08 July 2009

Mostly About Kershaw: Game 83 Unfair Win Shares

Does Matt Kemp lead the major leagues in outfield assists? Almost, but no. Kemp has 9 assists. Jay Bruce of the Reds has 10.

Kershaw has been amazing lately, but I have had two worries about him. First, that he might be a home park pitcher who struggles on the road, and second, that he doesn't go deep enough into games. Regarding the first point, Kershaw still has a pronounced home/road ERA split of 1.85 vs 4.80, but I don't think it's a big deal at this point. Kershaw had very bad starts in Houston and Colorado early in the season, followed by a poor start in Philadelphia. Since then, though, he's been great on the road, pitching gems in Miami, Anaheim and New York City, and pitching a solid game in Denver that in context could probably be called a gem as well.

In fact, if anything Kershaw's home/road performance has reversed over the last month and a half. In his last four road starts he's gone 6 innings twice and 7 innings twice, while he's reached 6 innings only once in his last 6 home starts. Weird, yeah? I don't know that it means much, though. With home/road splits I think one needs to wait at least a season, and then another season, before making firm conclusions. And then be ready to revise those conclusions after the third season.

There is another Minotaurean split that is worth more scrutiny, though. That's the difference in Kershaw's performance against right handed batters and left handed batters.

Kershaw just destroys lefty hitters. They have a batting average of 0.175 against him, with an OPS of 0.476. Righty hitters fare better, but not by a lot --- they are batting 0.214 against with a 0.668 OPS. This split, by itself, is not surprising. A southpaw should do better against left-handed hitters than against right handed hitters.

What's really interesting is his walk rate against each kind of hitter. Kershaw walks 7% of lefty batters he sees, which is good, and 16% of righty batters he sees, which is not good. Since opposing teams tend to stack their lineups with right-handed hitters against Kershaw, that 16% is the important ratio, and results in Kershaw walking an awful lot of the hitters he faces, which surprisingly doesn't often lead to runs being scored, but does lead to him using a lot of pitches and leaving games earlier than we'd like.

To me this is really fascinating, though, because I think it shows that Kershaw can have a low walk rate. It's not just intrinsic wildness. He is perfectly accurate against lefty batters. If he could just pitch to righty hitters the way he pitches to the lefties ...

Well, maybe not. It's probably not that simple. Is Kershaw's relatively low batting average allowed to right handed batters a result of his extreme care with them? If he reduced the walks given up to righty hitters would the benefit be wiped out by a lot of extra hits being given up as well? I just don't know. I suppose a big part of this is his curveball, how it breaks to the different hitters, how easily he can locate it effectively in the strike zone against right vs left.

Clayton Kershaw gets compared with Sandy Koufax a lot, for a lot of obvious reasons. Kershaw's numbers this year against left-handed hitters are still a relatively small sample, but if you do take them at face value, then he's there, right there with Koufax in his prime, at least against left handed hitters. These are the left-handed hitters good enough not to be platooned out of the lineup, remember. Kershaw is dominating them just the way Koufax did in his prime. Against right handed batters, obviously, Kershaw still has plenty of work to justify any comparison with Koufax. He might never get there, but I think he will. He's only 21, and already this good.


Unfair Win Shares ( Dodgers )

Kershaw -- 1
Kemp -- 1
Manny -- 1

Unfair Loss Shares ( Mets )

Pelfrey -- 2
Wright -- 1

No comments: