My Baseball World Counterpart

I just received quite an Amazon.com purchase tonight. Win Shares is a baseball statistical analysis book by the legendary Bill James. For those of you who don’t know who he is … which probably is everyone (why aren’t more people into the statistical dissection of major league American sports :(?), Bill James is a revolutionary force in the baseball community. He started a new wave of interest back in 1977 by creating sabermetrics – the scientific evaluation of statistical data in an attempt to determine why teams win and lose. Through his series of Baseball Abstract annuals, he introduced multiple generations of baseball fans to such stats and formulas as Runs Created, Range Factor, and the Pythagorean Winning Percentage.

Now, if you haven’t ever checked it out, give my NBA statistical analysis blog – NBA Sim – a gander. Even though it’s still very much in its infancy and still seeking its own style and voice, I’m pretty proud of all the work I’ve put into it so far. Hell, I already have my first fan! Anyway, with NBA Sim, I try to virtually represent NBA legends throughout the history of the league as accurately as possible using statistical analysis of their individual season stats. I’ve done a lot of research into the best way to boil down all contributions a player makes in a game into one number that can be compared to Legends in different playing positions and eras.

What’s cool is that this is the exact same mission Bill James tries to accomplish in Win Shares only for baseball instead of basketball players. The difference is the level of research and complexity. My system essentially just weights each individual stat (points, rebounds, assists, turnovers, etc.) and takes about a minute to compute for each basketball player. James’ Win Shares involves 6 massive steps (pitching, hitting, fielding, etc.), each consisting of about 10 mini-steps, to determine for each player. Instead of weighting individual stats, Win Shares compares a player’s individual stats against the league average and then adjusts each stat based on era, home ballpark, and other crazy contextual factors. I would estimate it takes about an hour to compute per player.

Although it’s for a completely different sport and using much different stats, I can appreciate all the time and effort James has put into his research. In fact, I love just reading his descriptions of computing all the different pieces of his formula. Looking ahead, I can actually see that I’m grooming myself to become a Bill James of the basketball world … after a few years of NBA Simming, I would feel confident enough to start to do some serious analysis into the sport. I would fully enjoy trying to come up with a basketball Win Share counterpart … am I crazy?

By the way … the NBA playoffs begin this weekend. I have all of my jersies at the ready, though I still need to get my bedroom tv hooked up to the cable network so I don’t have to annoy the living shit out of my roommates by watching 24/7 in the living room. In case you need to get pumped up for this, here you go:

(1) Detroit Pistons vs (8) Orlando Magic
(2) Cleveland Cavaliers vs (7) Washington Wizards
(3) Toronto Raptors vs (6) New Jersey Nets
(4) Miami Heat vs (5) Chicago Bulls

(1) Dallas Mavericks vs (8) Golden State Warriors
(2) Phoenix Suns vs (7) Los Angeles Lakers
(3) San Antonio Spurs vs (6) Denver Nuggets
(4) Houston Rockets vs (5) Utah Jazz

Comments:
  • Friday, April 20th, 2007 at 09:57 | #1

    this is the Doh Blog of the Century!!!

Leave a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>