Has Jhonny Peralta changed as a hitter?

Be advised that this column uses math pretty heavily, and discusses BABIP in some detail. If you'd like some reading material on BABIP to catch up, I'd recommend the Fangraphs Sabermetric Library entry as well as this piece from Big League Stew. I've also hyperlinked the Fangraphs Sabermetric Library in-text, so most of the confusing stuff should have some references.

One of the (many) scapegoats for the offensive woes in Detroit this season has been Jhonny Peralta. Peralta, who was hitting .250/.333/.388 heading into today's game, was a key cog in Detroit's offense last season, when he hit .299/.345/.478. A few people have argued that that slash line was a career high for Peralta, and that Tigers fans should expect some sort of decline this season. I understand where that argument comes from, but after examining Peralta's batted ball data, I have to disagree. Instead, this year's Peralta has shown himself to be a dramatically different hitter than he has been in past years; not a worse hitter, but a different one.

The first place to look when examining a hitter's performance in-season (beyond, of course, batting average/on base percentage/slugging percentage) is generally BABIP. Batting Average on Balls in Play can generally indicate whether or not a hitter has been lucky or unlucky on the season. Generally BABIP falls around .300 for any given hitter -- depending on whether they're more of a line drive, ground ball or fly ball hitter -- and Peralta is no different. His career BABIP is a bit higher, at .313, which is probably the function of a reasonably high career line drive rate (20.5%). For the season, Peralta has a BABIP of .288, which is reasonably close to his career numbers, but still his lowest career BABIP by 21 points. If we assume he regresses to his career mean, he'll probably have a BABIP of .310 to end the season which will mean a spike in batting average and on-base percentage.

The second place to look for hitters is (at least for me) ISO, or Isolated Power. This is a quick way to check power output on the season. Currently, Peralta has an ISO of .138, a bit below his career number of .159. This is odd, considering his HR/FB rate (percentage of times that a flyball turns into a home run) is right at his career norm of 11.1%. This piqued my interest a bit, so I started looking into the batted ball data and plate discipline data more carefully. What I found was intriguing. (see chart below).

When Jhonny Peralta makes contact this season, there is a 30.3% chance that the ball turns into a line drive. This is an absurdly high number. But line drives are the best kind of hit; there's less of a chance they can be converted into outs, and that means that his BABIP should be higher. I decided to run this batted ball data through the Fangraphs xBABIP calculator, and based on his batted ball data so far this season, Peralta should have a batting average on balls in play of .369 this season.

Obviously this screams regression; I'm no dummy. Peralta will, in all likelihood, have a LD% of lower than 30% at the end of the season. It's also possible that the line drives have been classified differently based on the parks that they've been recorded in (I had a Twitter conversation where someone pointed that out for me).

But the numbers are still curious. I'm a big fan of this Eric Seidman piece that points out when samples become reliable for statistics (or when we can start making assumptions based on skill rather than statistical noise). At 174 plate appearances, line drive rate and strikeout rate have both stabilized. What this means is that a good chunk of that 30.3% line drive rate is skill rather than statistical noise, and this indicates to me that Peralta may have changed as a hitter, prioritizing line drives and plate discipline as opposed to power numbers, which means that he may have a higher batting average and on-base percentage this season.

The K% might regress to career norms (factoring in Wednesday's game, it is back up to 19.1%) and the walk rate might as well (if Peralta doesn't walk again until his walk rates become a "stable" sample of 200 plate appearances his BB rate will drop to 9%, which is close to career numbers). But the batted ball data suggests that Peralta may have become a different hitter, and while it is not unlikely that his numbers could regress to career norms, he may have changed into a different player at the plate. This isn't to say he's a worse player; in fact, I'd argue that he's still getting unlucky. Just remember: Jhonny Peralta is not a good whipping boy. He has some skills- and those skills might be changing on us.

Be patient.