Soccer Analytics with Machine Learning (Haipeng Gao, Ari Joury, Weining Shen etc.)（Z-Library）

# Soccer Analytics with Machine Learning — Reading Guide ## 【One-Line Pitch】 A practical, soccer-driven introduction to machine learning and predictive modeling, this book teaches data science through the world's most popular sport—perfect for soccer fans curious about analytics and students or professionals who want to learn ML through real, messy problems rather than abstract exercises. ## 【Book Arc】 - **Opening (~0%–9%)**: The book opens with endorsements, preface, and audience framing—establishing why soccer is an ideal domain for learning machine learning, positioned during the 2026 FIFA World Cup to capitalize on global interest in the game. - **Early (~9%–25%)**: Front matter continues with conventions, code usage policies, and acknowledgments, then transitions into the core motivation: why soccer analytics matters and how the book will bridge methodological rigor with hands-on practice. - **Early (~25%–34%)**: The book introduces the soccer analytics landscape—the unique challenges of a low-scoring, high-variance sport, the types of questions analytics answers, and the decision contexts across clubs, from tactical preparation to player workload management. - **Middle (~38%–47%)**: Coverage expands to betting markets and the data-driven evolution of soccer—how bookmakers model uncertainty with Poisson goal models and machine learning, and how analytics shifted from outcome-based evaluation to process-based analysis. - **Middle (~47%–53%)**: The book explores the data revolution—event data (passes, shots, tackles with timestamps and coordinates) and tracking data (player and ball positions captured multiple times per second)—and introduces metrics like PPDA (passes per defensive action) for measuring style and defensive behavior. ## 【Key Takeaways】 - **Soccer is an ideal ML learning domain** (Early): The sport's low-scoring, high-variance nature creates rich, imperfect data that forces disciplined statistical thinking—perfect for learning how to separate signal from noise in real-world modeling. - **Analytics serves real decisions, not just curiosity** (Early): From tactical match preparation to substitution timing and pressing intensity evaluation, analytics supports coaching staffs with evidence-based insights across professional clubs. - **Player workload and injury risk are quantifiable** (Early): GPS tracking, heart-rate monitors, and training metrics allow analysts to answer questions about cumulative training load, match congestion effects on sprint output, and recovery patterns. - **Betting markets depend on calibrated probabilities** (Middle): Bookmakers like bet365 and DraftKings use statistical models—often building on classical Poisson goal models—to estimate match outcome probabilities, balancing accuracy with market dynamics and risk exposure. - **Process beats outcomes for evaluation** (Middle): Traditional league tables treat a 1–0 win against a title contender the same as one against a relegation candidate; modern analytics examines chance creation, defensive pressure, and possession movement instead. - **Event data and tracking data transformed the field** (Middle): Event logs record on-ball actions with timestamps and coordinates, while optical tracking systems capture player and ball positions multiple times per second—enabling analysis of off-ball movement and team shape. - **Metrics like PPDA quantify playing style** (Middle): Passes per defensive action measures how many opposition passes a team allows before engaging defensively, providing a concrete way to describe and compare defensive behavior. ## 【Reading Tips】 - **Skim the front matter** (~0%–19%): The preface, conventions, and acknowledgments are standard O'Reilly material—skip ahead unless you need code usage policies or contact information. - **Deep-read the analytics landscape chapter** (~25%–34%): This is where the book establishes its conceptual foundation—the unique challenges of soccer data and the decision contexts that motivate the rest of the book. - **Pay attention to the betting markets section** (~38%–44%): This is a practical, concrete application of statistical modeling that bridges classical methods (Poisson models) with modern machine learning approaches. - **Focus on the data evolution discussion** (~44%–53%): Understanding the difference between event data and tracking data—and metrics like PPDA—is essential for following the modeling chapters that likely come later. - **Note what the book explicitly excludes**: The authors state they deliberately focus on predictive modeling rather than covering every aspect of soccer analytics, so don't expect comprehensive coverage of business applications or advanced tracking-data methods. ## 【Coverage Limits】 This guide is based on excerpts covering roughly the first half of the book (approximately 0–53%). The later chapters—which likely contain the actual machine learning models, code examples, and hands-on exercises—are not covered in this sample. ##