What is xG Against (xGA)?

xG Against — often abbreviated as xGA or Expected Goals Against — is a statistical metric used in football analytics to measure the expected number of goals a team should have conceded based on the quality of chances they allowed. Rather than simply counting goals conceded, xG Against evaluates the probability that each shot faced would result in a goal, then sums these probabilities to provide an overall measure of defensive performance.

In essence, xG Against answers a critical question: "How many goals should this team have conceded, given the quality and quantity of chances their opponents created?" This shifts the focus from outcomes (actual goals conceded) to process (the quality of defending and chances allowed), providing a more nuanced view of defensive performance.

Definition and Core Concept

xG Against works by assigning each shot faced by a team a probability value between 0 and 1, representing the likelihood that shot would result in a goal. These individual probabilities are then summed across a match, season, or any time period to create the xGA figure.

For example, if a team faces five shots with xG values of 0.70, 0.15, 0.08, 0.03, and 0.02, their xG Against for that match would be 0.98 — meaning they should have conceded roughly one goal based on the quality of chances allowed.

The metric is built on historical data from millions of shots across professional football, allowing advanced models to calculate the probability of any given shot becoming a goal based on numerous factors. This probability model is the same one used to calculate xG (Expected Goals) for attacking teams — the only difference is that xGA applies it to shots faced rather than shots taken.

Why xG Against is Different from Goals Conceded

The distinction between xG Against and actual goals conceded is crucial. Goals conceded tells you the outcome — what actually happened. xG Against tells you the process — what should have happened based on chance quality.

Consider a match where a team concedes 3 goals but has an xGA of 1.2. This reveals a significant gap: the team allowed relatively low-quality chances (hence the low xGA), but their goalkeeper or defenders made errors, or the opposition was particularly clinical. Conversely, a team might concede 0 goals with an xGA of 2.5, suggesting they defended well or their goalkeeper made outstanding saves.

This distinction matters because:

• Goals conceded can be lucky or unlucky — a team might concede a world-class finish from a low-probability chance, or they might allow 10 high-quality chances and the opposition misses them all.
• xGA isolates defensive process — it shows whether the team's defensive structure, positioning, and tactical approach are actually preventing high-quality chances.
• Short-term variance is real — over a single match or even a season, luck plays a significant role in the goals conceded vs. xGA gap. Over multiple seasons, this gap narrows, revealing true defensive quality.

How is xG Against Calculated?

Understanding how xG Against is calculated requires insight into the probability model and the factors that influence shot values.

The Probability Model

xG Against is calculated using a machine learning model trained on historical shot data. The most widely used models are based on data from Opta Sports, which has recorded nearly one million professional football shots with detailed contextual information.

The model assigns a probability to each shot by analyzing:

1. Shot location — distance from goal, angle to goal line
2. Shot type — open play, header, free kick, penalty, etc.
3. Defensive pressure — whether the shooter was under pressure
4. Assist type — how the ball was delivered (cross, pass, dribble, etc.)
5. Body part — foot (left/right) or head
6. Goalkeeper position — whether the goalkeeper was in an advantageous position
7. Defensive density — number of defenders nearby

Each shot is assigned a value between 0 and 1, where:

• 0.0 = virtually no chance of scoring (e.g., a shot from 40 meters with three defenders blocking)
• 0.5 = 50% probability of scoring
• 1.0 = virtually certain goal (e.g., an open tap-in from two yards)

The xGA for a team is simply the sum of all individual shot probabilities they conceded.

Factors Influencing xGA Values

Different shots generate vastly different xGA values based on their characteristics:

A team's xGA accumulates shot by shot. If a team concedes 15 shots in a match with an average xGA of 0.12 per shot, their total xGA would be approximately 1.8 goals.

Worked Example: Calculating xGA for a Match

Let's walk through a realistic example. A team plays a match and their opponent takes 12 shots:

In this match, the team's xG Against is 1.83 goals. If they actually conceded 2 goals, they're performing roughly to expectation. If they conceded 0, they defended exceptionally well (or the goalkeeper made brilliant saves). If they conceded 3, they underperformed defensively or the goalkeeper had an off day.

What Does xG Against Tell You About a Team's Defense?

xG Against is one of the most revealing metrics for assessing defensive quality because it isolates the defensive team's actions from the outcome.

Reading the Numbers

A team's xGA over a season provides insight into their defensive structure and consistency:

Low xGA (0.8–1.2 per match average):

• The team is preventing high-quality chances
• Defensive structure is compact and organized
• Opponents are forced to take low-probability shots
• Suggests sustainable, strong defending

High xGA (2.0–3.0+ per match average):

• The team is allowing dangerous chances
• Defensive structure is disorganized or tactically vulnerable
• Opponents are getting into high-quality scoring positions
• Suggests defensive problems that may not be reflected in goals conceded if the goalkeeper is overperforming

Moderate xGA (1.3–1.9 per match average):

• Typical for most teams; defensive performance is average
• Some tactical vulnerabilities but not critical
• Sustainable over time

The key insight is that xGA reveals process independent of outcome. A team might sit top of the league with a respectable goals conceded total, but if their xGA is very high, they're likely benefiting from either goalkeeper heroics or opponent finishing failures. This is unsustainable — over time, actual goals conceded will regress toward xGA.

Defensive Structure and Tactics

A team's xGA is heavily shaped by their defensive system and tactical approach:

Defensive systems that reduce xGA:

• Compact defensive block — Teams that defend with narrow spacing between players force opponents to take shots from wider angles or greater distances, reducing xGA
• High pressing — Aggressive pressing prevents opponents from settling into shooting positions, reducing xGA
• Rest-defence structure — Strong defensive organization during transitions prevents dangerous counter-attacks
• Blocking and clearing — Defenders who actively block shots and clear danger reduce xGA

Defensive systems that increase xGA:

• High defensive line with poor coverage — Leaves space in behind for through balls and dangerous chances
• Disorganized pressing — Pressing without a clear structure leaves gaps in midfield
• Slow transition defense — Sluggish recovery allows opponents time and space to create chances
• Poor set-piece organization — Weakness at corners and free kicks inflates xGA

For example, Manchester City under Pep Guardiola typically has very low xGA (often under 1.0 per match) because their compact, pressing system forces opponents into low-quality shots. Conversely, a team playing a passive low-block might have higher xGA because they're allowing opponents into dangerous areas, even if they're defending the box well.

Identifying Defensive Weaknesses

xGA can reveal tactical vulnerabilities that aren't obvious from goals conceded alone:

• Weak flanks — If xGA is concentrated on wide areas, fullbacks may be vulnerable
• Vulnerable to crosses — High xGA from headers suggests set-piece or crossing weaknesses
• Transition defense problems — High xGA on counter-attacks indicates poor defensive organization during transitions
• Central vulnerability — High xGA from central areas suggests centre-back positioning issues

Coaches and analysts use xGA breakdowns (by location, assist type, game state) to diagnose and fix defensive problems.

How Does xG Against Evaluate Goalkeeper Performance?

One of the most important applications of xG Against is assessing goalkeeper performance — but it's crucial to understand its limitations.

The Goalkeeper Problem: Why xGA Alone Isn't Enough

xG Against measures the quality of chances faced, not the goalkeeper's ability to stop them. A goalkeeper who faces an xGA of 1.5 and concedes 1 goal is performing well, but xGA doesn't tell us how well — it doesn't capture shot-stopping ability, reflexes, positioning, or decision-making.

Consider two scenarios:

Scenario 1: A goalkeeper faces an xGA of 0.8 and concedes 0 goals.

• This suggests excellent shot-stopping, but we don't know if the goalkeeper made brilliant saves or if opponents simply missed.

Scenario 2: A goalkeeper faces an xGA of 2.5 and concedes 2 goals.

• This suggests they're overperforming (conceding fewer than expected), but we don't know if it's due to excellent positioning, luck, or last-second blocks by defenders.

xGA alone conflates defensive structure (how many chances are allowed) with goalkeeper skill (how well those chances are converted into goals). To isolate goalkeeper performance, we need a more advanced metric.

Introduction to Post-Shot xG (PSxG)

Post-Shot xG (PSxG) is a more advanced metric that measures how likely a shot is to become a goal after it's been taken, accounting for the shot's trajectory, placement, and speed.

Where xG Against measures shot quality before the ball is struck, PSxG measures shot danger after the ball is in flight. This allows us to isolate the goalkeeper's role: the difference between PSxG and goals conceded reveals how well the goalkeeper is actually shot-stopping.

Key differences:

For example:

• A goalkeeper faces an xGA of 1.8 but a PSxG of 1.2
• They concede 1 goal
• Interpretation: The defense allowed decent chances (xGA 1.8), but the goalkeeper positioned themselves well to reduce danger (PSxG 1.2), and then made a good save to concede only 1 goal
• Conclusion: Strong goalkeeping combined with decent defending

Using xGA and PSxG Together

The most complete picture of defensive and goalkeeper performance comes from using both metrics:

Example goalkeeper performance profiles:

1. xGA 1.5, PSxG 1.3, Goals 1.2 → Elite goalkeeper with good defense (overperforming on both metrics)
2. xGA 1.5, PSxG 1.6, Goals 2.1 → Weak defense but poor goalkeeper (underperforming on PSxG)
3. xGA 2.0, PSxG 1.4, Goals 1.3 → Strong goalkeeper compensating for weak defense
4. xGA 0.9, PSxG 0.95, Goals 1.2 → Elite defense but struggling goalkeeper

xG Against vs. Actual Goals Conceded: What's the Difference?

The gap between xG Against and actual goals conceded reveals important information about luck, goalkeeper performance, and defensive consistency.

When Goals Conceded Exceed xGA

If a team's goals conceded significantly exceed their xGA, one or more of the following is true:

1. Goalkeeper underperformance or errors

• The goalkeeper is making mistakes or poor positioning decisions
• Example: xGA 1.2, Goals 3 → suggests goalkeeper errors or deflections

2. Bad luck / Exceptional finishing

• Opponents are converting low-probability chances at an unusually high rate
• Example: xGA 0.6, Goals 2 → opponents finished clinically from rare opportunities

3. Deflections and own goals

• Shots are being deflected in unexpected ways
• Example: xGA 1.1, Goals 2 → one goal was deflected in unexpectedly

4. Set-piece vulnerability

• The team is weak at defending set-pieces, which xGA may underweight

Over a single match, this gap is normal and expected. Over a season, persistent underperformance (goals >> xGA) suggests either genuine goalkeeper issues or systematic defensive problems.

When Goals Conceded Are Below xGA

If a team's goals conceded are significantly below their xGA, they're overperforming defensively:

1. Elite shot-stopping

• The goalkeeper is making exceptional saves
• Example: xGA 2.2, Goals 1.0 → goalkeeper is performing brilliantly

2. Defensive blocks and last-second interventions

• Defenders are making crucial blocks or clearances
• Example: xGA 1.8, Goals 0.8 → strong defensive organization

3. Opponent finishing failures

• Opponents are missing high-quality chances
• Example: xGA 1.5, Goals 0.5 → opponents were clinical failures

4. Luck

• The ball is bouncing favorably; shots are hitting the post or going wide
• Example: xGA 1.2, Goals 0.3 → fortunate escapes

Like underperformance, overperformance is unsustainable over long periods. A team with xGA 2.0 but only 1.0 goals conceded will likely regress toward the xGA figure over time. This regression principle is crucial for predicting future performance.

When They're Equal

If goals conceded ≈ xGA, the team is performing at the expected level. Their defensive structure is allowing chances that correspond roughly to the goals they're conceding. This is sustainable performance and suggests:

• Defensive organization is consistent
• Goalkeeper is performing at a normal level
• No significant luck (positive or negative)
• Future goals conceded will likely track xGA

Common Misconceptions About xG Against

Several myths about xG Against persist in football discourse. Understanding these clarifications is essential for proper interpretation.

Misconception 1: "Low xGA Means Perfect Defending"

Reality: Low xGA indicates that opponents weren't allowed into dangerous positions, but it doesn't necessarily mean the defending was active or aggressive.

A team playing a deep, passive low-block might have very low xGA because they're sitting deep and forcing opponents to shoot from distance. However, this doesn't mean the defense is performing brilliantly — it means the team is absorbing pressure and limiting space. In contrast, a team pressing high and allowing more shots (higher xGA) might actually be defending more effectively because they're preventing dangerous positions from being reached in the first place.

Key insight: xGA must be interpreted in context of tactical philosophy. A low-block team will naturally have higher xGA than a pressing team, even if both are defending well.

Misconception 2: "xGA Perfectly Predicts Future Goals Conceded"

Reality: xGA is a probability measure, not a guarantee. Over a single match or even a season, significant variance is normal.

xG Against is based on historical probabilities, but actual outcomes involve randomness. If a team has an xGA of 1.5, we'd expect them to concede around 1.5 goals, but they might concede 0, 1, 2, or 3 goals — all within normal variance. Over multiple seasons and hundreds of shots, actual goals conceded will approach xGA, but short-term deviations are inevitable.

Statistical concept: The larger the sample size (more matches, more shots), the more reliable xGA becomes as a predictor. A single match's xGA is less predictive than a season's xGA.

Misconception 3: "xGA Doesn't Account for Goalkeeper Skill"

Reality: xGA doesn't isolate goalkeeper skill — that's intentional. It measures the quality of chances allowed, not how well those chances are saved.

This isn't a flaw; it's by design. xGA is meant to evaluate defensive structure and organization. To measure goalkeeper skill specifically, you need PSxG (Post-Shot xG), which accounts for shot trajectory and placement. Using xGA to evaluate goalkeepers is like using team xG to evaluate individual strikers — it's conflating team and individual performance.

Correct approach: Use xGA to assess defense, PSxG to assess goalkeeper shot-stopping ability, and the gap between PSxG and goals conceded to measure overall goalkeeper performance.

Where Did xG Against Come From?

Understanding the history of xG Against provides context for why it's become so important in modern football.

The Origins of Expected Goals

Expected Goals (xG) was invented in 2012 by Sam Green at Opta Sports. Green adapted concepts from baseball sabermetrics (particularly the idea of measuring quality rather than just outcomes) to football, creating a probability model for shot outcomes.

The motivation was straightforward: football's traditional statistics (shots, shots on target, goals) don't capture the quality of chances. A team might have 20 shots but if they're all from 30 yards, they're unlikely to score. Another team might have 5 shots but all from inside the box — they're far more likely to score. xG quantifies this difference.

The metric was revolutionary because it shifted analytical focus from outcomes (goals) to process (chance quality). This had immediate applications:

• Player evaluation: A striker with 10 goals but 15 xG is overperforming (lucky); a striker with 5 goals but 12 xG is underperforming
• Team assessment: A team's xG difference (xG for minus xG against) is more predictive of future league position than current points
• Tactical analysis: xG reveals whether a team's system is creating/preventing high-quality chances

xG Against emerged naturally as the defensive counterpart to xG — if we're measuring attacking quality, why not defensive quality?

Evolution and Adoption in Modern Football

From 2012 to the present, xG Against has evolved from a niche metric used by hardcore analysts to a mainstream tool:

2012–2016: Specialist use only

• Analytics companies (Opta, StatsBomb) develop models
• Academic interest grows
• Few clubs use it formally

2016–2020: Growing adoption

• Premier League clubs begin integrating xG/xGA into scouting and analysis
• Media outlets (BBC, Sky Sports) start reporting xG metrics
• Fantasy football platforms include xG data

2020–present: Mainstream acceptance

• xG is standard in match analysis across all major broadcasters
• Betting markets use xG for odds-setting
• Clubs publicly discuss xG in press conferences
• Player recruitment heavily influenced by xG metrics

Notable adoption examples:

• Liverpool: Used xG analysis to recruit Mohamed Salah in 2017; Salah's xG values convinced manager Jurgen Klopp
• Brighton: Known for heavy reliance on xG/xGA in recruitment and tactical analysis
• Manchester City: Uses advanced xG variants in their analytics framework
• Brentford: Famously built their promotion to the Premier League partly on xG-based recruitment

Today, xG Against is considered a standard metric for any serious football analysis, from coaching staff to fantasy football players to sports bettors.

How Do Coaches and Analysts Use xG Against?

xG Against has numerous practical applications for coaches, scouts, analysts, and bettors.

Defensive Evaluation and Tactical Analysis

Coaches use xGA to assess whether their defensive system is working:

Early-season detection of luck:

• A team might be 3rd in the league with only 8 goals conceded, but their xGA is 18
• This suggests they're getting lucky and will likely concede more goals as the season progresses
• The coach can preemptively address defensive issues before the league table reflects the problem

Identifying tactical weaknesses:

• If xGA is high from crosses, the team has set-piece or wide-area vulnerabilities
• If xGA is high from central areas, centre-back positioning or midfield coverage is the issue
• If xGA is high on transitions, defensive organization during turnovers needs improvement

Consistency assessment:

• A team with xGA 1.2 per match is more consistent than a team with xGA ranging from 0.5 to 2.5
• xGA variance reveals whether the system is reliable or dependent on individual performances

Recruitment and Player Scouting

Clubs use xGA data to identify and recruit defenders and goalkeepers:

Defender recruitment:

• Scouts look for defenders who consistently reduce xGA for their current team
• A defender who moves to a new club should reduce their new team's xGA if they're genuinely elite
• xGA data over multiple seasons reveals consistency and reliability

Goalkeeper recruitment:

• Clubs compare PSxG and goals conceded to identify elite shot-stoppers
• A goalkeeper with PSxG 1.8 but only 1.2 goals conceded is overperforming and likely to be a good signing
• xGA context helps understand whether a goalkeeper's goals conceded figure is due to their skill or their defense

Example: A Premier League club evaluates a centre-back from a lower league. They see that the defender's team conceded 45 goals (poor) but had an xGA of 52 (suggesting the defense was actually allowing dangerous chances). The defender didn't reduce xGA significantly, suggesting he's not elite. However, if the defender's team had xGA 38 with 45 goals conceded, he's underperforming on PSxG, suggesting goalkeeper issues rather than his own weakness.

Match Preparation and Opponent Analysis

Analysts use xGA to understand how opponents defend and where they're vulnerable:

Defensive vulnerability identification:

• If the upcoming opponent has high xGA from crosses, the team should plan wide attacks
• If the opponent has high xGA from central areas, central penetration is viable
• If the opponent's xGA is high on transitions, counter-attacking might be effective

Goalkeeper assessment:

• Understanding the opponent's goalkeeper's PSxG vs. goals conceded reveals whether they're over- or underperforming
• If the goalkeeper is overperforming (PSxG > goals conceded), expect them to regress; if underperforming, expect improvement or continued struggles

Tactical planning:

• If the opponent typically allows low xGA, a direct, counter-attacking approach might be more effective than trying to build play
• If the opponent allows high xGA, patient build-up play might expose their defensive structure

Limitations of xG Against

While xG Against is a powerful metric, it has important limitations that must be understood for proper interpretation.

What xGA Doesn't Capture

Goalkeeper positioning (pre-shot):

• xGA is calculated before the ball is struck, so it doesn't account for whether the goalkeeper is well-positioned or out of position
• A goalkeeper who's 10 yards off their line might make a shot easier to score, but xGA doesn't capture this
• This is why PSxG exists — to measure post-shot danger and goalkeeper positioning effectiveness

Individual defender performance:

• xGA is a team metric; it doesn't isolate individual defenders' contributions
• A centre-back might be excellent, but if fullbacks are poor, team xGA will be high
• Advanced metrics (pressure success rate, tackle success rate) are needed to evaluate individuals

Set-piece organization:

• xGA may underweight or overweight set-pieces depending on the model
• Some models treat all headers equally, but set-piece headers have different characteristics than open-play headers
• Teams with excellent set-piece defense might have higher xGA than their actual defensive performance merits

Contextual factors:

• xGA doesn't account for fatigue, injuries, or player availability
• A team with key defenders injured will have higher xGA, but it's not a reflection of system quality
• Fixture difficulty (playing top teams vs. bottom teams) affects xGA but isn't normalized in the metric

Sample Size and Statistical Variance

xG Against is a probability-based metric, so sample size matters enormously:

• Single match: xGA is highly volatile. A team might have xGA 0.5 one match and 3.0 the next.
• 5–10 matches: xGA begins to stabilize, but variance is still significant.
• Full season (30+ matches): xGA becomes reliable. Actual goals conceded should approximate xGA.
• Multiple seasons: xGA is highly predictive. Multi-season xGA is one of the best indicators of defensive quality.

Regression to the mean:

• If a team has an unusually low xGA one season, they're likely to have higher xGA the next season (not because they got worse, but because the first season was an outlier)
• Similarly, unusually high xGA regresses downward

Teams and analysts must be patient with xGA data. A single season isn't enough to draw firm conclusions; 2–3 seasons of consistent xGA data is far more reliable.

Context Matters: Tactical Philosophy

The same xGA value can mean different things depending on tactical approach:

Low-block team with xGA 1.8:

• This is normal and expected; they're absorbing pressure and forcing long-range shots
• Not a defensive problem

Pressing team with xGA 1.8:

• This is concerning; pressing should prevent opponents from reaching dangerous areas
• Suggests pressing structure is breaking down

Defensive philosophy context:

• A team pressing high will naturally have higher xGA (opponents get more shots) but fewer high-quality chances
• A team sitting deep will have lower shot count but higher xGA per shot (more dangerous chances)
• xGA must be interpreted alongside shot count and shot location data

FAQ — Frequently Asked Questions About xG Against

What is the difference between xG Against and xG Difference?

xG Against (xGA) measures the expected goals conceded by a team. xG Difference (xGD) is the difference between xG (goals expected to score) and xGA (goals expected to concede). xGD is often more predictive than league position because it shows overall attacking and defensive quality. A team with xGD of +5 (creating 5 more expected goals than they concede) is likely to climb the league, regardless of current position.

How does xG Against help evaluate goalkeeper performance?

xGA shows the quality of chances a goalkeeper faces, but doesn't isolate their shot-stopping ability. A goalkeeper facing xGA 1.5 and conceding 1 goal is overperforming, but we don't know if it's due to excellent positioning, luck, or defender blocks. PSxG (Post-Shot xG) is needed to measure goalkeeper shot-stopping specifically. The gap between PSxG and actual goals conceded reveals true goalkeeper performance.

Can a team have low xG Against but still concede many goals?

Yes, this indicates either goalkeeper underperformance, deflections, or exceptional opponent finishing. If a team has xGA 1.0 but concedes 3 goals, their defense is allowing low-quality chances, but they're being punished for it. This is unsustainable — over time, goals conceded regress toward xGA. It suggests either goalkeeper issues or that the team is getting unlucky.

What factors increase or decrease xG Against?

Factors increasing xGA (allowing more dangerous chances):

• Poor defensive structure or organization
• Allowing central shots (more dangerous than wide shots)
• Losing defensive duels
• Slow transition defense
• Failed pressing traps
• Large spaces between defensive lines

Factors decreasing xGA (preventing dangerous chances):

• Compact defensive block
• Good rest-defence structure
• Strong centre-backs
• Aggressive counter-pressing
• Forcing opponents wide
• Blocking shooting lanes
• Elite goalkeeping positioning

Is xG Against used by professional clubs?

Yes, extensively. Premier League clubs, European elite teams, and even lower-league clubs now use xGA for tactical analysis, recruitment, and match preparation. Clubs like Brighton, Brentford, and Manchester City are known for heavy reliance on xG metrics in their operations.

How does xG Against differ from traditional defensive statistics like tackles and interceptions?

Traditional stats (tackles, interceptions, clearances) measure actions taken, not effectiveness. A defender might make 10 tackles but still allow dangerous chances. xGA measures outcome — whether the defense is actually preventing high-quality shots. A team with low xGA and high tackles is defending effectively; high xGA with high tackles suggests defenders are reacting to problems rather than preventing them.

Why is post-shot xG (PSxG) important if we already have xG Against?

xGA measures pre-shot chance quality (before the goalkeeper has any influence). PSxG measures post-shot danger (after the ball is in flight), isolating goalkeeper positioning and reflexes. Together, they provide a complete picture: xGA shows if the defense is allowing dangerous chances, and PSxG shows if the goalkeeper is positioned to save them. This separation is crucial for understanding defensive and goalkeeper performance.

Related Terms

• Expected Goals (xG) — The attacking counterpart to xGA; measures expected goals scored
• Clean Sheet — A match in which a team concedes zero goals
• xA (Expected Assists) — Expected assists based on shot quality created
• Post-Shot xG (PSxG) — Expected goals based on shot trajectory and goalkeeper positioning
• Expected Goal Difference (xGD) — The difference between xG and xGA; indicates overall team quality