Football Analytics Course

Lesson 1 • Key Performance Metrics Overview

Introduces foundational metrics such as shots, passes, and possession before progressing to derived metrics. Connects raw counts to meaningful performance interpretation.

Lesson 2 • The Analytics Landscape in Football

Introduces the history and evolution of data use in football, from basic statistics to advanced modelling. Establishes why analytics matters for clubs, coaches, and scouts.

Lesson 3 • The Football Analytics Workflow

Outlines the end-to-end process from data collection to insight delivery. Students map each workflow stage to real club or media use cases.

Lesson 1 • Database Design for Football Data

Introduces relational and document-based database schemas tailored to football event and tracking data. Students design schemas that support efficient querying.

Lesson 2 • Data Governance and Ethics

Addresses player data privacy, consent frameworks, and responsible data use in professional football. Students apply ethical principles to data handling decisions.

Lesson 3 • Data Providers and Sources

Surveys commercial and open-source football data providers, covering coverage scope and licensing considerations. Students evaluate sources for fitness to specific analytical tasks.

Lesson 4 • Data Collection Methods

Covers manual tagging, semi-automated optical tracking, and wearable sensor collection. Students understand trade-offs in cost, accuracy, and latency for each method.

Lesson 1 • Hypothesis Testing for Football Data

Applies t-tests, chi-square tests, and non-parametric alternatives to football research questions. Students distinguish statistically significant findings from noise.

Lesson 2 • Descriptive Statistics in Football

Covers measures of central tendency, spread, and distribution shape applied to match and player data. Students summarise datasets and identify distributional patterns.

Lesson 3 • Probability and Uncertainty

Introduces probability theory as the basis for modelling uncertain football outcomes. Students calculate event probabilities and understand randomness in match results.

Lesson 4 • Correlation and Regression Basics

Teaches linear correlation and simple regression to identify relationships between football metrics. Students interpret coefficients and assess model fit.

Lesson 1 • Feature Engineering for Shot Models

Identifies and constructs shot-level features including distance, angle, body part, and assist type. Students engineer a feature set ready for model training.

Lesson 2 • Extending xG to xA and Post-Shot Models

Introduces expected assists (xA) and post-shot xG that incorporates shot placement. Students extend the xG framework to evaluate chance creation and finishing skill.

Lesson 3 • Building and Training xG Models

Walks through logistic regression and gradient boosting approaches to xG model construction. Students train models on historical shot data and tune hyperparameters.

Lesson 4 • Conceptual Foundations of xG

Explains the logic of assigning goal probability to individual shots based on situational features. Students articulate xG's value over raw shot counts.

Lesson 1 • Introduction to Passing Networks

Defines nodes, edges, and weights in the context of football passing networks. Students construct basic passing networks from event data.

Lesson 2 • Pressure and Defensive Spatial Analysis

Quantifies pressing intensity, PPDA, and defensive line height using event and tracking data. Students connect spatial defensive metrics to tactical outcomes.

Lesson 3 • Progressive Passing and Ball Progression

Defines progressive passes and carries, then measures how teams advance the ball through zones. Students identify progression patterns and bottlenecks.

Lesson 4 • Spatial Zones and Pitch Control

Introduces pitch zoning frameworks and physical pitch control models derived from tracking data. Students quantify territorial dominance and space creation.

Lesson 1 • Transfer Market Valuation Models

Builds regression and machine learning models to estimate player market values from performance data. Students assess model accuracy and apply valuations to recruitment budgets.

Lesson 2 • Adjusting for Context and Competition Level

Teaches league difficulty adjustments, minutes-played normalisation, and age-curve corrections. Students produce context-adjusted metrics for fair cross-league comparison.

Lesson 3 • Scouting Report Integration

Combines quantitative player ratings with qualitative scouting observations into unified reports. Students structure reports that support recruitment decision-making.

Lesson 4 • Similarity Scores and Player Profiling

Applies distance metrics and clustering to find statistically similar players across leagues. Students build player profiles and identify comparable transfer targets.

Lesson 1 • Opponent Profiling and Game Planning

Synthesises multiple analytical outputs into a structured opponent profile for pre-match preparation. Students produce a complete game plan report supported by data.

Lesson 2 • Transition Analysis

Measures attacking and defensive transition speed, structure, and effectiveness using event data. Students identify transition vulnerabilities and strengths in opponent teams.

Lesson 3 • Set Piece Analytics

Quantifies corner, free kick, and throw-in effectiveness using spatial and outcome data. Students design data-informed set piece routines and defensive schemes.

Lesson 4 • Formation and Shape Detection

Uses average position data and clustering to detect team formations and shape transitions. Students automate formation identification from tracking data.

Lesson 1 • Unsupervised Learning and Clustering

Uses k-means, hierarchical clustering, and dimensionality reduction to discover patterns in football data. Students segment players, teams, and playing styles.

Lesson 2 • Supervised Learning for Football Outcomes

Applies random forests, gradient boosting, and support vector machines to match and player outcome prediction. Students tune and evaluate classifiers on football datasets.

Lesson 3 • Model Deployment and Monitoring

Covers containerisation, API serving, and performance monitoring for production football models. Students deploy a model and set up drift detection pipelines.

Lesson 4 • Deep Learning for Tracking Data

Applies convolutional and graph neural networks to spatiotemporal tracking data. Students extract tactical patterns from raw positional streams.