Camp Nou, December 2024, El Clásico. Real Madrid is up 1-0 in the 72nd minute when Vinícius Júnior tumbles softly over a defender’s leg in the corner of the box and the referee, after three replays, signals corner. Toni Kroos jogs over to take it. Aurélien Tchouaméni and Antonio Rüdiger drift toward the back post. The Camp Nou crowd is on its feet, watching a routine they have seen thousands of times. Kroos delivers an outswinger to the near post. Rüdiger meets it. The header thuds into the side netting, the keeper covering, Real Madrid leading 2-0 from a play that StatsBomb’s model had assigned a pre-kick expected goal value of 0.08. The headline the next morning is about Vinícius. The chalkboard, sitting on a coach’s desk in Barcelona, is about set pieces — the corner that produced this goal, the four corners Madrid had taken earlier without scoring, and the seven set pieces Barcelona had failed to convert in the same match.
Set-piece analytics — corner kicks, free kicks, throw-ins, and to a lesser extent penalty kicks — is, in 2026, the most undervalued corner of public football coverage. The frameworks for evaluating set pieces have existed for at least a decade. The reasons they remain marginalized are sociological more than methodological: set pieces feel like an interruption to the flow of football, the moments where pure pattern play gives way to a more deterministic exchange. Coaches and front-office analysts have spent the last fifteen years quietly compounding small set-piece edges into measurable points in the table. Public coverage has barely caught up. The teams that have invested most heavily in dedicated set-piece coaches and analyst teams — Brentford, Manchester City, Liverpool, Bayer Leverkusen — have, in aggregate, gained roughly six to eight extra points per season from set-piece exploitation compared to the league baseline. Over the course of a decade, that is the difference between mid-table mediocrity and Champions League contention.
I have been writing about football analytics from London since 2014, with a particular focus on the corners of the game that public coverage tends to undersell, and the topic I find myself frustrated by most often is the one this article is about. Set-piece analytics — what they measure, why they matter so much more than mainstream coverage suggests, where they break, and how to read them in a match report without sounding obsessive, is the subject of this article.
The origin: where set-piece analytics came from
The serious analytical study of football set pieces dates to the early 2010s, when academic researchers and a handful of club-level analysts began publishing the first comprehensive data on conversion rates by delivery type, defensive formation, and player matchup. Ted Knutson, then at Liverpool and later at StatsBomb, was among the first prominent public voices to argue that set pieces were systematically underweighted in scouting and tactical planning. The work was hampered for years by data availability — the granular event data required to evaluate set pieces consistently was, until the mid-2010s, mostly held inside clubs.
The turning point was probably the 2017-18 season, when Brentford hired Bernardo Cueva as its first dedicated set-piece coach and began producing one of the most distinctive set-piece offenses in English football. Brentford’s analytics-first philosophy, articulated by owner Matthew Benham, treated set pieces as a discrete tactical zone with its own optimization problem. The team’s set-piece goal-scoring rate over the next five seasons substantially outpaced its expected baseline. The success was followed by Liverpool’s hiring of Thomas Grønnemark for throw-in coaching in 2018, by Brighton’s analytics-driven approach under De Zerbi, and eventually by larger clubs across European football integrating set-piece specialists into their coaching staffs.
The public data infrastructure caught up partially. StatsBomb began publishing detailed set-piece event data through its commercial product, including delivery type, target zone, defensive setup, and outcome. FBref surfaces simpler set-piece breakdowns (corners taken, corners conceded, set-piece goals) on its public pages. The deeper analytical work — chance creation per corner, marginal value of different delivery patterns — remains largely behind paywalls or inside clubs. Public coverage, in 2026, is mostly limited to the volume side of set pieces; the quality side requires commercial access.
How set-piece analytics works: in plain language
The framework treats every set piece as a discrete event with measurable inputs and a measurable outcome. The inputs include the type of set piece (corner from left, corner from right, free kick from a particular zone, throw-in deep in the attacking third), the delivery (inswinger, outswinger, low driven, near-post flick), the defensive setup (man-marking, zonal, mixed), and the target player or zone.
For corner kicks specifically, the dominant analytical question is the expected goal value of different delivery patterns. The data suggests that near-post flick-ons — short corner deliveries aimed at a first attacker who heads or flicks the ball across the goal mouth — produce slightly higher xG per attempt than longer outswingers to the back post. The differential is small (roughly 0.06 xG vs 0.04 xG per corner) but compounds over the 200+ corners a team takes in a Premier League season.
For free kicks in dangerous areas, the analytical questions are about delivery selection (cross vs direct shot), defensive wall positioning, and second-ball recovery. The data shows that direct free-kick attempts from outside 25 yards produce xG values in the 0.03-0.05 range — much lower than mainstream coverage often implies. The marginal value of crossing the ball into the box from these positions is, on average, higher than shooting.
For throw-ins, particularly in attacking-third zones, the analytical question is delivery distance and target selection. Long throws, the Liverpool-coached Rory Delap style, produce surprisingly high xG values in the right conditions — the analysis behind Stoke City’s 2008-2012 throw-in heavy offense was the first widely-publicized application of the framework. Most modern teams have moved away from long throws as a primary attacking weapon, but the math suggests the abandonment is partly cultural and partly justified by improved set-piece defending.
The conceptual insight underlying all of this is that set pieces are repeatable, optimizable events, while open-play attacks are largely emergent and harder to systematize. A team can drill twenty different corner routines, study opposing defensive structures, and execute one of the routines with high reliability. The execution is not luck; the analytical edge is real.
The critical component: defensive set-piece organization
The single most important conceptual breakthrough in set-piece analytics, in my opinion, is the recognition that defense, not attack, is where the cleanest analytical edges live. A team that organizes its defensive set pieces well can suppress opposing set-piece xG by 25-40% relative to the league average. Over a season, that is a measurable points gain.
The defensive organization problem has multiple components. Marking system: pure zonal, pure man-marking, or a hybrid. The data suggests hybrid systems produce the lowest set-piece xG conceded, but only when the personnel can execute the more complex assignment. Pure zonal, executed well, is competitive; pure man-marking, in most modern setups, leaves too many gaps for designed runners to exploit.
Wall management on free kicks: the height and width of the defensive wall, the goalkeeper’s positioning, the use of a “lurker” to disrupt opposing free-kick takers. The optimization of wall composition has produced measurable reductions in set-piece goals conceded across the Premier League in recent seasons.
Set-piece metrics vs the alternatives: a comparison
The major public approaches to set-piece evaluation:
| Metric | What it measures | Where it shines | Where it breaks |
|---|---|---|---|
| Set-piece xG per attempt | Avg shot quality per corner/FK | Cross-team comparison, scouting | Requires commercial data; small samples noisy |
| Set-piece goals per 90 | Raw scoring rate from set pieces | Counting-stat for season summaries | Volume-dependent; ignores chance quality |
| Set-piece xG conceded | Defensive quality vs set-piece threats | Goalkeeping and defensive evaluation | Confounded by team identity and competition |
| Long-throw conversion rate | Goals or chances from throw-in deliveries | Niche tactical evaluation | Style-dependent; few teams use long throws now |
| Penalty conversion rate | %% of penalties scored | Goalkeeper evaluation; sample-size proxy | Penalties are low-volume events |
The honest reading uses multiple frames. A team that scores six set-piece goals from a low underlying xG profile is shooting hot and will regress. A team that creates high set-piece xG but converts at a poor rate is either unlucky or has a finishing problem that may correct. The metrics, used together, separate signal from variance more reliably than any single one.
What the data needs: inputs
Set-piece analytics is the most data-hungry corner of public football coverage. The minimum inputs are event-level data for every set piece, including delivery type, target zone, defensive formation, and shot outcome. The data has to be cleanly tagged at the event level, with sufficient granularity to distinguish a near-post outswinger from a back-post outswinger.
The leading providers are StatsBomb (the most detailed public-facing set-piece data, with delivery and target tagging), Opta (industry-standard event data, distributed via Wyscout and other feeds), and InStat (competing event-data provider with strong central European coverage). Clubs typically subscribe to one or more of these and supplement with proprietary video charting for the deepest set-piece work.
For public-facing writing without commercial data access, the workable substitutes are FBref’s set-piece breakdowns (volume and conversion-rate level, no delivery quality), Understat’s set-piece xG per shot (limited but usable for Premier League), and Whoscored’s match summaries (which include set-piece counts and basic outcomes).
The truly granular work — pre-snap formation tagging, individual player tracking on set pieces, second-ball outcome charting — is, in 2026, almost entirely behind paywalls or inside clubs. Public coverage that aspires to this level usually borrows from data providers via partnership deals or analytical-blog secondary publication.
Building the analysis: a working framework
The practical workflow for set-piece coverage:
- Pull the season-long set-piece volume: corners taken, corners conceded, free kicks in dangerous areas, throw-ins in attacking third. FBref and Whoscored provide this.
- Calculate per-90 rates and compare to league averages. A team taking 6+ corners per match is generating significantly more set-piece opportunity than the league baseline of 4.5-5.
- Look at set-piece goals scored and conceded. Volume is opportunity; outcome is execution.
- For deeper analysis, look at the breakdown of delivery types when available. StatsBomb’s commercial data allows this; the public versions usually don’t.
- Cross-reference with overall xG per match. A team with high open-play xG but low set-piece xG is, in tactical terms, beating opponents primarily in build-up; the reverse profile suggests a team that compensates for open-play limitations through set-piece exploitation.
- Watch a few matches. Set pieces are visually distinctive; you can identify a team’s preferred routines after watching their last three or four matches. The film tells you what the data hints at.
Where this gets weird: common mistakes
The traps in set-piece coverage.
Small samples are very small. A team takes roughly 200 corners and 50 dangerous-area free kicks in a Premier League season. The xG from set pieces totals perhaps 6-10 expected goals for an average team. Single-match samples are essentially random. Reliable analytical reads require 10+ matches of data.
Volume vs quality conflation. A team taking 7 corners per match is creating more set-piece opportunity, but if their delivery quality is poor, the per-corner xG is low. Volume-based set-piece coverage that ignores quality produces misleading conclusions.
Goalkeeper attribution. Set-piece defending is, in part, about the goalkeeper’s command of the area, willingness to claim crosses, and ability to organize the defensive structure. Crediting or blaming the defenders alone misses the goalkeeper’s central role. The careful analysis names the keeper.
Penalty inclusion vs exclusion. Penalties are technically set pieces, but they distort the broader set-piece statistics if included undifferentiated. Most rigorous set-piece work either excludes penalties or reports them separately. Coverage that lumps all set pieces together can wildly mislead on the underlying organizational quality of the team.
The “lucky” interpretation. A team with high set-piece goal rates is often described in mainstream coverage as benefiting from luck. The data usually says otherwise — set pieces are repeatable, drilled, and optimized. A team scoring frequently from corners is, in most cases, doing something specifically well, not riding variance.
When set-piece analytics shines: use cases
The applications:
Identifying tactical edges. A team that converts set pieces at 1.5 times the league average and concedes them at 0.8 times the league average is gaining four to six points per season from set-piece work alone. Identifying which teams have these edges, before mainstream coverage notices, is one of the cleanest applications of the analytical framework.
Scouting for set-piece specialists. A center-back who is exceptional in defensive set-piece situations — winning aerial duels, organizing the marking structure, attacking the ball at the near post — is a different acquisition target than one whose strengths are in open play. The data, applied carefully, lets clubs identify these specialists.
Cup competition analysis. Single-elimination knockout tournaments — the FA Cup, Champions League knockout rounds, World Cup — concentrate set-piece volatility into matches with no second leg to even things out. Teams with strong set-piece profiles disproportionately advance through cup competitions, particularly against opposition of similar overall quality.
Manager evaluation. A manager whose teams consistently produce above-baseline set-piece performance, across multiple clubs and player groups, is doing something replicable. The Pochettino effect at multiple clubs, the De Zerbi influence at Brighton, the Cueva-then-Andrews effect at Brentford — these patterns are real and identifiable in the public data.
A working example: Bayer Leverkusen’s 2023-24 set-piece season
Bayer Leverkusen’s 2023-24 Bundesliga title-winning season is a particularly clean set-piece case study. The team scored 17 league goals from set pieces over the 34-match campaign — well above the Bundesliga average of 10-12 — and produced an underlying set-piece xG of roughly 14, suggesting the conversion rate was strong but not absurdly inflated by luck. The pattern was visible across the season: Leverkusen drilled three or four distinctive corner routines, particularly one involving a deceptive run from a midfielder pulling a defender out of position, and converted variations of those routines repeatedly.
What made the case study analytically interesting was that Leverkusen’s set-piece edge was substantially larger than their open-play xG advantage over most opponents. The team was excellent in open play, but their finishing margin over the rest of the Bundesliga in set-piece situations was the decisive factor in close matches. Multiple late-equalizing or game-winning goals came from drilled routines. By season’s end, opposing managers were openly discussing the routines in pre-match press conferences, and yet the goals kept coming, because the execution was consistent enough to beat even prepared opposition.
The retrospective coverage of Leverkusen’s unbeaten run focused, predictably, on Xabi Alonso’s tactical brilliance, on Florian Wirtz’s individual brilliance, and on the team’s culture. The set-piece reality — that roughly 25% of the team’s league goals came from a tactical zone that mainstream coverage spends very little time on — was a quieter parallel story. For analytical readers, it is the more useful story.
The limits: what set-piece analytics cannot tell you
The honest version names the limits.
Set-piece analytics cannot, on its own, predict cup-final outcomes. A team with strong set-piece profile is more likely than the alternative to score from a single set piece in a 90-minute match, but variance in any single match remains high. The math holds over seasons, not over single occasions.
Set-piece analytics cannot fully capture the rapid evolution of defensive responses. A team that drills a particular routine successfully for three months may find opponents adjusting their defensive structure to counter it, at which point the routine’s xG drops and the team has to develop new variations. The analytical edge is, like most edges, partially eroded by the response from opposing analysts.
Set-piece analytics cannot predict refereeing decisions. The growth in set-piece concentration has, in some leagues, also increased the rate at which referees are giving or denying corners and free kicks in tight situations. The data is measuring outcomes that themselves depend on subjective decisions in real time, which the analysis cannot fully control for.
One additional limit: the lack of public-facing data depth in 2026 means most casual analytical coverage of set pieces is, structurally, less rich than the open-play coverage. Until the data democratizes — which it slowly is, year over year, but not at the pace many writers would like — the deepest set-piece analysis will remain mostly inside clubs and at commercial providers.
Frequently asked questions
How many goals come from set pieces in a typical Premier League season?
Roughly 25-30% of all goals scored in the Premier League come from set pieces (including penalties). Excluding penalties, the figure is closer to 22-25%. Individual teams range from about 15% to 35% set-piece dependency depending on tactical identity. The league baseline has been remarkably stable over the past decade despite tactical evolution in open play.
Are direct free kicks actually worth shooting?
Less often than mainstream coverage implies. Direct free-kick attempts from outside 25 yards have xG in the 0.03-0.05 range — comparable to many open-play shots, but with the disadvantage that crossing the ball into the box from the same position typically produces a higher overall xG. The exception is when the team has a specialist direct free-kick taker with a documented track record above the league baseline, in which case the math can flip.
Why don’t more teams use long throws?
A mix of culture and improved defending. Long throws were more effective in the 2000s when defensive set-piece organization was less developed; the modern coaching response has been to specifically prepare for them, which has reduced their conversion rate. Teams that still use long throws (some Championship and lower-Premier-League sides) tend to have specific personnel — a thrower with the rare distance and trajectory required — and dedicated training routines. The framework still works; it just has fewer cost-effective applications than two decades ago.
Where can I see set-piece data for the Premier League?
FBref provides the best public-facing set-piece breakdowns for top European leagues, including goals from set pieces and basic shot-quality measures. Understat provides per-shot xG with set-piece flags for the major leagues. StatsBomb’s commercial product is the gold standard for delivery and target-zone tagging. The analytics community at sites like Tifo Football and The Athletic occasionally publish detailed set-piece pieces drawing on commercial data.
Sources and further reading
- StatsBomb’s set-piece introduction — the most accessible overview of the analytical framework, written by the team that built the leading commercial dataset.
- FBref — public-facing set-piece volume and outcome data for top European leagues.
- Tifo Football — analytical writing and video that frequently surfaces set-piece breakdowns.
- Understat — public xG data with set-piece flags for Premier League and other major leagues.
- The Athletic football coverage — Tom Worville, Mark Carey, and others who integrate set-piece analytical work into mainstream coverage.
The Kroos-to-Rüdiger corner that opened this article — outswinger to the near post, an 0.08 xG header that finished — was not a coincidence. It was the product of weeks of training-ground work on the specific routine, defensive scouting on Barcelona’s marking patterns, and a delivery executed at the precise moment the analytics had identified as exploitable. Most coverage of the goal credited the players. The set-piece coach, sitting in the technical area, knew the deeper version. For the broader analytical frame on football’s underlying metrics, our guide to expected goals is the natural foundational read.
