2004 Win Shares Have Arrived

The 2004 Win Shares, which boil a player’s contribution to his team down to a single number, can be seen here.

Bill James invented Win Shares as a simple way to compare baseball players. The idea was to develop a statistic that allows you to compare shortstops and outfielders, starters and relievers, relievers and shortstops, etc. Since the win is the ultimate measure of success, James developed a stat that measures each player’s contribution to his team’s wins, or Win Shares.

Although the Win Share methodology is extremely complex, the output is simple: one number that represents the number of wins contributed by that player. Actually, Win Shares is the number of wins contributed by that player multiplied by three. Why? To make it simpler. By multiplying the result by three, Win Shares provides enough meaningful distinction between players. It’s enough to say that Albert Pujols contributed 41 Win Shares last year. You don’t have to say that he contributed 41.1. The extra decimal point doesn’t add any accuracy.

There are three types of Win Shares: batting (which includes basestealing), fielding and pitching. Over an entire league, batters receive a little less than half of all Win Shares, and fielding and pitching receive slightly more than half. This will vary quite a bit for separate teams, depending on their relative strengths.

So everyday players tend to garner more Win Shares than pitchers, because they are credited with both batting and fielding Win Shares. Shortstops and catchers generally receive more fielding Win Shares than other positions, because they play more crucial fielding positions. Strikeout pitchers receive more Win Shares than pitchers who allow balls in play, because they don’t depend on their fielders as much.

That’s why we like the Win Shares system. It’s simple and intuitive, and it allows you to compare players and pitchers across baseball history. Here are some other things to keep in mind about Win Shares.

  • Win Shares adjusts for contexts. For instance:
    • It adjusts for ballpark and conditions. A run is harder to score in Dodger Stadium than in Coors Field, so hitters receive more credit for what they accomplish in Dodger Stadium.
    • Fielding stats are adjusted for lefthanded and righthanded pitchers, as well as groundball and flyball pitching staffs. So Win Shares evens out the opportunities presented to fielders on the left side of the field vs. the right side, as well as infielders vs. outfielders.
  • Win Shares measures contribution, not true talent. If a player has a fluky year and hits, say, 50 home runs, he receives full credit for those home runs. The fact that he’s never going to do it again doesn’t matter.
  • Similarly, if a player hits well in the clutch, then Win Shares gives him credit for that, even if hitting in the clutch is not likely to be a consistent skill.
  • Ace relievers will receive credit for more Win Shares per inning pitched than a starting pitcher will. That’s because the innings an ace reliever pitches (late and close) are more important than most of those pitched by a starter.
  • If a player plays for a team that happens to win a lot of games, he generally won’t get credit for more Win Shares than if he had played for a lousy team. Win Shares is fair to all players, in that sense.
  • Having said that, a player will get credit for more Win Shares if his team happens to exceed its Pythagorean win/loss record. And vice versa.

Add these up, and you can see that Win Shares are a great player comparison tool. I like using them for Hall of Fame discussions, or MVP arguments. Also, they’re good for contract evaluations. Conversely, they’re not well-suited for evaluating a player’s true talent, so they’re also not good for forecasting a player’s or team’s future performance.

Even though we’ve done a lot of research into alternative approaches to Win Shares, this version follows the original Bill James methodology. I plan to roll out newer approaches over the next couple of weeks, including the “WSAA” (or Win Shares Above Average) approach. As those changes come about, I’ll let you know about them.

Also, keep in mind that it’s kind of early to be looking at Win Shares. You’ll see some funky numbers here, including some fielders with no fielding Win Shares, or batters with no batting Win Shares.

References & Resources
You can buy the Win Shares book from Stats Inc.

Here’s a full, detailed explanation of how Win Shares are calculated. The Baseball Graphs site also includes 2003 Win Shares, as well as a lot of detailed Win Shares research.

Using Recurrent Neural Networks to Predict Player Performance
Technology is rapidly advancing possibilities in decision-making.

My thanks go out to Pete Simpson, who taught me all about Win Shares and whose spreadsheets are the source of today’s Shares. Also thanks to Joe Dimino for helping me with the in-season methodology.

Print This Post
Dave Studeman was called a "national treasure" by Rob Neyer. Seriously. Follow his sporadic tweets @dastudes.

Comments are closed.