Set-Piece Expected Goals Explained
A practical guide to set-piece expected goals, including shot classification, chance chains, delivery, routines, opposition effects and sample-size limitations.
Set-piece expected goals measures the quality of shots generated from dead-ball situations such as corners, indirect free kicks and, depending on the data provider, throw-ins and direct free kicks. A shot valued at 0.20 xG would historically be scored approximately 20% of the time under the assumptions of that particular model.
The metric can reveal an attacking or defensive set-piece advantage that ordinary goal totals conceal. However, set-piece xG usually values the final shot rather than every action that created it. Delivery accuracy, blocking movements, first contacts and second-ball recoveries may therefore determine the outcome without receiving separate xG credit.
It should be interpreted alongside opportunity volume, routine design, opponent behaviour and sample size—not as a complete measure of set-piece quality.
What Is Set-Piece Expected Goals?
Set-piece expected goals, commonly shortened to set-piece xG, is the expected-goals value assigned to shots originating from specified dead-ball situations.
For each qualifying shot, an xG model estimates the probability of a goal by comparing the attempt with relevant historical shots. Common inputs include:
- shot location and angle;
- distance from goal;
- body part;
- assist or delivery type;
- whether the chance was a header;
- the position of defenders and the goalkeeper where tracking data is available; and
- other contextual information included by the provider.
A team’s set-piece xG is normally found by adding the values of all shots classified as set-piece attempts:
Total set-piece xG = xG of shot 1 + xG of shot 2 + ... + xG of shot n
The underlying principles are the same as conventional expected goals. The important additional question is how the provider decides where a set-piece sequence starts and ends.
Which Situations Count as Set Pieces?
There is no universal set-piece xG definition shared by every data provider. Depending on the dataset, the category may include:
- corners;
- indirect free kicks;
- direct free kicks;
- long or attacking throw-ins;
- shots following an initial clearance; and
- follow-up attempts after penalties.
Penalties should normally be separated from non-penalty set pieces. Their high individual xG value can dominate the total and obscure how effectively a team creates from corners and free kicks.
Provider definitions also affect second phases. One model may classify a shot as set-piece xG when it follows an initial clearance and immediate return into the area. Another may decide that the original phase has ended and classify the same attempt as open play.
Comparisons are meaningful only when the underlying definition remains consistent.
Set-Piece xG Is the End of a Chance Chain
A set-piece shot is the final measurable output of several connected actions. A useful analysis follows the complete chain rather than looking only at the xG total.
| Stage | Example event | What can be measured? | Main analytical question |
|---|---|---|---|
| 1. Opportunity | Attacking team wins a corner | Corner volume, location and game state | How often does the team earn useful set-piece opportunities? |
| 2. Delivery | Inswinging ball targets the central six-yard area | Target zone, height, speed, curve and accuracy | Does the ball reach the intended dangerous area? |
| 3. Routine | Near-post runner moves a defender and creates space centrally | Runs, blocks, spacing and defensive reactions | Does the design create an advantageous matchup? |
| 4. First contact | Attacker heads the ball back across goal | First-contact winner, location and direction | Does the attacking team control the initial contest? |
| 5. Second ball | Loose ball falls to an attacker eight metres from goal | Recovery location, defensive pressure and time to shoot | Is the team structured to exploit the next action? |
| 6. Shot | Central shot valued at 0.28 xG | Shot location, angle, body part and pressure | How likely was this particular attempt to become a goal? |
| 7. Outcome | Goalkeeper saves the attempt | Goal, save, block, miss or post | Did the outcome differ from the underlying chance quality? |
In this example, the team records 0.28 set-piece xG even though it does not score. The delivery and routine helped create the chance, but a conventional shot-based model normally assigns the xG to the shot rather than dividing it between the taker, runners and first-contact player.
The opposite problem can also occur. A poor delivery might be cleared badly and produce a high-quality shot through defensive error. The resulting xG is real—the chance existed—but it may exaggerate the repeatable quality of the attacking routine.
This is why set-piece xG belongs inside the wider set-piece analytics process.
Different Ways to Measure Set-Piece xG
The most useful version of the metric depends on the question being asked.
| Measure | Calculation | What it helps explain | Main limitation |
|---|---|---|---|
| Total set-piece xG | Sum of qualifying shot xG | Overall chance value created or conceded | Heavily affected by opportunity volume and minutes played |
| Set-piece xG per match | Total set-piece xG ÷ matches | Average contribution to match-level chance creation | Fixture and game-state differences remain |
| Set-piece xG per opportunity | Total set-piece xG ÷ corners and relevant free kicks | How much chance value each opportunity produces | Depends on a consistent opportunity definition |
| Set-piece xG per shot | Total set-piece xG ÷ set-piece shots | Average quality of the attempts created | Ignores how frequently opportunities become shots |
| Set-piece xG difference | Set-piece xG for − set-piece xG against | Net attacking and defensive set-piece performance | Can be unstable in small samples |
Volume Versus Efficiency Example
Suppose two teams each create 4.0 set-piece xG:
- Team A generates it from 80 relevant opportunities: 0.050 xG per opportunity.
- Team B generates it from 50 opportunities: 0.080 xG per opportunity.
Their total output is identical, but it was produced differently. Team A may be good at earning corners and free kicks, while Team B extracts more chance value from each opportunity.
That distinction matters when projecting future matches. Opportunity volume depends partly on possession, territory and opponent behaviour. Efficiency may depend more on delivery, personnel and routine execution.
Why Delivery Matters
A set-piece routine cannot create a shot unless the ball reaches a usable area or the team retains possession after the initial action.
Delivery analysis can consider:
- inswinging, outswinging, driven or floated deliveries;
- near-post, central, far-post and edge-of-area target zones;
- delivery height and speed;
- the taker’s preferred foot;
- short-corner variations;
- whether the goalkeeper can claim the ball; and
- how often the first defender clears it.
A team can record low set-piece xG because its deliveries repeatedly fail to beat the first defender. Another may reach good areas but lose the aerial contest. The identical xG outcome can therefore have different underlying causes.
Corner volume should also be kept separate from corner quality. The guide to corner markets and statistics explains why territory, tactical width, crossing and blocked shots affect how many corners a team wins. Those factors do not automatically reveal what the team does with them.
Why Routines and Personnel Matter
Set pieces are structured interactions rather than isolated crosses. Teams can use movement to:
- create separation for a primary target;
- move a dominant defender away from the delivery zone;
- block or delay a marker within the laws of the game;
- crowd the goalkeeper’s starting area;
- attack a planned flick-on zone;
- create a cutback after a short corner; or
- prepare for a clearance and second phase.
Personnel determines which routines are plausible. Delivery quality, aerial reach, timing, physical strength, movement and second-ball anticipation can all change the probability that an opportunity becomes a good shot.
A change of taker or the absence of a primary aerial target can make historic team averages less relevant. Conversely, a new routine or specialist coach may create a genuine structural improvement, although several successful matches are not enough to establish its size or durability.
Opposition Set-Piece Defence Changes the Estimate
Set-piece xG is not produced by the attacking team alone. It emerges from a matchup between the attacking routine and the defensive system.
Defensive factors include:
- zonal, player-to-player or hybrid marking;
- aerial strength and first-contact success;
- goalkeeper starting position and claim behaviour;
- protection of the six-yard area;
- players stationed at the posts;
- the height and positioning of the defensive line;
- how quickly defenders react to second balls; and
- whether players are left forward to discourage attacking overloads.
A team’s season-long attacking set-piece numbers should not be transferred unchanged to every opponent. A routine designed to isolate a small player may be less useful against a physically dominant defensive side. A goalkeeper reluctant to leave the goal line may create a different opportunity from one who claims crosses aggressively.
Good prediction work therefore asks whether the attacking strength and defensive weakness interact. Two independent rankings—strong set-piece attack and weak set-piece defence—are a starting point, not the conclusion.
How Set-Piece xG Affects Team Scoring Probabilities
A team’s expected scoring rate can be separated into components:
Total expected goals = open-play xG + set-piece xG + penalty expectation + other modelled components
Suppose a baseline model gives a team:
- 1.05 expected open-play goals;
- 0.22 expected non-penalty set-piece goals; and
- 0.08 expected goals from its modelled penalty expectation.
Its total expected scoring rate is 1.35 goals.
If opponent-specific analysis increases the non-penalty set-piece component from 0.22 to 0.34, the revised total becomes 1.47. That change can affect match-result, correct-score and team-total probabilities.
The adjustment should be cautious. Set-piece mismatches are already visible to sophisticated markets, while an apparent historical weakness may be driven by a few unusually valuable chances.
Why Sample Size Is a Major Limitation
Set-piece goals and high-quality chances are relatively infrequent. A team may take many corners but generate only a small number of shots from dangerous central locations.
This creates several problems:
- one major chance can dominate a short run of set-piece xG;
- one penalty or classification difference can distort the total;
- goals can diverge substantially from xG over limited samples;
- personnel and routines may change before the sample becomes stable;
- fixture difficulty can produce unequal opportunities; and
- red cards and unusual game states can inflate or suppress volume.
Suppose a team produces 1.10 set-piece xG across five matches, but 0.65 comes from one goalmouth scramble. The total is accurate under the model, yet the remaining opportunities contributed only 0.45 xG. Projecting the headline average without inspecting its distribution would create excessive confidence.
Analysts should examine longer samples, individual chances and opportunity-level measures. Recent data may better reflect current routines, but older data provides more observations. Neither window should be accepted automatically.
Provider and Classification Differences
Set-piece xG totals from different providers may disagree because their models and event definitions differ.
Possible sources of disagreement include:
- whether throw-ins are included;
- whether penalties are grouped with other set pieces;
- when a second phase becomes open play;
- how blocked shots are treated;
- which shot characteristics enter the xG model;
- whether defender and goalkeeper locations are available; and
- how own goals and deflections are recorded.
Decimal differences should not be treated as proof that one provider is wrong. The analyst should establish what each number measures before comparing teams, competitions or seasons.
Common Set-Piece xG Mistakes
Using Goals Instead of Chance Quality
A team can score several set-piece goals from modest chances or fail to score from strong opportunities. Goals describe outcomes; xG helps assess the quality of the shots behind them.
Treating xG as a Complete Routine Score
Conventional xG usually values shots. It may not directly credit a delivery that narrowly misses every runner or a movement that creates space but does not produce an attempt.
Ignoring Opportunity Volume
High total set-piece xG may reflect an unusually large number of corners rather than exceptional efficiency.
Ignoring the Opposition
Attacking and defensive averages are matchup inputs. They do not prove that the same advantage will appear against different marking systems or personnel.
Including Penalties Without Saying So
Penalties carry high xG values and can overwhelm the non-penalty set-piece signal. Definitions should be explicit.
Overreacting to Small Samples
A few goals, one large chance or a short run against weak opponents can make set-piece performance appear more stable than it is.
Double-Counting the Adjustment
If a team-strength model already includes historic set-piece performance, adding the entire set-piece advantage again can count the same evidence twice.
How GoalIQAI Uses Set-Piece xG
Set-piece xG should adjust a wider match assessment rather than replace it. A structured process is:
- Separate open-play, penalty and non-penalty set-piece performance.
- Confirm the provider’s classifications and sequence window.
- Measure opportunity volume, shot-creation rate, xG per opportunity and xG per shot.
- Review deliveries, first contacts, routines and second phases on video.
- Assess the opponent’s marking system, goalkeeper and available personnel.
- Check whether the sample is driven by one or two extreme chances.
- Adjust the team’s scoring probability cautiously and avoid double-counting.
- Compare the revised probability with the market rather than treating the mismatch as a guaranteed outcome.
This fits within the broader football match-analysis framework: establish the baseline, identify relevant contextual evidence, estimate its effect and preserve uncertainty.
Prediction articles should use set-piece xG when the evidence identifies a material matchup that changes scoring probabilities. A generic observation that one side “is good from corners” is not sufficient.
Key Takeaways
- Set-piece xG measures the quality of shots generated from defined dead-ball situations.
- Provider definitions differ, particularly for penalties, throw-ins and second phases.
- Conventional xG normally values the eventual shot rather than every delivery, run or first contact.
- Total xG should be separated from xG per match, per opportunity and per shot.
- Delivery quality, routines, personnel and second-ball structure help explain how the shots were created.
- Opposition marking, aerial ability and goalkeeper behaviour affect the matchup.
- Set-piece samples can be dominated by a small number of high-value chances.
- Open-play, penalty and non-penalty set-piece performance should be analysed separately.
- A set-piece mismatch matters only when it changes the estimated scoring distribution materially.
Related Guides
- Set-Piece Analytics Explained
- What Is Expected Goals?
- Corners Betting Explained
- Team Totals Betting Explained
- How Professional Football Bettors Build a Match Analysis Framework
- Football Betting and Analytics Knowledge Base
Stay Ahead of the Market
Subscribe to the GoalIQAI newsletter for evidence-based football analysis and practical guides to performance data, probability and betting markets.