Football Analytics Glossary
Plain-English definitions of every football analytics term used in the Goalii app and on this site. If you've ever wondered what xG actually is or how a draw fits into 1X2, this page is the reference.
Expected Goals (xG)
Also known as: xG
Expected goals (xG) is a metric that estimates the probability of a shot becoming a goal, based on factors such as shot location, body part used, type of assist, and goalkeeper position. A match's total xG is the sum of every shot's expected goal value. Goalii uses xG as a primary driver for Over/Under and BTTS predictions because it captures attacking quality more reliably than raw shot counts.
Expected Goals Against (xGA)
Also known as: xGA
Expected goals against (xGA) is the xG total of shots that a team allows its opponents to take. It is a measure of defensive quality independent of save percentages. Goalii uses xGA in combination with xG to estimate how many goals each team is likely to concede in an upcoming fixture.
Over/Under
Also known as: Alt/Üst, Total Goals
Over/Under is a football betting market that asks whether the total number of goals scored in a match will be more (Over) or fewer (Under) than a stated threshold, typically 1.5 or 2.5. Goalii predicts Over/Under outcomes from expected goals, recent form, and league context. The market is settled at the end of regulation time.
Both Teams to Score (BTTS)
Also known as: BTTS, KG Var / KG Yok
Both Teams to Score (BTTS) is a football betting market with two outcomes: Yes (both teams score at least one goal) and No (one or both teams fail to score). BTTS is independent of total goals — a 1-1 draw and a 3-2 thriller are both BTTS Yes. Goalii predicts BTTS by combining offensive xG with opponent xGA and historical scoring patterns.
1X2
Also known as: Match Result, Maç Sonucu
1X2 is the three-way football match outcome market: 1 (home win), X (draw), 2 (away win). The market settles on the result at the end of regulation time. 1X2 is the most-traded football market and one of the hardest to predict because draws cluster in a narrow scoreline band. Goalii reports 1X2 probabilities separately and uses a Poisson-based scoreline model to estimate each outcome.
Poisson Distribution
Also known as: Poisson model
A Poisson distribution is a probability model for counting events that occur independently at a constant average rate. In football, the Poisson model is used to estimate the probability of each possible scoreline given each team's expected goals. Goalii's 1X2 prediction is derived from the Poisson scoreline distribution after applying live signal adjustments.
Recent Form
Also known as: Form
Recent form is a weighted summary of a team's performance over its most recent fixtures, typically the last five to ten. Goalii calculates form using xG and xGA rather than raw points so that lucky wins and unlucky losses are properly contextualised. Recent form is weighted by opponent strength so beating a stronger side counts more than beating a weaker one.
Confidence Grade
Also known as: A/B/C grade
Goalii's confidence grade is a three-tier label (A, B, or C) attached to every prediction that summarises how much signal the model has on the fixture. A is the highest confidence, reserved for fixtures where the model has strong, consistent signals; B is moderate; C is preliminary. Goalii publishes the hit-rate separately for each grade so users can audit how often each tier lands.
Hit-Rate
Also known as: Hit rate, Accuracy, Strike rate
Hit-rate is the percentage of predictions that turn out correct over a defined window. Goalii publishes its rolling 90-day hit-rate per market (Over/Under, BTTS, 1X2) and per confidence grade. Hit-rate is not the same as profitability — a 60% hit-rate with poor odds selection can still lose money — but it is the cleanest measure of predictive accuracy.
Backtest
Also known as: Backtesting
A backtest is a simulation of how a prediction model would have performed on historical matches, using only data that was available before each fixture. Backtests are used to estimate a model's real-world performance before it is deployed live. Goalii backtests every market change before publication and exposes the results in-app under Methodology.
Head-to-Head (H2H)
Also known as: H2H
Head-to-head (H2H) refers to the historical record between two specific teams. H2H stats include total matches, goals scored, BTTS rate, and Over/Under tendencies. Goalii uses H2H as a supporting signal but weights recent form more heavily because team composition changes over time. A 5-year H2H from before both clubs changed managers is less informative than recent form.
Related pages
Last updated