Smart-Recruit
Using Machine Learning to recruit footballers
Every club needs a striker. For a striker to excel, his preferred playstyle and physical attributes must match the club tactics. Using data available on StatsBomb, I created a model that can sugest which strikers best fit the team according to chosen playstyle.
Client
RV College of Engineering
Services
Machine Learning
Industries
Sports
Date
May 2023
Researching the topic involved a thorough exploration of football analytics, specifically focusing on striker performance metrics. I collected and analyzed data from StatsBomb, identifying key playstyle indicators and physical attributes crucial for strikers. The integration of this model aids clubs in recruiting strikers aligned with their tactical preferences, optimizing team performance and strategy.
The final outcome demonstrates the successful implementation of a robust xG model using Gradient Boosting, providing a reliable tool for clubs to assess and recruit strikers tailored to their playstyle. The high accuracy achieved (92%) underscores the effectiveness of the approach, contributing valuable insights to enhance decision-making in football player recruitment.