Pedro Satorre-Mulet

My MSc. Data Science thesis was supervised by Dr. Gabriel Anzer – Head of Football Data Analytics @ RB Leipzig:

“Shot Classification & Goal Probability Estimation Using Graph Neural Networks” (specifically, Heterogeneous Graph Transformers – HGTs)

Cutting-edge, state-of-the-art MSc. Data Science thesis’ research project – successfully demonstrated statistically that HGTs classify, estimate, and calibrate better football shots & their respective goal probability (i.e. xG), compared to conventional xG models.

Dr. Gabriel Anzer’s Recommendation Letter

This project was also selected by David Sumpter (and his committee) as 1 of the 4 finalists for Twelve Football’s  “Pitch to the Pros #3”.

In parallel to my thesis project mentioned above, I also implemented:

• Spearman’s Pitch Control surfaces (both in static .png and in animated/.mp4 video formats) to smoothly visualize the areas of the pitch that are under control by each of the teams and which areas would be under dispute for the ball. 
• Expected Possession Value-Added surfaces + contours (both in static .png and in animated/.mp4 video formats) to smoothly visualize which areas of the pitch the ball should be displaced to, at every instant in time, by the team in possession, to maximize their xG within that same play.
  • I’m very happy to provide live demos of any of these.

I got selected into Twelve Football’s 1st edition of the course (AI) “Context Engineering for Football” – we had to implement a hands-on project as part of the course, to develop an AI Analyst agent; I implemented a CBs & CB-Pairings AI Analyst agent chat on a Streamlit app.

• once again, I’m very happy to provide a live demo of this project. 

I enjoy teamwork and promoting a camaraderie spirit among colleagues. I also enjoy myself when given autonomy and freedom to propose and implement my initiatives.