|Are Jimmy Rollins’s struggles finally over? (Icon/SMI)|
The article I was planning for today didn’t get finished in time, so I thought we’d have a theoretical discussion on mechanical adjustments, regression to the mean, and Phillies’ shortstop Jimmy Rollins.
As I noted on Buy on the Rumor last night, I filled in last minute on the Fantasy Baseball Roundtable Radio Show. One of the questions posed was “What is wrong with Jimmy Rollins?,” which spurred an interesting discussion when I mentioned Rollins’s claim that he found a mechanical glitch in his swing over the weekend. I didn’t get to articulate my point as well as I would have liked, or say as much about it as I would have liked, so I thought I’d talk a little more today.
Mechanical adjustments and regression to the mean
When I mentioned the mechanical adjustments Rollins claimed to have made, our good friend Mike Podhorzer immediately jumped in, wondering if it was just the typical BS we often hear from struggling players. He noted Rollins’s “unlucky” .239 BABIP, saying that a number that low is bound to come back up. Analysts often call this “regression to the mean” or “regression to a player’s true talent level,” but I posed a different view of what this actually means.
Sure, Rollins’s BABIP is very low and is almost certain to rise, but the reason that it’s so low to begin with may not be sheer bad luck. While we try to be as objective as possible and focus mostly on the numbers, we have to remember that we are dealing with human beings who are most certainly not focused only on the numbers. These are professional baseball players who have access to scores of video footage and are likely constantly evaluating themselves on a micro-level and making adjustments accordingly.
What we call “regression to the mean” may not simply be a matter of luck and sample size, but is likely also caused, in part, by players making adjustments (at least for some players). After all, while Rollins’s true talent may have been something like a .350 wOBA coming into the season, if his swing is different now, how can we expect him to perform to his previous “true talent level”? If it’s a different swing, it’s a different player, at least to some degree.
Maybe Rollins’s BABIP was low because there was a problem with his swing, but because he’s a professional baseball player he was bound to fix it, causing the BABIP to rise to it’s normal level. This, in turn, would cause analysts to classify Rollins’s initially poor BABIP as “bad luck” in hindsight, but perhaps there was actually more to it than that.
The specific case of Rollins — BABIP
Now, of course, the possibility also exists that this was just BS coming from a struggling and/or unlucky player. So let’s examine Rollins’s claim and see if the numbers back it up.
Scrutinizing footage of recent games, Rollins discovered a mechanical issue occurring at the moment his bat made contact with the ball: His swing was flat, meaning that his bat dropped less than an inch at contact, causing him to get under the ball and lift it in the air.
“Hitting pop-ups, usually it’s something right at the point of contact when your swing is coming off the plane,” he said. “Everything is good right before the point of contact – your timing and everything else – but you’re flattening out your swing right at that point.
“As opposed to keeping the angle of the bat above the ball, it might just get a little flat where it’s level with the ball. When you finish the swing, if you kind of go down, that creates the underneath-the-ball effect, and you pop up.”
Rollins then held his thumb and forefinger about an inch apart: “It’s literally between here and right there, and that’s a pop-up,” he said.
He closed his fingers and continued: “That’s a line drive.”
So, is Rollins hitting more pop-ups and fewer line drives? You bet:
+------+---------+--------+-----+--------+ | YEAR | LAST | IF FB% | LD% | OF FB% | +------+---------+--------+-----+--------+ | 2009 | Rollins | 6.0 | 18 | 34 | | 2008 | Rollins | 3.6 | 24 | 27 | | 2007 | Rollins | 3.5 | 20 | 41 | | 2006 | Rollins | 3.9 | 19 | 33 | | 2005 | Rollins | 3.7 | 24 | 28 | | 2004 | Rollins | 3.7 | 21 | 32 | +------+---------+--------+-----+--------+
Waaaay more pop-ups, actually, and the fewest line drives of his career (or at least as far back as 2002, the earliest we have batted ball data for). And while it’s obviously a very small sample, Rollins hit two line drives in last night’s game (50 percent). Since pop-ups become outs 98 percent of the time, this definitely has something to do with Rollins’s BABIP.
Also worth noting is that, according to our early look at HITf/x data, posted by Mike Fast at THT Live yesterday, Rollins was among the worst hitters in the majors in terms of Speed Off Bat during the month of April (280th out of 303). As HITf/x is brand new and we don’t have anything from 2008 to compare that to, we can’t say for sure that this isn’t the norm for Rollins, but there’s a very good chance that it is not. Speed Off Bat very likely has a high correlation with BABIP, and given Rollins’s .300+ career mark coming into the year, I very much doubt he’s among the worst in the league at hitting the ball hard.
As a side-note, I’m getting super excited for HITf/x. If we had it right now, we’d not only be able to check how Rollins was doing during his early season slump, but we’d also be able to check his Speed Off Bat numbers over the next week or two and see if this mechanical change does appear to be legit. Even if his BABIP is only .200 over the next couple weeks, having the HITf/x data would let us look below the surface and potentially say that “Yes, Jimmy Rollins has made changes and has simply been unlucky since then. Buy!”
The specific case of Rollins — Power
Finally, during the show, another good friend of ours, Patrick DiCaprio, conceded that perhaps a mechanical problem was to blame for Rollins’s BABIP, but he couldn’t see how it might be to blame for his power loss. While I’m no mechanics expert, I suggested that perhaps it was a matter of the batter shifting his weight improperly or something similar. Now, having a chance to read Rollins’s exact explanation of the mechanical change, I think I have a better explanation.
Rollins’s HR/FB is way down this season at 4.7 percent. To compare, it was 7.2 percent last season and above 10 percent in 2006 and 2007 (and tHR believed it should have been above 10 percent last year as well). If you look at the batted ball table above, however, you’ll also notice that Rollins’s outfield fly rate is very high, the second highest of his career. This is usually a good thing for a power hitter (more flies equals more opportunities for home runs), but for Rollins, in this specific instance, it may not be.
Rollins was never a guy who blasted the ball over the fence to begin with, so if he really is getting under the ball too much, altering the trajectory of his fly balls could have a significant impact on his home runs. If his fly balls are being hit too high up instead of being hit on a straighter line out, logically, fewer of them are going to be clearing the fences. They’re going to be landing in the middle of the outfield instead of on the warning track or in the stands.
This could also further explain the BABIP. The more time the ball is in the air (as would be happening if Rollins is hitting the ball higher up), the more time the fielders have to get under it and catch it. Fly balls are the easiest batted balls to field to begin with (aside from pop-ups), and Rollins may have been making things even easier for fielders.
So what do you guys think? Am I trying to hard to find a reason to be optimistic about Rollins (full disclosure: he was my most expensive hitter in LABR NL)? Am I simply engaging in a form of logical fallacy? Or does all this make enough sense to believe Rollins might be on the rebound?