Football Analytics Glossary: xG, PPDA, Elo and More

By Tactiq AI · 2026-09-08 · 8 min read · AI & Football

Football analytics has its own vocabulary, and most of it is simpler than it sounds. This glossary defines the key metrics in plain English, grouped by what they actually measure, with a link to a full guide for each. If you only learn one, start with expected goals.

Shooting and chance quality

xG (Expected Goals). The probability that a shot becomes a goal, based on distance, angle, body part and chance type. Add it up across a match and you get how many goals the chances were worth, regardless of whether they went in. Full guide: What Is xG.

xA (Expected Assists). The chance quality a player creates for others. It credits the pass that sets up a shot with the xG of that shot, so it rewards creators even when teammates miss. Full guide: What Is xA.

npxG (Non-Penalty Expected Goals). xG with penalties removed. Penalties are worth about 0.79 xG each and are not a repeatable skill in open play, so npxG is a cleaner read on how a team or striker creates from the run of play. Full guide: What Is npxG.

SCA and GCA (Shot- and Goal-Creating Actions). The two offensive actions that lead directly to a shot (SCA) or a goal (GCA). They capture build-up contribution that xG and xA miss, like the pass before the assist. Full guide: SCA and GCA Explained.

Pressing, possession and territory

PPDA (Passes Allowed Per Defensive Action). A pressing gauge: how many passes the opponent completes before the defending team makes a tackle, interception or foul. Lower means a more aggressive press. Full guide: PPDA Explained.

Field Tilt. The share of final-third possession a team has, a measure of territorial dominance that plain possession percentage hides. A side can have less of the ball but more field tilt. Full guide: Field Tilt Explained.

Progressive Passes and Carries. Passes and dribbles that move the ball meaningfully toward goal. They show who actually advances play rather than who passes sideways. Full guide: Progressive Passes and Carries.

PAdj (Possession-Adjusted Stats). Defensive numbers like tackles and interceptions adjusted for how much the ball a team sees. A side that always has possession will record few tackles without being weak defensively; PAdj corrects for that. Full guide: Possession-Adjusted Stats.

Team strength, modeling and luck

Elo Ratings. A single number for team strength that rises and falls after every match, weighted by result and opponent quality. It is the backbone of many match-probability models. Full guide: Elo Ratings in Football.

Poisson Distribution. The statistical model most often used to turn expected goals into a scoreline probability, and from there into home, draw and away percentages. Full guide: Poisson and Goal Modelling.

xPts (Expected Points). The points a team deserved from its chances, built by running each match through its xG. The gap between xPts and real points is the clearest measure of luck over a season. Full guide: xPts and Deserved Tables.

Brier Score. A grade for how well a model's probabilities match reality. Lower is better, and it is the honest way to check whether a forecast that says 60 percent is right about 60 percent of the time. Full guide: Brier Score Explained.

Putting it together

These metrics are most useful in combination. Chance quality (xG, npxG) tells you whether a result was earned, pressing and territory (PPDA, field tilt) tell you how it was shaped, and modeling (Elo, Poisson) turns it all into a probability you can read before kick-off. Calibration (Brier) keeps the model honest.

If you want to see the reasoning in action, our step-by-step guide to analyzing a match walks through how these fit together, and our guide to choosing a free analytics app covers what to look for in a tool that surfaces them.

Frequently Asked Questions

What is football analytics?
Football analytics is the use of statistical models to measure things the scoreline hides: chance quality, pressing intensity, territorial control, and team strength. Instead of asking who won, it asks how, and how repeatable the result was.
What does xG mean in football?
xG (expected goals) is the probability that a given shot becomes a goal, based on factors like distance, angle and chance type. Summed across a match, it estimates how many goals a team should have scored from the chances it created.
What is the difference between xG and xA?
xG measures the quality of shots a player takes. xA (expected assists) measures the quality of chances a player creates for others. A high-xA, low-xG player is a creator; the reverse is a finisher.
What is PPDA in football?
PPDA (passes allowed per defensive action) measures pressing intensity. A low number means a team challenges the opponent after very few passes, so lower PPDA equals a more aggressive press.
Which football stats actually predict results?
Chance-quality metrics like xG and npxG are the most predictive over time because they strip out finishing luck. Possession and shot counts predict less well on their own. Calibration, measured with a Brier score, is how you check whether a model's probabilities hold up.
Where can I see these stats for real matches?
Tactiq applies expected goals, win probabilities and form context to fixtures across 1,200-plus competitions, with a free tier of eight analyses per day on iOS and Google Play.

Related Articles