Sports Analytics in Practice with R. Ted Kwartler

186

      195  187

      196  188

      197  189

      198 190

      199  191

      200  192

      201  193

      202  194

      203  195

      204  196

      205  197

      206  198

      207  199

      208  200

      209  201

      210  202

      211  203

      212  204

      213  205

      214  206

      215  207

      216  208

      217  209

      218  210

      219  211

      220  212

      221  213

      222  214

      223  215

      224  216

      225  217

      226  218

      227  219

      228  220

      229 221

      230  222

      231  223

      232  224

      233  225

      234 226

      235  227

      236  228

      237  229

      238  230

      239  231

      240  232

      241  233

      242  234

      243  235

      244  236

      245  237

      246  238

      247  239

      248  241

      249  242

      250  243

      251  244

      252  245

      253  246

      254  247

      255  248

      256  249

      257  250

      258 251

      259  253

      260 254

Preface

      Sports is one of the few places where the data and outcomes are well known. Unlike medicine which requires significant subject-matter expertise or business where the data is proprietary in most cases, sports knowledge is relatively accessible, and the data and outcomes are public. As a result, sports analytics serves as a great entry point for many aspiring data scientists and analytics professionals. For the novice, this book demonstrates the many facets and uses of countless techniques applicable outside of sports. It should have more than enough topics and examples to aid learning for general practice. For the avid R programmer and sports fan, the book likely has some new functions and techniques which may be less well known. These readers will delight in improving and expanding the demonstrated methods once the core concepts are understood. Finally, for those already in the sports analytics world the techniques and individual chapter topics can serve as a reference and starting point in their professional analysis. For instance, much of the use cases in the chapters can be adjusted to specific sports or updated by more recent underlying data.

      This book has been a long journey in the making. Originally the book’s scope was centered on individualized chapters demonstrating analytical techniques within a sports context. The goal is that a reader inherits various tools that act as a foundation for analysis to build upon and add complexity with subsequent analyses as the reader’s technical acumen and sports interests grow. Each chapter is meant to be a standalone reference as the reader explores and learns. This also frees up the reader to focus on topics of interest. For example, a reader may not want to learn about natural language processing so could skip that chapter altogether to focus on another subject such as optimizing a fantasy football lineup. The book’s undertaking grew in complexity due to a personal commitment to demonstrate concepts on diverse data sets including Paralympic athletes, female soccer and basketball, and less US-centric popular sports including cricket in addition to the more typically demonstrated sports analyses of men’s football, baseball, and basketball. My goal is to make the subject accessible and relevant to many in the analytics field despite this effort slowing the book’s creation. Keep in mind a chapter’s concepts can be applied to many sports domains. For example, the text analysis applied to cricket fan forum posts can easily be applied to men’s basketball fan tweets or forum posts. Each chapter’s takeaway is meant to be a broadly useful tool, not a brittle or narrowly focused analysis. Additionally, the book was delayed due to the pandemic’s effect on the sports-world. Admittedly the shortened seasons, canceled games, and other changes that created outlier statistics pales in comparison to the pandemic’s hardship and humanistic impact outside of sports. Despite these challenges, the book’s end result was worth the delay. The final product covers many diverse concepts, and data, encouraging analytics professionals to enjoy the intersection of sports and analysis.

The book’s supporting website is www.rstatsbook.com. The site contains data and scripts along with any code revisions necessary as functions and packages change. Redundantly, data is shared via git repository at www.github.com/kwartler/Practical_Sports_Analytics.

Author Biography

       Ted Kwartler

      Adjunct Professor, Harvard University

Ted Kwartler is the VP, Trusted AI at DataRobot. At DataRobot, Ted sets product strategy for explainable and ethical uses of data technology in the company’s application. Ted brings unique insights and experience utilizing data, business acumen, and ethics to his current and previous positions at Liberty Mutual Insurance and Amazon. In addition to having four DataCamp courses, he teaches graduate courses at the Harvard Extension School and is the author of Text Mining in Practice with R.

      Analytics