Everybody and his mother knows that Jimmy Rollins has gotten off to a very poor start this year at the plate. As a result of this, a good deal of personal criticism has been leveled against Rollins, the content of which I won't repeat here -- suffice it to say it's been unfair. But the underlying facts aren't open to debate. Right now, Jimmy is pulling down a slash line of .229/.295/.283 for a wRC+ of 66. That's bad, and it's probably cost the Phillies at least one win in the early going.
This has attracted some recent gloom-and-doom commentary from some respectable non-idiot sources: for example, the estimable David Hale and the generally-okay Fangraphs. But the timing of this is curious, because if you look closely, Rollins hasn't been swinging the bat that badly of late.
This is one of the errors you'll see systematically from commentators (sabermetric or not) who purport to be able to cover all of MLB at once just by looking at players' season stats. I'm obviously talking about Fangraphs more than Hale here. The basic story of Rollins' season is that he got off to an unimaginably awful start in April, but has been better at the plate over the last couple of weeks. A national sabermetrics writer seeing Rollins' awfulness in April would have known to discount them due to sample size. But if Rollins' numbers are awful on April 30, then his season stats are obviously still going to be awful on May 23 -- expecting otherwise would mean falling for the gambler's fallacy. Yet the same national sabermetrics writer will invariably parachute in on May 23, look at Rollins' stat sheet, and say: "Hey, the overall sample size is bigger now, so now I can draw some conclusions!" When in fact, it may be the case that no new supporting data has been generated since April 30, when the same writer would have thought those conclusions to be premature.
Back to Rollins. The first question we need to ask is: if Rollins isn't cooked, then what should his offensive numbers looks like? This question is harder to answer than it sounds, because Rollins has been a completely different hitter at some points of his career than at other points. I'm not even talking about his 2007 MVP season, which is clearly an outlier that we can throw out for present purposes. Before 2007, he was a low-walk, high-strikeout, moderately high-BABIP doubles-and-triples hitter. Ever since 2007, his walk rate has risen (except in 2009), his strikeout rate and BABIP have plummeted, and his XBH stats have yo-yo'd back and forth. But I would say that the best baseline for Rollins would be some sort of amalgam of his numbers from 2008, 2010, and 2011.
Basically we're looking for a walk rate near 10%, a K rate under 10%, and an ISO around .140. If he gets those peripherals down and his BABIP doesn't drop to anything ridiculous, his results at the plate will be fine. Especially when you consider that the overall run environment in MLB is more pitcher-friendly in 2012 than it was in any of those years.
So far in 2012, Rollins has a 8.6 BB%, a 16.2 K%, and an .054 ISO. The BB rate isn't too far off, but the other numbers are. And yet, things have been improving recently. Eyeballing his game log, I think the most recent low point of Rollins' season came on May 4, when he went 0 for 5 and struck out 3 times against one Stephen Strasburg. Here's what his splits look like through and since that game (SSS).
|Thru May 4||113||5.3||18.6||.028||.286||.229||.268||.257|
|Since May 4||72||13.9||12.5||.098||.255||.230||.338||.328|
I don't want to oversell these numbers. Clearly, he still isn't where he needs to be yet. But his BB rate is where it needs to be (better than that, in fact). Meanwhile, his K rate and his ISO are at least getting closer. If not for the dropoff in his BABIP that has coincided with the improvement in his other stats, then his improvement would have been more noticeable.
But wait, you might say -- Rollins has had a BABIP around .250 before, in both 2009 and 2010. Why should we assume that his recent .255 BABIP has been unlucky? To see that, take a look at his batted ball stats.
|Thru May 4||43.4||19.3||37.3||16.1||.286|
|Since May 4||40.0||22.0||38.0||21.1||.255|
The formula seems pretty straightforward. The higher Jimmy's LD rate goes, the higher his BABIP goes. That correlation has held true in each of these splits -- until this year. So I don't think his BABIP will stay down if he keeps swinging the bat like this.
In short, yes there's reason to worry about Rollins, but there's also reason for hope. What he's been doing recently hasn't been too bad. He just needs to bring his power numbers up another tick, and his strikeouts down another tick. If he can do that, his rest-of-year stats will be fine.