With this novel method, SciSports can identify the most effective football players and the best possible line-up for the 2017/2018 season. ‘If you can assign a value to each action of each player on the pitch, it will help you decide if the player is the right fit for the team. It is a great source of information if a professional football club is considering whether to sign a player,’ explains Jan van Haaren, the Head of Data & Analytics at SciSports. ‘Most metrics only focus on rare events like shots and goals alone, but our metric addresses all actions based on their impact on the match outcome.’
The dream team of 2016/2017
Using the metric, the company is therefore able to name, for example, the most effective players during the past and the ongoing football season.
Based on analyzing the five main European leagues (English Premier League, Spanish Primera Division, German 1. Bundesliga, Italian Serie A, and the French Ligue 1), SciSports identified the ‘Team of the 2016/2017 Season’. Their results reveal Lionel Messi, Kylian Mbappé, and Eden Hazard as the most effective players during the season and the best possible line-up for 2016/2017 according to SciSports’ metric is as follows:
The offensive line features Eden Hazard, Lionel Messi and Kylian Mbappé. The midfield consists of Kevin De Bruyne, Isco, and Cesc Fàbregas. The defensive line includes Vlad Chirices, Leonardo Bonucci, Markus Suttner and Lukasz Piszczek and the goalkeeper should be Jordan Pickford.
The best line-up of this season
In the paper, SciSports also presents the best possible line-up for the 2017/2018 season. Curious? Here it is:
MIT Sloan Sports Analytics Conference
The paper, submitted on behalf of SciSports to the MIT Sloan Sports Analytics Conference, was written by Jan van Haaren (SciSports), Lotte Bransen (SciSports), Tom Decroos (KU Leuven) and Jesse Davis (KU Leuven). Although accepted for the MIT conference, it wasn’t selected as one of the top six papers in its category and therefore cannot be presented during the event. SciSports plans to make the full paper public in the foreseeable future.
