[Basketball Prospectus received a free copy of this book to review.]
For years, Wayne Winston flew under the radar. In NBA circles, Winston was best known for being the co-creator of WINVAL, the innovative first stab at adjusted plus-minus in the NBA. Since Winston and collaborator Jeff Sagarin (the NCAA ratings guru) were using their numbers to advise the Dallas Mavericks, he maintained a low public profile.
That changed in late September, when Winston published Mathletics: How Gamblers, Managers, and Sports Enthusiasts Use Mathematics in Baseball, Basketball, and Football. In conjunction with the book's release, Winston did a four-part interview series with TrueHoop's Henry Abbott entitled "Mark Cuban's Stat Expert Isn't Bashful." Indeed, Winston was not shy, most famously in a subsequent post where he said to Abbott he would not recommend to the Mavericks the addition of Kevin Durant even if he was available for free (because of Durant's poor adjusted plus-minus, a topic subsequently explored on Basketball Prospectus).
If your only experience with Winston is his interview series with TrueHoop, you may be surprised by what you read in Mathletics. The book has few of the bold conclusions Winston offered Abbott. Instead, Mathletics reads as something of a primer for advanced statistics across all three major sports, exploring their application to a variety of different questions as hinted at in the book's subtitle.
Mathletics is broken up into four sections, starting with one apiece for baseball, the NFL and the NBA. For each sport, Winston begins by exploring the basic fundamentals used by analysts before applying them and techniques borrowed from formal statistics to answer questions like the probability of Joe DiMaggio's 56-game hit streak and whether the NBA's rookie-scale contracts divvy up salary efficiently to draft picks.
Baseball statistics get the broadest treatment, with Winston citing a variety of different sources in addition to some of his and Sagarin's own methods to calculate linear-weights and win expectancy ratings. The NFL gets the least attention of the three major sports, and Winston spends nearly as much time evaluating playcalling and coaching decisions as he does rating football players.
While the NBA is the last of the three major sports covered in Mathletics, it is clear from the book that Winston's most extensive work has been in analyzing basketball players. Here, Mathletics is less of a survey, devoting a chapter apiece to Dean Oliver's Four Factors at the team level and linear-weights ratings (primarily PER and Win Score) before focusing on Winston's work with adjusted plus-minus. Naturally, he offers his strongest statements on the topic of the NBA, using adjusted plus-minus both to evaluate individuals and lineups. (This was the crux of the work he and Sagarin did for the Mavericks before the team opted not to renew their contract this season, instead moving consultant Roland Beech into a larger role.)
The last section deals primarily with team ratings and other questions that largely center on gambling. While there is a natural overlap between the gambling and analytics communities--both of them interested in gaining a better understanding of team performance, though for different reasons--most statistical analysis shies away from gambling implications, so Winston addresses some topics here that are relatively new for a widespread audience. I found his breakdown of how successful a gambler would need to be to make money (the standard is higher than 50 percent because of the vigorish charged by bookmakers) especially interesting.
Throughout Mathletics, Winston strikes a pedagogical tone befitting his job as a professor in the Indiana University School of Business. He walks readers through the math, working to keep it from being too complex. If readers wish to follow along with Winston's calculations, he explains how to replicate them in Excel, and sample files referenced in the book are available on Winston's Web site.
One of Mathletics' strengths is the variety of different resources Winston calls upon to provide insight. At the end of the book is an extensive bibliography of key books, research papers and Web sites where readers can find more information.
If anything, Mathletics may at times be too comprehensive of an overview of the existing literature. For fans who already read the Prospectus sites, Football Outsiders, John Hollinger and others on a regular basis, that information may be largely review. While the more focused studies offer new information for readers who are well-versed in modern sports analytics, Mathletics is best suited for newcomers who are just beginning to understand the field. It is also a helpful introduction for anyone who is familiar with the state of analysis in one of the three major sports but wants to broaden their understanding of the techniques used by the others. In that regard, Mathletics fills a new and useful niche.
For NBA observations and quick analysis, follow Kevin on Twitter at @kpelton.
Kevin Pelton is an author of Basketball Prospectus.
You can contact Kevin by clicking here or click here to see Kevin's other articles.