Create

Design Project Gallery

Project Search
Search projects by keyword, program, course, or submission year.

Soccer Defender Analytics – An analysis of Imbalance in Soccer Player Evaluations

Project Description:

In this project, we investigated the undervaluation of soccer defenders in player rating systems by leveraging data analytics and machine learning techniques. Using player performance data from the top five European leagues across multiple seasons, we built predictive models to estimate player ratings based on defensive-specific metrics. Our analysis revealed a systemic bias that favored attacking players, often overlooking defensive contributions. To address this, we developed an adjusted rating system that reweights defensive actions such as tackles, interceptions, and blocks to better reflect a defender’s true impact on the game. Visualizations including bar charts, box plots, and scatter plots helped communicate these findings, showing how defenders’ rankings improved after adjustment. The project not only highlighted the limitations of traditional player evaluations but also provided a data-driven solution to promote fairer recognition of defensive talent in professional soccer. This work offers valuable insights for clubs, analysts, and fans alike.

Project Photo:

The image shows a defender (wearing a red kit) performing a slide tackle on a forward (wearing a white kit) in a UEFA Champions League final. This highlights the role of defenders in high-stakes matches, underscoring their importance in both stopping attacks and contributing to their team's overall performance.

Defender executing a crucial slide tackle on a forward during a UEFA Champions League final, exemplifying the vital role of defenders in high-pressure matches.

Project Poster

Open full size poster in new tab (PDF)

Project Poster Summary:

The project “Soccer Defender Analytics” focuses on addressing the imbalance in soccer player evaluations, where defenders often receive less recognition than attackers. The team developed a data-driven rating system using machine learning models trained on over 40 data features from around 6,000 professional players. This system predicts and adjusts defender ratings, highlighting defensive outliers and offering a more equitable approach to player evaluations. The project aims to provide a fairer, analytical framework for recognizing defensive excellence in soccer.

Course Faculty

Project Mentors, Sponsors, and Partners