loss).
Match expected goals
The number of goals the model expects from each team in a specific match. It's the basis for calculating all the probabilities.
Goals − xG
The difference between goals scored and expected goals. Strongly positive means a team is scoring more than it creates and is often due to slow down.
Poisson model
The statistical model that, starting from expected goals, calculates the probability of a team scoring 0, 1, 2, 3… goals. Step-by-step example.
Over/Under
Shows whether the match's total goals go over (Over) or stay under (Under) a line: 0.5, 1.5, 2.5, 3.5, 4.5. Over/Under guide.
PPG (points per game)
The average points won per match: 3 for a win, 1 for a draw. It lets you compare teams that have played different numbers of matches. Read more.
PPG for every team: Serie A · Premier League · La Liga · Bundesliga · Ligue 1.
First and second half
Goal statistics split by half: how often a team scores or concedes before and after the break.
News Radar
The Scout Specialist feature that checks injuries, suspensions and doubtful players from the official lists, links them to the right match and recalculates the probabilities. Discover the Radar.
Regression to the mean
The tendency of extreme values to drift back towards the average over time: a very good or very bad run rarely lasts. Read more.
Correct score
The exact final score. Even the most likely one rarely goes above 12–15%, which is why it's better to look at aggregate probabilities.
Head-to-head (H2H)
Past meetings between two teams. Read them with caution: squads and managers change, and there are often only a few matches.
Change threshold
In the news Radar, the minimum number of percentage points a probability must change by for you to get a notification.
xGA (expected goals against)
The expected goals conceded to opponents: they measure the quality of the chances a team allows.
See these numbers for today's matchesEvery term in the glossary has its own column in the dashboard. Ask Scout AI: try it free, no account needed.
Try it free