Win Share Notesby Dave Studeman
May 24, 2005
Based on the number of inquiries and requests we keep receiving, readers seem to really enjoy Bill James's Win Shares. We're happy to oblige with a running list of 2005 Win Shares, because we also feel there is much to glean from them. But there are some issues with Win Shares, particularly this early in the season.
One of the issues that we've tried to address is the "replacement level" issue, which James acknowledged in his original Win Shares book. As a result of some research I conducted during the offseason, I feel that I've got a pretty good cut at the number of Win Shares a bench player would accrue, given a certain amount of playing time. For each player, then, we've calculated the equivalent number of Win Shares a bench player would accrue, given that player's playing time, and each Win Share above the bench level is called, appropriately enough, Win Shares Above Bench (or WSAB).
I personally think that WSAB is a better measure to use than straight Win Shares, but I'll refer to both in this article, like so: 5/3 (WS/WSAB). As a reminder, one-third of a win equals one Win Share. And you can read more about Win Shares in last year's introductory article as well as this review at the end of last season.
With that in mind, here are some of the things I noticed in this early-season cut:
Jose Reyes has zero Win Shares.
The Mets' young heralded shortstop has zero Win Shares, as in zero batting AND zero fielding Win Shares. He's actually three Win Shares below bench level, for a combined Win Share score of 0/-3.
At the time I ran the Win Shares analysis, Reyes was batting .250/.276/.363, so I wasn't surprised that he had zero batting Win Shares. But I was surprised that he had zero fielding Win Shares. To understand why, compare his assist total to that of his partner-in-youth on the left side of the Mets' infield, David Wright.
Met shortstops had 79 assists at the time of the analysis, while Met third basemen had 95. That is the reverse of what you'd usually expect to see. Shortstops usually gobble up more assists than third basemen. So Reyes accrued zero Win Shares, while Wright tied for the major league lead among third basemen with 1.6 fielding Win Shares.
However, neither player's zone rating reinforces that sort of ranking, and I expect that this will correct itself over the entire season.
By the way, two other positions receive zero fielding Win Shares: Marlin first basemen (Carlos Delgado) and Mariner second basemen (Brett Boone). As I said, don't make strong judgements based on these rankings yet.
Chris Snyder has the most fielding Win Shares in the majors.
Chris Snyder is the catcher for the surprising Diamondbacks, and he's having quite the year behind the plate. Consider:
- He's thrown out 40% of baserunners.
- In fact, baserunners have attempted to steal a base against him only 0.2 times a game.
- Plus, the Diamondback ERA when he is catching is 3.86; vs. 6.05 when backup Koyle Hill is catching.
Of course, he isn't the "reason" that the Diamondbacks have done so well. If you want to pick one particular surprise among the surprising Dbacks, Win Shares nominates Craig Counsell, who is having a fine year in the field and at bat and is tied for third in the league at 9/7.
The second-most fielding Win Shares belong to the Pirates' backstop combination of David Ross and Humberto Cota. As of the middle of last week, Ross had thrown out four of six potential basestealers and the Pirate staff had a 3.04 staff ERA when he was catching.
Some NL pitchers really hurt themselves with the bat.
As you know, pitchers bat in the National League, not in the American League. This creates some problems with Win Share comparisons across leagues. For instance, two of this year's three worst-hitting pitchers (according to Win Shares) were in the American League last year, where they didn't bat, and their 2005 Win Shares totals will suffer in comparison.
- The Phillies Jon Lieber is 7th in the NL with 6.4 pitching Win Shares, but he has -1.3 batting Win Shares (tied for worst in the league) for an overall rank of 14th (5/3).
- Another AL refugee, Pedro Martinez, has 5.4 pitching Win Shares, which ranks 11th in the NL, but he also has -1.3 batting Win Shares gives him 4/2, which is 19th in the league.
It's "fun" to look at the difference between fielding and batting Win Shares for other notable players, too, such as the Nationals' Cristian Guzman (-2.3 batting Win Shares vs. 1.0 fielding Win Shares) and the Pirates' Jack Wilson (-2.1 batting vs. 1.9 fielding).
Half of the major league shortstops aren't yet playing above the level of a bench player.
I looked at the top thirty players at each position, according to playing time, and then looked at his WSAB. Here's a list of how many at each position are not playing above a bench level.
- Nine catchers
- Nine first basemen
- Ten second basemen
- Fifteen shortstops
- Twelve third basemen
- 26 of 90 outfielders
- Six relief pitchers
- 56 of 150 starting pitchers
These numbers will decrease over the full season, but teams with good shortstops have a definite advantage over other teams this so far this year.
The Dodger firstbase platoon is the tops.
I certainly don't want to step into the middle of the controversy over who should be playing firstbase for the Dodgers. But I do want to point out that the Dodger firstbase platoon of Hee Seop Choi and Olmedo Saenz ties for first in National League Win Shares, when combined, with Derrek Lee. And Lee is the leading contender, along with Bobby Abreu, for National League MVP.
Saenz has also seen some time at third, so this isn't a pure comparison; the two players combine for five expected Win Shares, while the NL firstbase leaders all have four. But the Dodgers have two fine firstbasemen. Unfortunately, only three NL teams (Colorado, Cincinnati and the Giants) have less pitching Win Shares than the Dodgers.
References and Resources
I want to thank Bryan Donovan for the tremendous job he did creating code that allows you to sort and re-sort Win Shares to your heart's content.
Dave was called a "national treasure" by Rob Neyer. Seriously. Comments about this article can be sent to him through the miracle of e-mail.