Following the launch of TransferLab v3, Analytics FC are pleased to announce another new data model which will feature in the platform. Introducing ClubStrength+, a brand new method for adjusting player performance output based on club strength, replacing our previous, league-based model.
This game-changing approach brings a new level of accuracy and sophistication to assessing relative player performance. It will change the way football performances are measured, providing richer context and enabling our clients to gain a better understanding of a player’s true abilities.
How Does It Work?
Many of our services, including our flagship TransferLab data-scouting platform model, use our powerful on-pitch algorithm to evaluate the contribution of each player in a match.
This approach, known as Goal Difference Added (GDA), enables Analytics FC to quantify each player’s offensive and defensive contributions, both positive and negative, over the course of a 90-minute match. By evaluating the balance between positive and negative contributions, we can determine a player’s overall impact on their team’s probability of scoring or conceding and thus their overall goal difference.
Previously, we used a model which calculated an ‘Exchange Rate’ to adjust player output based on the quality of competition they were playing in, calculating how that output would translate from league to league. However, in consultation with our clients and industry specialists, we have now launched a new model that calculates the global ranking for each club and adjusts the output based on the strength of clubs, not leagues.
Key Solutions
This new system solves a number of issues that have been problematic in previous models:
Domestic domination: The model facilitates better measurement of players playing for dominant domestic teams, and especially those who also play in continental competitions. This means, for example, a more sophisticated method for distinguishing the performances of a player like Kylian Mbappe against Lorient or against Real Madrid.
Inter-league sample size: The new method also addresses inter-league issues that arise with models like ELO, which lack sample size due to the lack of inter-league play between teams on different continents. Analytics FC’s ranking system interweaves all leagues across continents together, enabling a much fairer comparison.
Intra-league differences: It also addresses intra-league differences, such as a Celtic/Rangers scenario in Scotland, where it can be difficult to rank an entire league when a small number of clubs are significantly better than the league’s mean or median.
Mid-Season Split (Championship Round vs Relegation Round): Furthermore, ClubStrength+ eliminates the issue of the league split that occurs in the second half of the season in many countries, which can skew data and mislead scouts about players who have only performed well in 2nd half of the season in the ‘relegation’ group.
Bullies: Finally, the new model also addresses the issue of bully-boy players who inflate their numbers by destroying bottom half teams. ClubStrength+ works at match event level so every action is adjusted for the strength of the opposition club being played.
A Leap Forward
Analytics FC’s new method of adjusting player performance output based on the level at which a player is playing is set to revolutionise the way football performances are measured. This new approach provides a greater level of accuracy and context, enabling executives, coaches, scouts, and fans to gain a better understanding of a player’s true abilities. By using this new model, the company has addressed a number of issues that have plagued previous models, enabling a fairer comparison across leagues and continents.
This is a significant step forward in the world of football analytics, and it will be interesting to see how it is adopted by teams, coaches, and fans around the world.
If you want to learn more about our TransferLab platform contact us here, or book a free trial of the by clicking here.
Header image copyright IMAGO / Andy Rowland / PMI
