Following the deadline for submissions, the Opta Forum judges spent the early weeks of this year reviewing dozens of presentation proposals for the first pro research track of the 2023 event, which will be taking place next month.
The judges and question setters in each practitioner-led category scored proposals based on four key criteria: innovation, methodology, relevance and application. By the end of the process, five projects had been chosen for either presenting or exhibiting to an invited audience of industry delegates attending the event, which is being held at The Mermaid, London, on Tuesday 21st March.
The Opta Forum, which is being staged for the tenth consecutive year in 2023, remains a key date in the football analytics calendar, providing presenters with a platform for showcasing innovative approaches to key questions relating to performance analysis and recruitment. A number of previous event presenters have gone on to work in various roles across the professional football industry, as highlighted here.
For the third successive year, the event has partnered with the French Football League (LFP), who will be making historical Ligue 1 Uber Eats tracking data available to help presenters attain deeper insights into performance.
In addition to these research presentations, the Opta Forum will also be incorporating a second stage for the first time this year, which will feature a number of fan-track sessions focused around applications of data to enhance audience engagement and multi-platform media content.
The full line-up for the first pro track of the 2023 Opta Forum, listed in no specific order, is as follows:
Stage Presentations
Guillaume Hacques – Destabilising a Set Defence: Identification of Symmetry-Breaking Collective Movements
This project was chosen in the first of the Opta Forum’s two practitioner-led submission categories, set by Vignesh Jayanth of Stade Rennais, which applies tracking and event data with the objective of measuring the collective moments of an attacking team, when facing a set defence, based on how effective the collective movement is at causing shape disruption.
When a team in possession is facing an opponent well set in their defensive shape, collective movement off-the-ball is crucial in helping open-up spaces in the key areas of the pitch which the defending team is looking to protect. This is typically either between, or in behind, their defensive line.
This presentation will identify scenarios where movements by the team in possession resulted in their opponents being pulled out of their shape, as well as establish how the coordinated movement of two or more players on the attacking team caused it. The effectiveness of this collective movement will then be measured based on key attacking outcomes.
Based in Normandy, France, Guillaume possesses a PhD in in skill acquisition in sport and has previous experience as a coach and performance analyst in French football. He is currently based at the University of Rouen, where he works as a post-doctoral researcher.
Pegah Rahimian and Laszlo Toka – A Team and Player Rating Model, Based on Penetrative Pass Performance
Possessing players with the ability to complete passes that can penetrate an opponent’s defence is a valuable commodity in football, as they have the potential to take several opposition players out of the game and get the ball closer to the opposition goal. However whilst these passes have potential high rewards, they also carry a high risk of losing possession.
Pegah and Laszlo’s project proposes the creation of a new metric that measures the expected penetrative value of a pass, which can help quantify the likelihood of a potential penetrative pass resulting in goal scoring opportunities.
Once established, the metric will be applied to rank the ability of players and teams, based on their ability to complete penetrative passes, convert penetrative pass situations into goals and also determine a defensive value for players and teams, based on their ability to stop penetrative passes from occurring.
Pegah was a finalist in last year’s research-paper competition at the MIT Sloan Sports Analytics Conference and is currently completing a PhD at Budapest University of Technology and Economics. Alongside her studies, she works as a researcher at High-Speed Network (HSN Lab). Laszlo is an associate professor at Budapest University of Technology and Economics.
Tomasz Piłka, Michał Zaręba, Mateusz Tylka, Jakub Kaczmarek, Bartłomiej Grzelak and Tomasz Górecki – An Analysis of On-Field Distribution Zones for Offensive Actions in the FA Women’s Super League
In the second of the Forum’s practitioner-led submission categories, set by the FA’s Alex Thomas and Laura Seth, the chosen project will apply Opta event data from the last three FA WSL seasons to identify the zones of the pitch where key offensive actions occur, as well as highlighting stylistic differences compared to men’s competitions.
The research, led by Tomasz Piłka, will be split into five areas: offense, defence, player positions, flow of play and passing accuracy. His team will then apply their own xGBoost algorithm to attribute values to the different events, which occur in individual zones, across each area to help identify which events are most significant.
The research group comprises academics and students from Adam Mickiewicz University in Poznań, Poland. Tomasz Górecki is a professor, and Tomasz Piłka an assistant professor, at the university’s Faculty of Mathematics and Computer Science. Both are involved in research in artificial intelligence, machine learning, and their practical applications.
Michal is a PhD student, while Bartłomiej, Jakub and Mateusz are completing their Master’s degrees in Data Analysis and Computer Science. In addition to their roles at the univeristy, Tomasz Piłka and Bartłomiej also conduct work on behalf of Ekstraklasa club Lech Poznan.
Poster Exhibitors
Javier M. Buldú and Álvaro Novillo – Introducing Offside Control: a Pitch Control Parameter to Evaluate Offside Performance
Javier and Álvaro’s poster will display a new methodology which aims to quantify the performance of a team’s defensive offside line.
By analysing how defensive and attacking players interact with an offside line, they will look to define a new parameter which quantifies the level of risk a defending team allows with the positioning of their line, and the potential danger which the attacking team can exploit. In doing so, they can identify which players and teams are effective at generating risk from beating offside lines and which defenders are effective at reducing danger.
As well as taking into account the field coordinates of the offside line, contextual aspects of matches, including the game state and time remaining, will also be taken into account within the methodology.
Possessing a PhD in Applied Physics, Javier is the coordinator of the Complex Systems Group at the King Juan Carlos University in Madrid, as well as being the Principal Investigator of the Laboratory of Biological Networks at the Center for Biomedical Technology. Álvaro has an MSc in Data Science and is also based at King Juan Carlos University, where he conducts analytical research using event and tracking datasets.
Henrick Biermann, Franz-Georg Wieland and Weiran Yang – Extension of xCounter through continuous metric and new comprehensible features
After introducing the concept of Expected Counter (xCounter) in a presentation at the Opta Forum last year, Henrick, Franz-Georg and Weiran will be returning to the Opta Forum in 2023 to exhibit a poster which builds on their existing work by introducing new metrics, including a continuous metric in relation to predicting the success of counter attacks, as well as counter-aware marking by opponents.
By incorporating these new metrics into their existing methodology, the poster will illustrate how teams can apply xCounter to develop more robust strategies to either prevent or facilitate counter attacks. They will also aim to exhibit the outcomes of a player model, which ranks players individually based on their propensity to lose the ball in dangerous situations.
Henrick and Weiran are both based at the German Sport University in Cologne, working as Research Assistants in the Institute of Training and Computer Science. Franz-Georg is a Doctoral student and research associate at the University of Freiburg.
Click here to watch their original presentation from last year’s event.
Stats Perform would like to thank everyone who submitted a proposal in track one of the competition, as well as the panel of judges and question setters, and congratulate the five people who will be presenting or exhibiting at the 2023 Opta Forum.
The remaining presenters and exhibitors, who entered the Opta Vision track of the research competition, will be announced next month.
