Advanced Basketball Data Science

Advanced Basketball Data Science: With Applications in R is the essential next step for anyone looking to push basketball analytics beyond standard metrics. Expanding on the foundation of Basketball Data Science (2020), this book takes readers into the fast-evolving world of advanced statistical modeling, machine learning, and modern computational techniques applied to the game. From lineup optimization and clutch-performance analysis to player tracking, pose estimation, and ball-trajectory modeling, the book shows how cutting-edge data can reveal the hidden patterns that shape decision-making on and off the court. Readers learn not only what to analyze, but how to build robust, reproducible workflows using real data, fully executable R code, and a structured project environment. Designed for analysts, coaches, researchers, and graduate students, this volume translates complex concepts into actionable tools that can immediately elevate scouting, strategy, and performance evaluation. Whether you aim to understand spatial tendencies, quantify player impact, or model scoring probabilities with machine learning, this book provides the framework to do so with clarity and confidence. Advanced Basketball Data Science is where rigorous methodology meets practical basketball insight, an indispensable resource for anyone committed to understanding the game through the power of data. . Combines advanced statistical methods, machine learning, and computer-vision techniques to provide a unified and cutting-edge framework for basketball analytics. . Offers fully reproducible workflows - complete with datasets, R code, and additional functions - which enable readers to directly apply and extend all analyses. . Integrates real-world case studies from diverse data sources (play-by-play, tracking, pose estimation, ball trajectories) to demonstrate how rigorous methodology translates into actionable basketball insights.

Paola Zuccolotto is a Full Professor of Statistics at the University of Brescia, with nearly 30 years of experience in scientific research in the field of Statistical Sciences. Her research focuses on multivariate data analysis, data mining, prediction using statistical models and machine learning algorithms, and time series analysis. Her work spans a wide array of applications, with particular emphasis on finance, social sciences, genetics, and sports, especially basketball analytics. She is the Director of the Big&Open Data Innovation Laboratory (BODaI-Lab, bodai.unibs.it) and has established extensive collaborations with international research teams. In 2016, together with Marica Manisera, she co-founded the international network Big Data Analytics in Sports (BDsports, bdsports.unibs.it). She is also a co-author of the book Basketball Data Science: With Applications in R (CRC Press). She has published numerous scientific articles in international journals and books and has been an active participant in several international conferences, including as a keynote speaker. Marica Manisera is an Associate Professor of Statistics at the University of Brescia and has almost 25 years of experience in scientific research in the area of Statistical Sciences. Her main research interests include multivariate analysis, data mining techniques, and predictive modeling based on statistical approaches and machine learning methods. Her research activity covers different domains, with a strong focus on social sciences and sports, particularly basketball analytics. She currently serves as Chair of the Special Interest Group on Sport Statistics of the International Statistical Institute and as Associate Editor of the Journal of Sports Analytics, and she has developed long-standing collaborations with international research groups. In 2016, together with Paola Zuccolotto, she co-founded the international network Big Data Analytics in Sports (BDsports, bdsports.unibs.it). She is also co-author of the book Basketball Data Science: With Applications in R (CRC Press). Her publication record includes numerous articles in international journals and book chapters, and she regularly participates in international scientific conferences. Marco Sandri is a freelance statistician with nearly 30 years of experience in scientific research. His work spans time series analysis, machine learning, biostatistics, and sports analytics. He has collaborated extensively with international research teams in medicine and biology, with an emphasis on the modeling and prediction of complex phenomena. In parallel, he has conducted long-standing research in sports data science, analyzing performance determinants across several sports, with a particular emphasis on basketball analytics. He is a co-developer of the BasketballAnalyzeR R package and previously authored a chapter in Basketball Data Science: With Applications in R (CRC Press). He has taught courses on statistics, machine learning, and data analysis, and has published over 300 peer-reviewed articles in international journals.

Foreword Preface 1 Getting started: overview and supporting materials I Analyzing and comparing game splits 2 Beyond individual skills 3 Drilling down on clutch splits: measuring performance when it matters most 4 The race to the finish: exploring the relationship between season segments and final rankings II Decoding motion 5 Understanding players' spatial dynamics 6 Athletic motion kinematics analysis 7 Tracking and analyzing ball trajectories III Spatial performance analysis 8 Basketball performance maps based on court segmentation 9 Scoring probability maps via Machine Learning algorithms Bibliography R packages Index