Thursday, July 31, 2014

On the Coffee Table: Understanding Sabermetrics

Title: Understanding Sabermetrics: An Introduction to the Science of Baseball Statistics
Authors: Gabriel B. Costa, Michael R. Huber and John T. Saccoman
via Barnes & Noble
As long-time visitors now, I'm a big baseball fan.  The Orioles are my team and the last few years have been kind as far as recent history goes.  Like many baseball fans, I am fascinated by statistics, particularly the sabermetric approach which looks beyond traditional measures like batting average in order to determine a player's actual value to his team.  I've written about sabermetrics before, most recently here.

Understanding Sabermetrics is different from other books I've read on the subject - Moneyball and Baseball Between the Numbers - which were written by journalists.  Costa, Huber and Saccoman are all mathematics professors.  As such, their book delves deeper into the nitty gritty of statistical analysis.  It's a worthy study, to be sure, and I'm a numbers geek so it's fun but Understanding Sabermetrics isn't as readable as the others.  You really have to care about the subject matter to get much out of it.

That said, I'm glad to have read the book and I'll keep it around for quick reference.  It's short and therefore easy to flip through to find a particular formula.  The layout is like a text book, including problems to work through at the end of each chapter.  I was tempted at first but then reminded myself that I'm not actually a college math student.  I can just check the answers in the back and no one will care.  The authors even include materials from the sabermetrics courses they teach at their colleges, a primer for those looking to design their own.

The book's editing leaves something to be desired.  I caught one flat out mistake in their analysis but I was bothered more by the number of unanswered questions.  I realize they want us (not it) to do some of the work but the choice of what to answer and what not seemed arbitrary, or even sloppy.

This is definitely not the starter book if you're interested in learning about sabermetrics.  Moneyball is the better intro.  Still, I'll keep Understanding Sabermetrics on the shelf with my other baseball geek books.

22 comments:

  1. That looks interesting. But it's short you say? I'm always on the look out for any book on math and numbers, so I might check that out.

    ReplyDelete
    Replies
    1. Sure, it's short. But again, if you're interested, start with Moneyball - a far more engaging read. Then you can head to the sabermetric websites and go to town on your own. Fair warning: this stuff is highly addictive!

      Delete
  2. I'm not a fan of numbers, for sure--my husband might like this one, though.

    ReplyDelete
    Replies
    1. Again, start him with Moneyball. If he likes it, Baseball Between the Numbers is great, too. This one third.

      Delete
  3. Well, if it was written by math guys, you may have your answer to why it wasn't well written. Those two things don't usually go together.

    My wife read Moneyball after we saw the movie, but, then, she's a data analyst.

    ReplyDelete
    Replies
    1. Is she a baseball fan or does she just like playing with numbers? Freakonomics is a lot of fun, too.

      Delete
    2. She has read Freakonomics.
      Mostly, it's a numbers thing, though she does call herself an A's fan, now. Really, though, our only interest in baseball is because of our daughter and all of her softball.

      Delete
    3. If you ever get your hands on a copy of The Sports Gene, I think you and your family would at least enjoy the first chapter - all about softball.

      Delete
    4. I listened to interview with the author on NPR. That's a book I want to read.

      Delete
    5. Fascinating stuff. I think as a parent of an aspiring athlete, you'd get a lot out of it.

      Delete
  4. I'm not sure what the Red Sox are doing this season, but I don't think sabermetrics is on the table. Which is funny because they won last year thanks to playing that way.

    ReplyDelete
    Replies
    1. The Cespedes deal is huge, though. Wow! 13 games back and having given up two starting pitchers, they're probably not looking to close the current year big but as a move for the future, very bold!

      Delete
  5. My testosterone laden family members would be all over that, but I appreciate the heads up - Moneyball sounds like it is the thing.

    ReplyDelete
    Replies
    1. Yes. If they should read it, please let me know what they think.

      Delete
  6. I gave you a shout out on my blog today! You don't have to do the tag unless you want. I'm a no-pressure tagger!

    ReplyDelete
  7. I am a huge baseball fan but math is not my forte. I might check out Moneyball as I really want to see the movie and books that are movies are my niche!

    ReplyDelete
    Replies
    1. And I have never seen the movie!

      Who's your team, ceslavie22?

      Delete
  8. Numbers never lie. I love using spreadsheets at work - you can usually tell what's going on with a company by studying the spreadsheet. It just makes sense that the same theory applies with sports, or with anything, really. I'm more of a basketball person than baseball, and I know Coach Cal used numbers for his `tweak` to get Kentucky on their run to the NCAA championship this spring.

    ReplyDelete
    Replies
    1. Ah, but see, you can use numbers to say almost anything you want, whether it's true or not.

      But seriously, part of the point of sabermetrics is that for decades - really, over a century - people have been paying attention to the wrong baseball statistics. Batting average doesn't tell you nearly as much as on-base percentage. Stats like wins, saves, RBI are too dependent on the efforts of other people to tell you much about what an individual player is contributing. The saber crowd has sought to develop more meaningful measures.

      Other sports - basketball, hockey, soccer, American football - are starting to catch up but baseball has a considerable advantage over the others: mountains and mountains of data. Major League teams play 162 games a year, each. Every play of each one has been well documented for several generations. Far more than any other team sport, professional baseball is a statistician's gold mine.

      Delete