A Goal Scoring Probability Model for Shots Based on Synchronized Positional and Event Data in Football (Soccer)
Due to the low scoring nature of football (soccer), shots are often used as a proxy to evaluate team and player performances. However, not all shots are created equally and their quality differs significantly depending on the situation. The aim of this study is to objectively quantify the quality of...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-03-01
|
| Series: | Frontiers in Sports and Active Living |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fspor.2021.624475/full |
| _version_ | 1818603389894787072 |
|---|---|
| author | Gabriel Anzer Gabriel Anzer Pascal Bauer Pascal Bauer |
| author_facet | Gabriel Anzer Gabriel Anzer Pascal Bauer Pascal Bauer |
| author_sort | Gabriel Anzer |
| collection | DOAJ |
| description | Due to the low scoring nature of football (soccer), shots are often used as a proxy to evaluate team and player performances. However, not all shots are created equally and their quality differs significantly depending on the situation. The aim of this study is to objectively quantify the quality of any given shot by introducing a so-called expected goals (xG) model. This model is validated statistically and with professional match analysts. The best performing model uses an extreme gradient boosting algorithm and is based on hand-crafted features from synchronized positional and event data of 105, 627 shots in the German Bundesliga. With a ranked probability score (RPS) of 0.197, it is more accurate than any previously published expected goals model. This approach allows us to assess team and player performances far more accurately than is possible with traditional metrics by focusing on process rather than results. |
| first_indexed | 2024-12-16T13:22:24Z |
| format | Article |
| id | doaj.art-482c55c6b1304283bc854b1c81fe0295 |
| institution | Directory Open Access Journal |
| issn | 2624-9367 |
| language | English |
| last_indexed | 2024-12-16T13:22:24Z |
| publishDate | 2021-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Sports and Active Living |
| spelling | doaj.art-482c55c6b1304283bc854b1c81fe02952022-12-21T22:30:19ZengFrontiers Media S.A.Frontiers in Sports and Active Living2624-93672021-03-01310.3389/fspor.2021.624475624475A Goal Scoring Probability Model for Shots Based on Synchronized Positional and Event Data in Football (Soccer)Gabriel Anzer0Gabriel Anzer1Pascal Bauer2Pascal Bauer3Sportec Solutions AG, Subsidiary of the Deutsche Fußball Liga (DFL), Munich, GermanyInstitute of Sports Science, University of Tübingen, Tübingen, GermanyInstitute of Sports Science, University of Tübingen, Tübingen, GermanyDFB-Akademie, Deutscher Fußball-Bund e.V., Frankfurt am Main, GermanyDue to the low scoring nature of football (soccer), shots are often used as a proxy to evaluate team and player performances. However, not all shots are created equally and their quality differs significantly depending on the situation. The aim of this study is to objectively quantify the quality of any given shot by introducing a so-called expected goals (xG) model. This model is validated statistically and with professional match analysts. The best performing model uses an extreme gradient boosting algorithm and is based on hand-crafted features from synchronized positional and event data of 105, 627 shots in the German Bundesliga. With a ranked probability score (RPS) of 0.197, it is more accurate than any previously published expected goals model. This approach allows us to assess team and player performances far more accurately than is possible with traditional metrics by focusing on process rather than results.https://www.frontiersin.org/articles/10.3389/fspor.2021.624475/fullexpected goalsXGpositional dataevent dataapplied machine learningfootball |
| spellingShingle | Gabriel Anzer Gabriel Anzer Pascal Bauer Pascal Bauer A Goal Scoring Probability Model for Shots Based on Synchronized Positional and Event Data in Football (Soccer) Frontiers in Sports and Active Living expected goals XG positional data event data applied machine learning football |
| title | A Goal Scoring Probability Model for Shots Based on Synchronized Positional and Event Data in Football (Soccer) |
| title_full | A Goal Scoring Probability Model for Shots Based on Synchronized Positional and Event Data in Football (Soccer) |
| title_fullStr | A Goal Scoring Probability Model for Shots Based on Synchronized Positional and Event Data in Football (Soccer) |
| title_full_unstemmed | A Goal Scoring Probability Model for Shots Based on Synchronized Positional and Event Data in Football (Soccer) |
| title_short | A Goal Scoring Probability Model for Shots Based on Synchronized Positional and Event Data in Football (Soccer) |
| title_sort | goal scoring probability model for shots based on synchronized positional and event data in football soccer |
| topic | expected goals XG positional data event data applied machine learning football |
| url | https://www.frontiersin.org/articles/10.3389/fspor.2021.624475/full |
| work_keys_str_mv | AT gabrielanzer agoalscoringprobabilitymodelforshotsbasedonsynchronizedpositionalandeventdatainfootballsoccer AT gabrielanzer agoalscoringprobabilitymodelforshotsbasedonsynchronizedpositionalandeventdatainfootballsoccer AT pascalbauer agoalscoringprobabilitymodelforshotsbasedonsynchronizedpositionalandeventdatainfootballsoccer AT pascalbauer agoalscoringprobabilitymodelforshotsbasedonsynchronizedpositionalandeventdatainfootballsoccer AT gabrielanzer goalscoringprobabilitymodelforshotsbasedonsynchronizedpositionalandeventdatainfootballsoccer AT gabrielanzer goalscoringprobabilitymodelforshotsbasedonsynchronizedpositionalandeventdatainfootballsoccer AT pascalbauer goalscoringprobabilitymodelforshotsbasedonsynchronizedpositionalandeventdatainfootballsoccer AT pascalbauer goalscoringprobabilitymodelforshotsbasedonsynchronizedpositionalandeventdatainfootballsoccer |