What the Hell are Expected Goals?

Back to Guide

Expected Goals (xG) is an analytics metric in Soccer that often sparks debate among fans and analysts. While some view it as a useful way to evaluate modern soccer performance, others consider it a flawed statistic. Read on to learn exactly what xG means, how it is calculated, and why it remains a frequent topic of discussion within the soccer community.

What is an expected goal (xG)?

Expected Goals (xG) is used in modern sports analytics to evaluate a fundamental question: how high was the quality of a specific scoring opportunity? Rather than relying on guesswork, this metric assesses the probability of any shot taken on the soccer field. By analyzing an extensive database of historical match information, xG offers a data-driven calculation on a scale from 0 to 1, where a higher number indicates a higher probability of a goal, and a lower number represents a more difficult chance.

For example, a close-range tap-in from 10 yards out might be assigned an xG value of 0.55, indicating a 55% statistical probability of scoring. Conversely, a long-range attempt from 30 yards out might carry an xG of just 0.05, representing a 5% chance. If a player takes both of these shots in a match and scores one goal, their performance would exceed the statistical expectation, resulting in 1 actual goal compared to an expected total of 0.6 xG. To achieve this level of detail, the metric accounts for a variety of situational variables that extend beyond simple distance from the net.

Working out expected value in Soccer

While different analytical providers use unique criteria to calculate Expected Goals (xG), most major models focus on a similar set of core variables. Distance from the net is only a starting point. To determine the statistical probability of a scoring opportunity, an xG model typically evaluates several specific details:

• The Angle: Evaluates whether the shot was taken directly from the center of the field or from a more challenging, narrow angle.
• The Body Part Used: Notes whether the shot was a standard kick or a header.
• The Dominant Foot: Accounts for whether the player used their preferred foot or their weaker one to take the shot.
• The Assist Type: Considers how the opportunity was created, such as through a targeted through ball, a cross into the box, or an individual dribble.
• The Defensive Pressure: Examines whether the shooter was in a clear one-on-one situation or surrounded by opposing players.
• The Goalkeeper and Defender Positioning: Tracks the exact locations of the defenders and the goalie at the moment the shot was taken.

By analyzing these specific details alongside a large database of historical match data, the model helps reduce the impact of luck in the evaluation process. When it looks at those scoring attempts, it compares the specific action against thousands of similar past scenarios to determine a consistent statistical probability.

What’s up with expected goals?

In soccer analytics, Expected Goals (xG) is a helpful metric used to evaluate performance and assess future trends. Many analysts and sportsbooks look beyond the final score, using xG datasets to focus on the underlying quality of chances created by players and teams on the field.

Traditional statistics like possession percentage and total shots can sometimes be misleading. In contrast, xG offers a clearer picture of the scoring opportunities a team actually generated versus the outcomes that may have been influenced by luck. The 2022 World Cup provides a clear example of this dynamic, particularly in Argentina’s unexpected 2-1 loss to Saudi Arabia.

While Argentina’s 70% possession was a notable headline, it did not fully reflect the specific flow of the game. The xG data offered a different perspective, showing 2.24 expected goals for Argentina compared to 0.14 for Saudi Arabia. Saudi Arabia managed to convert low-probability chances while utilizing a high defensive line and persistent pressing. In this context, the xG metric highlights the final score as a notable statistical anomaly rather than a reflection of typical match trends.

Plotting out xG

To make raw data easier to interpret, analysts frequently use xG maps to visualize a team’s or player’s performance on the field. These maps are utilized by both fans and coaching staffs to evaluate trends and consistency over time. Consider Erling Haaland’s 2022–2023 debut season as an example. He finished the year with 36 goals against an expected goals total of 28.7, and an xG map illustrates exactly how his scoring opportunities unfolded in the Premier League.

The map highlights several key details about his performance:

• High Efficiency: An average of 1.05 xG per game.
• Shot Quality: A solid 0.23 xG per individual shot.
• Shot Location: The specific areas on the field where his attempts originated.
• Finishing Success: The conversion rate for each of his shots.
• Overall Volume: A total of 123 shots taken over the course of 35 matches.

These maps can track more than just individual scoring opportunities. They also measure Expected Assists (xA), which evaluates the probability that a player’s passes will lead directly to a goal. By combining xG and xA, analysts calculate Expected Goal Involvement (xGI), a comprehensive metric used to assess a player’s overall offensive impact on the field.

xG in sports betting

Ultimately, Expected Goals (xG) is just one analytical tool among many, rather than a definitive solution for sports betting. Having access to the data does not automatically grant a wagering edge, as major sportsbooks look at the exact same xG statistics and tables when setting their lines. Instead, strategic bettors use xG alongside other data to find potential value by spotting discrepancies between a team’s underlying performance and its market price.

For example, if a team like Newcastle United during the 2023–2024 season is generating high xG figures but sportsbooks like Beazt Sports are still offering favorable odds, it could be an indicator worth exploring. However, relying entirely on xG is generally not recommended. Bookmakers account for a wide range of complex variables when adjusting their boards.

To build a consistent strategy, it is helpful to look at the complete picture by reviewing recent form, analyzing injury reports, and cross-referencing that information with xG metrics. Because xG is subjective and depends heavily on the specific model used, it can vary across different data providers. It serves as a helpful reference point, but it should be treated as a guide rather than a guarantee.

The pitfalls of expected goals

While discussion often focuses on the utility of Expected Goals (xG), it is also important to consider its limitations. Because xG relies on underlying statistical assumptions, interpreting the data requires a closer look at the context behind the numbers. Consider a scenario where a team’s front office is evaluating two different forwards based on their past performances.

Comparing their respective time in the Premier League, with Erling Haaland playing 66 games and Timo Werner appearing in 69, the statistical outcomes show a noticeable difference. Haaland scored 63 goals from an expected total of 64.41 xG, performing very close to his statistical baseline. On the other hand, Werner recorded 12 goals from an expected 23.81 xG. A quick glance at these numbers might lead to the conclusion that one player is highly efficient while the other struggles under pressure, but xG data does not always tell the whole story.

The metric generally focuses on the shot itself rather than the broader human and situational elements that influence a match. For example, xG models typically do not account for coaching philosophies, specific tactical roles, the quality of service from teammates, or psychological factors.

Over a multi-season sample, variables like team chemistry, tactical adjustments, and individual fitness can significantly affect a player’s final output. While xG provides a helpful measurement of the opportunity, it cannot fully capture the individual variables of the player executing the shot.

The future of xG?

It is clear that Expected Goals (xG) has become an established part of modern soccer analytics. However, this is likely just the beginning of how data will be used in the sport, as the industry continues to move toward more comprehensive evaluation models. Basic statistics like expected goals and expected assists are increasingly being combined into Expected Goal Involvement (xGI), providing a single, consolidated metric to evaluate a player’s overall attacking impact.

Future models are expected to expand this analytical approach to defensive and transition plays as well. This could include metrics such as expected interceptions, expected dribbles, expected tackles, and expected saves. These detailed data points will likely serve as the foundation for new artificial intelligence and machine learning applications in sports science.

As these systems develop, they will become increasingly useful for tracking player performance, supporting development programs, and refining predictive models to provide a more accurate outlook on the game.