2025 Fall, Introduction to Football Data Analytics

If you’re new to football data analytics, we recommend starting with these foundational resources.

Soccermatics (by David Sumpter): The best introductory book on the intersection of football, mathematics, and data science. You can find related code and lecture materials on the author’s blog.
Friends of Tracking (YouTube Channel): The premier YouTube channel for learning how to handle football data, especially tracking data and event data. The Python tutorials are exceptionally detailed.

This is your essential bookmark list for tracking the latest trends and drawing inspiration from high-level analysis.

Hudl StatsBomb: The company that sets the standard for modern football data analytics. They offer insightful articles and the industry’s best free data (Open Data).
The Analyst: A fantastic source for professional analysis and stunning visualizations based on Opta data. Topics range from match predictions to historical analysis.
FBref: The most comprehensive website for statistics on football players, teams, and leagues. It’s highly reliable as it is powered by StatsBomb data.
Tifo Football (YouTube): Explains complex tactics and data analytics concepts with stylish animations. It’s the best resource for visual learners.

Karun Singh’s Blog: The blog of the analyst who pioneered advanced analytical concepts, including the Pitch Control model.
Analytics FC: A blog that breaks down advanced football analytics concepts, including Expected Goals (xG), in an accessible way.

These are the powerful tools that will elevate your data analytics journey.

Python: The ‘Libero’ of the data science world. It’s easy to learn and has a powerful ecosystem of libraries.
R: A language with particular strengths in statistical analysis and visualization.

pandas: The core library for handling dataframes. Essential for data cleaning and manipulation.
numpy: The fundamental library for numerical computation and array operations.
matplotlib & seaborn: For creating beautiful and informative data visualizations.
mplsoccer: The king of football data visualization. Makes it incredibly easy to draw pitches, pass maps, shot maps, and more.
scikit-learn: Used for applying machine learning models to predict player performance, classify team styles, and more.
statsmodels: A library for statistical modeling and hypothesis testing.

tidyverse: A collection of packages for the entire data analysis workflow, including data manipulation (dplyr) and visualization (ggplot2).
ggsoccer: A package that helps you easily create football visualizations using ggplot2.
StatsBombR: An R package that makes it easy to access and work with the StatsBomb API and open data.

Get your hands dirty with real data to build your skills!

StatsBomb Open Data: Provides incredibly detailed and extensive event data for free, including Lionel Messi’s entire Barcelona career, World Cups, and women’s football. (Highly Recommended!)
FBref: Almost all data on the website can be easily copied as a table or collected via Web Crawling.
Understat: Provides Expected Goals (xG) data for leagues, teams, and players.
Kaggle Datasets: A great place to find various football datasets shared by users worldwide, including FIFA game data and European league data.
Hudl Wyscout (Paid): A professional data platform used by top clubs. Check for free sample data or student access programs they might offer.

For those who want to delve deeper into theory and the latest research trends.

Google Scholar: Search for keywords like Football Analytics, Soccer Analytics, and Expected Goals to find the latest research papers.
“Soccermatics” (Book): The book by Professor David Sumpter, mentioned earlier, is a great resource for understanding the mathematical principles behind football analytics.
Friends of Tracking GitHub: A repository containing numerous papers, code, and tutorials related to tracking data analysis.