## Mathletics

With the football calendar at its darkest nadir--no games, no signings, no draft, just some off-season workouts--and basketball season getting into full swing, maybe it's a good time to broaden our statistical horizons. If you're new to sports analytics or want to become more familiar with the methods used in other sports than football, I recommend Wayne Wilson's book Mathletics.

Wayne is a professor of decision science at Indiana University and has consulted for the Dallas Mavericks for several of the last few seasons. Basketball is his wheelhouse, but Mathletics covers baseball and football analytics as well. The book is really two things. It's a primer on the various principles and techniques used in sports analysis, and it's a how-to book on how to use Excel to do the actual computations. It's perfect for the guy who wants to grab some data off the Web and get his feet wet crunching numbers.

Reading Mathletics felt like taking a course in sports math, which is how I suspect the book got its beginnings. There are a large number of short bite-sized chapters, each dealing with a different stat or technique which likely correspond to lessons in Wayne's intro course at IU. A sample of the chapters in the baseball section include topics such as Pythagorean wins, runs created, linear weights, evaluating pitchers, win probability, park factors, platoon effects, and streakiness. (I have to credit the chapter on Range Factor with inspiring my development of Tackle Factor.)

The section on football is the shorter of the three. Mathletics touches on many of the major concepts in football analysis such as expected points and run-pass balance. Wayne summarizes several of the landmark football papers including the Romer 4th down study and the Massey-Thaler draft study. One of the chapters is based my linear regression model for estimating team wins, and Wayne has kind remarks for Advanced NFL Stats throughout. Those of you who enjoyed my intro post on game theory will like Chapter 23. One of my very first articles here was 'What Makes Teams Win', and Chapter 18 offers many of the same findings using a slightly different approach. Chapter 26 shows why the NFL's overtime system is broken, which is one of my pet causes. So if you see football through the same lens I do, and I suspect you do if you're reading this, you'll enjoy the football section.

For basketball, Wayne is a proponent of the 'adjusted +/-' stat and has used it consulting for the Mavs. I have a lot of respect for the approach in the basketball section of the book for two reasons. First, Wayne doesn't try to make his approach out to be mystically accurate, magically complex, or proprietary in any way. He says, "Here's why it works, and here's how to do it yourself if you're interested." Second, he gives fair and billing to alternative approaches, such as those by Wages of Wins author Dave Berri, basketball analyst John Hollinger, and those at 82games.com.

For a more in-depth review, particularly of the baseball parts, I recommend Phil Birnbaum's excellent write-up.

Note: My main criticism is the choice of example teams to show why the RPI college basketball ranking system doesn't work. He chose the Navy as a team so bad that a win against them can actually drop your ranking. True, Navy is ranked 247 in the country this year, but Indiana, where I'm told they take their basketball seriously, is 222.

### 4 Responses to “Mathletics”

1. Dave says:

I read the NFL part of this a few months ago (planning to finish the rest later), and I'd agree that it's worth reading. Some of the analytical techniques he gives seem somewhat simplified, but they are not bad starting points to improve upon; and you can substitute in more advanced statistical techniques where appropriate.

2. johnnyjohnnywu says:

Awesome! Just ordered it on Amazon...

3. JDTapp says:

Did you happen to see the essay by Yeon-Koo Chein of Columbia University in The Economists' Voice (a book series compilation of essays) last year about auctioning off field position in overtime? Instead of a coin toss, each team would bid where they'd be willing to take the ball at. The team willing to take the ball deepest in their own territory wins. In included a pretty good microeconomic analysis of the problem.

4. mr. parker says:

Not quite sure where this comment belongs...

Would it be useful to to know what the predicted outcome of a game is before we calculate the likelihood that the team who is winning has of winning. Does that make a difference?