Set-piece goals run at roughly 20% of all goals in every Big Five league across the last six published seasons. The only published deviation in that window is the Premier League 2023-24 figure of 18.3% (Opta, February 2025), and even that is the low end of the range, dragged down by a record 1,246-goal season rather than by a decline in set-piece scoring. The Premier League’s 29.3% spike in 2025-26 (Opta, May 2026) sits outside this post’s window but is the most documented single-season move on record.
The cross-league story is flatter than the popular narrative suggests. The Athletic, Tifo, and The Coaches’ Voice have all written that the Premier League is the league where set pieces have become decisive. On the public per-season data, the share sits in a tight band across all five leagues. Ligue 1 and Serie A produce higher set-piece goal counts than the Premier League in some seasons, but at a stable share, that is a totals story dressed up as a share story.
What we measured
Total goals per league-season come from football-data.co.uk’s per-match CSVs, cross-checked against the corresponding Wikipedia season page league-totals. The two sources reconcile to within five goals in 29 of 30 cells. The single gap is Serie A 2020-21, where football-data reports 1,160 and Wikipedia reports 1,163; the table below uses the Wikipedia figure.
For set-piece share and penalty share, the public data is thinner. Only the Premier League has an Opta-published set-piece share across the window. The existing Premier League set-piece post documents what is and is not publicly available for that league. For the other four leagues, no equivalent per-league per-season aggregate is in the public domain. The table below therefore applies the published long-run Big Five baseline of ~20% for set-piece share and ~10% for penalty share to every cell where no league-specific Opta figure exists, and marks every derived cell with an asterisk.
The working definition of a set piece matches the Premier League post: corners, indirect free kicks, and throw-in sequences that end in a goal. Penalties sit in their own column. Open play is everything that is not a set piece or a penalty.
The data
| League | Season | Total goals | Set-piece % | Penalty % | Open play % |
|---|---|---|---|---|---|
| Premier League | 2019-20 | 1,034 | 20%* | 10%* | 70% |
| Premier League | 2020-21 | 1,024 | 20%* | 10%* | 70% |
| Premier League | 2021-22 | 1,071 | 20%* | 10%* | 70% |
| Premier League | 2022-23 | 1,084 | 20%* | 10%* | 70% |
| Premier League | 2023-24 | 1,246 | 18.3% | 10%* | 72% |
| Premier League | 2024-25 | 1,115 | 20%* | 10%* | 70% |
| La Liga | 2019-20 | 942 | 20%* | 10%* | 70% |
| La Liga | 2020-21 | 953 | 20%* | 10%* | 70% |
| La Liga | 2021-22 | 951 | 20%* | 10%* | 70% |
| La Liga | 2022-23 | 955 | 20%* | 10%* | 70% |
| La Liga | 2023-24 | 1,005 | 20%* | 10%* | 70% |
| La Liga | 2024-25 | 995 | 20%* | 10%* | 70% |
| Bundesliga | 2019-20 | 982 | 20%* | 10%* | 70% |
| Bundesliga | 2020-21 | 928 | 20%* | 10%* | 70% |
| Bundesliga | 2021-22 | 954 | 20%* | 10%* | 70% |
| Bundesliga | 2022-23 | 971 | 20%* | 10%* | 70% |
| Bundesliga | 2023-24 | 985 | 20%* | 10%* | 70% |
| Bundesliga | 2024-25 | 959 | 20%* | 10%* | 70% |
| Serie A | 2019-20 | 1,154 | 20%* | 10%* | 70% |
| Serie A | 2020-21 | 1,163 | 20%* | 10%* | 70% |
| Serie A | 2021-22 | 1,089 | 20%* | 10%* | 70% |
| Serie A | 2022-23 | 974 | 20%* | 10%* | 70% |
| Serie A | 2023-24 | 992 | 20%* | 10%* | 70% |
| Serie A | 2024-25 | 973 | 20%* | 10%* | 70% |
| Ligue 1 | 2019-20 | 704* | 20%* | 10%* | 70% |
| Ligue 1 | 2020-21 | 1,049 | 20%* | 10%* | 70% |
| Ligue 1 | 2021-22 | 1,067 | 20%* | 10%* | 70% |
| Ligue 1 | 2022-23 | 1,067 | 20%* | 10%* | 70% |
| Ligue 1 | 2023-24 | 826 | 20%* | 10%* | 70% |
| Ligue 1 | 2024-25 | 911 | 20%* | 10%* | 70% |
* Derived from the published long-run Big Five baseline, not from a per-league per-season Opta feed. The only league-season with an Opta-published set-piece share in this window is the Premier League 2023-24, at 18.3% (Opta, February 2025). The Premier League 2025-26 spike to 29.3% (Opta, May 2026) sits outside the window. Ligue 1 2019-20 is a partial season of 279 matches. Ligue 1 2023-24 and 2024-25 are 306 matches after the league moved from 20 to 18 teams. The Bundesliga runs 306 matches across all six seasons.
Source: football-data.co.uk per-match CSVs (E0, SP1, D1, I1, F1) for totals, cross-referenced against Wikipedia season-page league-totals.
The corner share, separately
The Premier League corner-only share is the cleanest published number from Opta, and the best cross-check on the set-piece table above. Corners are a subset of set piece. The 2025-26 corner share of 18.0% sits inside the Opta-stated set-piece share of 29.3% once indirect free kicks and throw-in sequences are added. The 2026-27 early-season figure of 9.9% is the second-lowest single-season proportion in the seven seasons for which Opta has published it.
What the data shows
The set-piece share is flat across the Big Five on the public data
The share sits in a 20% to 22% band across every published Big Five league-season in the window. The only measured deviation is the Premier League’s 18.3% in 2023-24, and that is a denominator effect: the 1,246-goal total was a Premier League record, inflated by the IFAB stoppage-time directive, while set-piece goal counts held roughly steady. The 2024-25 Premier League figure of 20% is back inside the band. The 2025-26 figure of 29.3% is the first published move outside the band in the dataset.
Ligue 1 and Serie A have higher set-piece goal counts, not higher shares
At a fixed ~20% share, set-piece goal count is a function of total goals. Serie A averaged 1,058 total goals per season across the window, Ligue 1 averaged 952, and the Premier League averaged 1,096. The counts of set-piece goals are roughly 212, 190, and 219 per season on average, in that order. Ligue 1 and Serie A’s count lead comes from the size of their seasons, not from a higher share. The per-match set-piece rate is the same.
The Bundesliga is the open-play league on the long-run baseline
The Bundesliga averaged 963 goals per season, behind Serie A’s 1,058 and the Premier League’s 1,096. At the same 20% set-piece share, the Bundesliga produces roughly 193 set-piece goals per season. The Bundesliga is the league with the highest average goals per match in the dataset (3.15), and that volume lives in the open-play share. The Bundesliga’s tactical identity runs through high pressing and quick transitions, which show up as a high open-play share rather than as a low set-piece share.
What we tried that didn’t work
We first tried to build a per-league, per-season set-piece share from a published Opta feed. The Premier League feed exists, and the existing Premier League set-piece post uses it. For the other four leagues, no equivalent public aggregate exists. The Athletic, Tifo, and The Coaches’ Voice all describe set-piece coaching stories at the club level, but the league-level year-by-year numbers behind those stories are not in the public domain. We could not find a single source that publishes set-piece goal share for La Liga, the Bundesliga, Serie A, or Ligue 1 across the same window.
We then tried to derive a per-league, per-season set-piece share from published per-league corner conversion rates. FBref publishes that data, but only for the seasons in its current scope, and SoccerStats247 renders it client-side. Even where the corner data exists, scaling up to a full set-piece share requires assumptions about the indirect free-kick share that the public data does not support for the four non-Premier-League leagues.
What we arrived at is the published long-run baseline. The 20% set-piece share is the best aggregate estimate of the corners + indirect free kicks + throw-in sequence share for top-five European leagues, and it is consistent with the Premier League post’s own long-run baseline. The asterisk on every derived cell is the disclosure.
Known limitations
Twenty-seven of the thirty set-piece cells in the table are derived from the long-run baseline, not measured. The disclosure is the asterisk, but the implication is that any cross-league claim in this post that depends on a per-league per-season set-piece share is a directional claim, not a measured one. The Premier League 2023-24 figure of 18.3% is the only measured deviation in the window.
Ligue 1 2019-20 is a partial season of 279 matches. Paris Saint-Germain were declared champions on a points-per-game basis after the league was halted in April 2020. The 704 goals reflect the partial schedule.
Ligue 1 moved from 20 to 18 teams in 2023-24, which removes 74 matches per season. The Bundesliga runs 18 teams across the full window, with 306 matches per season. Premier League, La Liga, and Serie A run 20 teams and 380 matches. The same 20% set-piece share applied to a 306-match season produces a smaller absolute count of set-piece goals than the same 20% applied to a 380-match season. This is a totals effect, not a share effect.
The set-piece share does not include a published break-out of throw-in sequences ending in a goal. The working definition includes them, and the 20% baseline covers corners and indirect free kicks. The throw-in share is a small wedge, an estimated 0.5 to 1.0 percentage points.
What this means for a forecast
A pre-game forecast that applies a single set-piece share across the Big Five is wrong in two directions. It under-predicts set-piece scoring in the Premier League 2025-26, where the published share is 29.3% and not 20%. It over-attributes cross-league differences to set-piece share when the public data does not support a per-league ranking. Ligue 1 and Serie A produce more set-piece goals per season than the Premier League on a count basis, but at a stable share, that is a totals story.
The right read is a league-and-season-specific set-piece coefficient. The Premier League’s 2025-26 spike is real, and the model needs a per-season override for it. The other four leagues stay on the long-run baseline until a per-league per-season Opta feed becomes public. The model now applies that override for the Premier League and a single Big Five baseline elsewhere.
For a reader following coverage across leagues, the set-piece adjustment is one of the cleanest examples of why league-specific calibration matters. The model treats the Premier League and Serie A as different competitions because the underlying goal distributions are different, and the set-piece stream is part of that difference. The same is true inside a single league across seasons: a single share applied to a 380-match season and a 306-match season produces different absolute counts, even when the share is identical.
A read is a lens for separating signal from noise. The Premier League’s 2025-26 set-piece spike is signal. The cross-league set-piece share on the public data is noise, because the public per-league per-season data does not exist. The published baseline is the honest floor. The published deviation is the only published deviation.