xT in football stands for Expected Threat and measures how much an action changes the attacking value of a possession as the ball moves across the pitch. Rather than focusing only on shots or assists, the model assigns different threat values to areas of the pitch and evaluates how passes and carries move the ball between them. This makes it possible to identify which actions increase attacking danger before a shot is taken, giving analysts a clearer view of how players contribute to progression, chance creation and overall attacking play.
How xT in Football Works and What It Measures
xT in football works by assigning a threat value to different areas of the pitch according to how likely possession in each area is to eventually lead to a goal. When a player completes a pass or carries the ball from one zone to another, the model measures the change between the starting value and the value of the destination. Actions that move the ball into more dangerous areas therefore generate positive xT, while actions that reduce the immediate threat can produce a lower or negative value, depending on the implementation of the model.
This approach allows analysts to evaluate actions that take place before the final pass or shot. A midfielder who breaks a line with a progressive pass or a winger who carries the ball towards the penalty area may increase attacking threat even if neither action directly produces an assist. In this sense, xT helps quantify how possession develops rather than simply recording its final outcome.
That is also the main difference between xT and xG. Expected Goals evaluates the quality of a scoring opportunity once a shot occurs, whereas Expected Threat focuses primarily on how passes and carries move possession into areas with greater attacking potential.
| Metric | What it measures | When it adds value |
|---|---|---|
| xT | Change in attacking threat as the ball moves between areas of the pitch | During possession progression through passes and carries |
| xG | The probability that a shot will result in a goal | When a shot is taken |
xT in football measures how passes and carries change attacking threat as possession moves across the pitch
How xT in Football Is Calculated from Pitch Zones
To calculate xT in football, the pitch is divided into a grid of zones and each zone is assigned a value representing the attacking threat associated with having possession there. In Karun Singh’s original Expected Threat model, the pitch was divided into a 16 × 12 grid, creating 192 zones, although the resolution can be changed depending on the amount of event data available.
Historical match data is then used to estimate what usually happens when a team has the ball in each zone. The model considers the probability of shooting, the probability of scoring if a shot is taken, the probability of moving the ball and the zones to which successful moves are made. Passes and carries therefore connect different areas of the pitch through a transition matrix based on previous actions.
These probabilities are combined iteratively. A zone close to goal may initially have a high value because shots from that location are more likely to score, but xT also propagates that threat backwards through the possible passes and carries that can eventually lead to dangerous positions. This is what allows the model to value stages of an attack that occur several actions before the final shot.
Once the xT surface has been calculated, valuing a successful ball-moving action becomes relatively straightforward. The xT generated by the action is the difference between the threat value of its destination and its starting position.
For example, imagine a pass starts in a zone valued at 0.021 xT and reaches another valued at 0.074 xT:
The pass therefore generates +0.053 xT because it moves possession into a considerably more threatening area. This simple calculation is what turns the underlying pitch model into a practical measure of attacking progression.
How xT in Football Values Passes and Carries
One of the main strengths of xT in football is that it can value actions that shape an attack before a shot takes place. Instead of rewarding only the final pass or the finish, the model captures how a player changes the attacking potential of a possession by moving the ball into a more or less dangerous area. This makes xT in football especially useful for analysing progressive passes and carries that often go unnoticed in more traditional statistics.
A simple example is a lateral pass in a team’s own half. The action may help maintain possession and circulation, but if it barely changes the level of danger, the increase in xT will usually be very small. By contrast, a line-breaking pass that takes the ball from a safer zone into a central area between the opposition’s midfield and defensive lines can produce a much larger xT gain, because it moves possession into a more threatening position.
The same logic applies to carries. A player who drives forward with the ball into the final third or towards the edge of the box may generate significant xT even without delivering an assist. In that situation, the value comes from transporting possession into a zone from which the team is more likely to create a dangerous attack. On the other hand, a backward pass can reduce immediate threat if it moves the ball away from a promising attacking area.
The key point is that xT in football does not judge whether an action looks spectacular, but whether it changes the attacking value of possession. That is why two passes completed successfully can have very different meanings in analysis. The examples below are purely illustrative, designed to show how the model interprets different actions rather than to represent values from a specific data provider.
| Action | Starting zone | Final zone | xT change | Interpretation |
|---|---|---|---|---|
| Lateral pass | 0.021 | 0.023 | +0.002 | Small increase in attacking threat |
| Line-breaking pass | 0.021 | 0.074 | +0.053 | Strong attacking progression into a more dangerous area |
| Carry towards the box | 0.035 | 0.091 | +0.056 | Significantly increases attacking danger |
| Backward pass | 0.074 | 0.040 | −0.034 | Reduces the immediate threat of the possession |
How to Interpret xT in Football for Players and Teams
Once the model has assigned values to individual actions, xT in football can be aggregated to analyse how players and teams contribute to attacking progression. However, the overall figure needs context to be meaningful, so analysts usually break it down by type of action, playing time and the areas of the pitch where threat is generated:
- xT from passes measures how much threat a player creates by successfully moving the ball between zones through passing. A high value can highlight players who regularly find team-mates in more dangerous positions, particularly through progressive or line-breaking passes.
- xT from carries isolates the value generated when a player progresses with the ball. This can reveal players who increase attacking threat through movement rather than distribution, such as wingers driving towards the penalty area or midfielders carrying possession through pressure.
- Total xT combines the threat generated across relevant actions. It provides an overall measure of attacking progression, but totals are influenced by playing time, possession and the number of opportunities a player receives.
- xT per 90 standardises production according to minutes played, making comparisons between players with different amounts of game time more meaningful. It should still be considered alongside role and team context rather than used as a standalone ranking.
Analysts can also examine where xT is generated and how much action volume is required to produce it. A player creating threat repeatedly from deep areas is performing a different function from an attacker accumulating a similar xT closer to goal. Likewise, a team may generate its attacking threat mainly through central progression, wide combinations or carries into advanced areas.
For that reason, xT in football should always be interpreted within tactical context. A full-back, central midfielder and winger occupy different spaces and receive different types of possession. Two players can therefore record similar xT figures while contributing to attacks in completely different ways.
How Analysts Use xT in Football for Performance Analysis and Scouting
For analysts, xT in football becomes most useful when it is connected to tactical behaviour rather than treated as an isolated number. Once the source of attacking threat is identified, the metric can help explain how a player or team progresses possession, where dangerous actions originate and which patterns repeatedly move the ball into higher-value areas.
Its practical use can be divided into three main areas:
- Player analysis: xT can identify players who consistently increase the attacking value of possession through passes or carries. Analysts can then review where those actions start, which zones they reach and whether the player creates threat through progression, combination play or movement into advanced areas.
- Team analysis: aggregated xT can reveal how a team builds attacking threat across different phases of possession. Analysts can compare central and wide progression, identify recurring routes into dangerous zones and examine whether certain structures or combinations repeatedly produce high-value actions.
- Scouting: xT provides another layer for comparing players with similar roles by showing how they contribute to moving attacks forward. However, comparisons become more meaningful when analysts consider position, minutes played, team possession, tactical role and the zones in which each player operates.
The metric also has important limitations. xT values depend on how the model is constructed, including the size of the pitch zones, the historical event data used and the methodology applied to estimate threat. Different providers or models may therefore produce different values, while contextual information such as defensive pressure, player positioning or the quality of available passing options may not always be fully represented.
For this reason, professional analysis combines metrics such as xT with video, tactical context and other performance indicators. Learning to connect data with specific moments of the game is also central to areas such as set-piece analysis, where positioning, execution and context must be studied together. This analytical approach is developed further in the MSc Degree in Set Pieces in Football, which focuses on applying data and football analysis to specialised game situations.
