Skip to content
View matteovezzoli's full-sized avatar
2 followers · 7 following

Block or report matteovezzoli

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
matteovezzoli/README.md

Hi, I'm Matteo 👋

I'm a Sports Data Analyst and Scientist based in Milan, Italy. I specialize in turning complex spatial, event, and tracking data into actionable tactical insights and predictive models. With a strong academic background in statistical modeling, my primary focus is pioneering advanced football analytics.

🚀 What I'm Working On

  • Contextual Football Scouting (In Progress): A data-driven scouting platform designed to mitigate "Team Bias" by isolating individual talent from team tactical structures. Utilizing UEFA Euro 2024 StatsBomb 360° and Transfermarkt data, the project implements spatial metrics via Convex Hull, Expected Possession Value (EPV) to spot line-breakers, decision-quality evaluation under pressure, off-ball movement tracking, and a within-role similarity model ("Spatial DNA").
  • Serie A CB Scouting Engine: An interactive Streamlit dashboard built on PCA and clustering algorithms to categorize and profile Italian top-flight center-backs.
  • Technical Scouting & Match Analysis (Girona FC Recruitment Pipeline): Developed professional scouting deliverables and tactical assessments tailored to the specific recruitment data standards of Girona FC's analytical department.
  • Expected Goals (xG) Pipeline: End-to-end development and calibration of xG models using Logistic Regression, Random Forests, and XGBoost.

🎓 Education & Background

  • Postgraduate in Sports Analytics | Barça Innovation Hub & Universitat Central de Catalunya (2026 - Present)
    • Focusing on processing open, event, and tracking data to solve specific tactical and performance problems using Python and SQL.
  • M.Sc. in Statistics, Business Analytics | University of Bologna
    • Graduated 110/110 cum laude.
    • Master's Thesis focused on multivariate time-series forecasting benchmarks (VAR, Random Forest, XGBoost, LightGBM, CatBoost, SVM, LSTM), which later led to a research collaboration and a co-authored scientific paper.
  • B.Sc. in Statistical and Economic Sciences | University of Milano-Bicocca
    • Thesis project: Predictive modeling for NBA player salaries.

💼 Professional Experience & Scouting Tasks

  • University of Bologna (Research Fellow): Managed the full data pipeline—from curation and machine learning modeling to final software implementation and visualization—for a scientific publication.
  • Nomisma (Market & Data Analyst): Designed and deployed a predictive model in R that was officially presented at Vinitaly 2025.

🛠️ Tech Stack & Skills

  • Languages: Python (Pandas, NumPy, Scikit-learn, XGBoost), R, SQL (MySQL), SAS, MongoDB
  • Analytics: Time Series Forecasting, Predictive Modeling, Machine Learning (Regression & Classification), Clustering, Web Scraping
  • Tools: Git/GitHub, Power BI, Excel, Streamlit

📫 Connect with me

Pinned Loading

  1. SerieA-DefensiveScouting-Engine SerieA-DefensiveScouting-Engine Public

    A complete Data Science pipeline (Jupyter Notebook) and interactive Streamlit web app for tactical football scouting of Serie A defenders using PCA and K-Means clustering.

    Jupyter Notebook 1

  2. ArMat-Analytics/Contextual-Football-Scouting ArMat-Analytics/Contextual-Football-Scouting Public

    A data-driven football scouting platform mitigating Team Bias. It leverages 360° spatial data, EPV grids to evaluate a player's true decision-making under pressure.

    Jupyter Notebook 2