A Passing Metrics Glossary: The Terms Used to Measure How Teams and Players Move the Ball
Passing stats in football are the counts, rates and models used to describe how a player or team moves the ball: how often, how accurately, how far forward and with what effect on chances. Match statistics, lineups and player pages on RubiScore supply the context of role, minutes and opponent that these numbers need.
This glossary treats the passing family as one system. It groups more than 20 terms by the question each answers: how much and how cleanly, how far forward, what the pass created, what kind of pass it was and what it says about the team. Fixtures, results and live match statistics are available at https://rubiscore.com for placing any passing figure in the match it came from.
Where the same label carries different definitions, the glossary quotes two major event-data providers, Opta and Wyscout. Concepts with dedicated explainers, such as expected pass completion, deep completions and pass networks, get a single line.
Volume and Accuracy: How Much and How Cleanly?
Pass: Opta defines a pass as the attempted delivery of the ball from one player to a teammate. Its count includes goal kicks and corners or free kicks played as passes, while crosses, goalkeeper throws and throw-ins are logged as separate events.
Completed pass: In Opta's wording, a pass that goes to a teammate directly without a touch from an opponent. Wyscout's test is slightly different: a pass is successful if the next touch is by a teammate.
Pass completion percentage: Completed passes divided by attempted passes, often called pass accuracy. Opta notes that it usually excludes crosses, which are measured separately.
Passes per 90: Attempted or completed passes scaled to 90 minutes, so that substitutes and starters can be compared. Very small samples still produce unstable rates.
Pass completion over expected: The actual completion rate minus the rate a model predicts from the difficulty of each pass.
Direction and Progression: Which Passes Move the Ball Forward?
Forward-pass share: The proportion of a player's passes that travel forwards rather than sideways or backwards, classified from each pass's start and end coordinates.
Progressive pass: A pass that moves the ball significantly closer to the opponent's goal, and the clearest case of one label carrying several definitions. Opta counts a completed pass in the attacking two-thirds of the pitch that moves the ball at least 25 per cent closer to goal. Wyscout measures from the start of the pass to the next touch: at least 30 metres closer when both points are in a team's own half, 15 metres when the pass crosses halfway and 10 metres when both are in the opponent's half.
Progressive passing distance: The total distance a player's completed passes move the ball towards the opponent's goal line, which rewards range as well as frequency.
Pass into the final third: Wyscout defines it as any pass that starts outside the final third and whose next touch occurs inside it, with throw-ins excluded.
Pass into the penalty area: A pass that starts outside the opponent's penalty area and is next touched inside it. In Wyscout's version, crosses count, but entries made by dribbling with the ball do not.
Line-breaking pass: FIFA's Football Language defines a line break as a player progressing the ball and breaking one or more unit lines of the opposition team shape, by playing through, around or over them. Because it depends on where defenders stand, it usually needs positional data rather than ball events alone.
Switch of play: A long pass from one flank to the other, usually to attack the less crowded side. Providers set their own minimum lateral distance.
Deep completion: A non-cross pass into the zone closest to the opponent's goal, which Wyscout sets at within 20 metres.
Chance Creation: What Did the Pass Lead To?
Assist: Opta defines it as the final touch from a teammate that leads to the recipient scoring, and awards none for own goals or goals scored directly from free kicks or corners. Wyscout's assist is the last action by a player of the scoring team before a goal, which can include an own goal. Opta also records a broader fantasy assist, covering cases such as a penalty won.
Key pass: In Opta's data, the final pass to a teammate who then shoots without scoring; key passes plus assists make up chances created. Wyscout's key pass is narrower, a pass that immediately creates a clear goalscoring opportunity, so its shot assist is the closer match. Shot-creating actions extend credit to the two actions before a shot.
Second assist: Wyscout defines it as the last action by a player of the scoring team before a teammate's assist. Opta uses the term more loosely for a pass or cross instrumental in creating a chance.
Expected assists (xA): A model-based value for chance creation, defined in two ways. Opta's xA measures the likelihood that a completed pass becomes an assist, using factors such as pass type, end point and length, so a pass can earn xA even if no shot follows. Wyscout gives a shot assist the xG value of the shot it set up.
Through ball: Opta describes it as a pass splitting the defence for a teammate to run on to. Wyscout's through pass is played into the space behind the defensive line.
Cut-back: A pass played backwards from near the byline to a teammate arriving in or around the penalty area. Opta logs it as a pull-back, made by a player in the opposition box who reaches the byline, with crosses excluded.
Cross: Opta defines it as a ball played from a wide position towards teammates in a central area close to goal. Wyscout counts open-play crosses only and treats the outer 23 metres of a normalised 68-metre pitch width as each flank. Cross completion is the share of crosses that find a teammate.
Type and Context: What Kind of Pass Was It?
Short, medium and long pass: Length bands differ by provider and by unit. Wyscout tags a long pass as a ground pass longer than 45 metres or a high pass longer than 25 metres, and treats every other pass as short or medium.
Long ball: A long, usually aerial pass aimed at territory rather than a precise target. Opta's closest qualifier is the launch, a long high ball into space or an area for players to chase.
Pass under pressure: A pass made while an opponent is closing down the passer. Only datasets that log pressing events can flag it, and the threshold varies.
Open-play and set-piece passes: Goal kicks and set-piece passes can sit inside the same totals as open-play passes. Separating them stops a set-piece taker's deliveries from distorting an open-play profile.
Team Level: How Does a Side Pass as a Unit?
Possession share: The percentage of the ball a team had, calculated by some providers from each team's share of passes and by others from time on the ball. In the first leg of the 2012 Champions League semi-final between Chelsea and Barcelona, UEFA credited Barcelona with 72 per cent of possession, while Opta, whose figure was then based on passes, gave Chelsea 20 per cent.
Passes per sequence: The average number of passes in a team's sequences, which Opta defines as passages of play belonging to one team that end with a defensive action, a stoppage or a shot. High values point to patient build-up.
Directness: How quickly a team moves the ball towards goal. Opta's direct speed divides the metres a sequence progresses towards the opponent's goal line by its duration in seconds.
Pass network and pass sonar: Visual summaries rather than counts, mapping passing links between teammates and the direction mix of a player's passes.
How Do the Passing Terms Relate?
The families form a chain. Volume says how often a player passes, accuracy how cleanly, progression how far forward and creation what the pass led to. A sound reading usually needs a term from each link.
- Volume is not value. A centre-back can attempt more passes than anyone on the pitch and still move the ball forward less than a teammate who passes half as often.
- Completion percentage alone misleads. Wyscout's glossary puts the average success rate at about 86 per cent for short and medium passes and about 50 per cent for long passes, so a player's mix of pass types largely sets the rate.
- Role and team context decide what is normal. A full-back in a crossing side and a deep playmaker in a possession side produce different profiles by design. Lineups and player pages on RubiScore show the position and minutes behind a passing figure, the first check before two profiles are compared.
- Adjust for minutes and for the ball. Per-90 rates handle playing time, but a team that dominates possession gives its players more chances to pass. FIFA's technical analysts normalise data per 30 minutes in and out of possession, noting that the ball is generally in play for about 60 of the 90 minutes.
Why Do Providers' Numbers Differ?
Two sources can publish different passing figures for the same match without either being wrong. The usual reasons are:
- What counts as a pass. Including or excluding crosses, throw-ins and set pieces changes the total.
- What counts as a success. Reaching a teammate directly and being next touched by a teammate are not identical tests.
- Thresholds. Progressive passes, long passes and switches depend on distances set differently, in metres or yards.
- Attempts or completions. Opta's progressive pass counts completions only; Wyscout counts attempts and reports an accuracy rate.
- Models. An xA built on pass features and one built on the following shot's xG value the same pass differently.
The practical rule is to check the definition before the number. A passing statistic is most informative when its provider, threshold and denominator are known, and when it is read alongside the player's role, minutes and the state of the match.
