Ben Zobrist's Weird Platoon Splits
Ben Zobrist has always been a switch hitter. Ben Zobrist has always had a slight platoon split. Up until 2012 he hit righties to the tune of a .337 wOBA while hitting lefties at a .356 clip. We can go one better and use our knowledge of regression to form an expectation of how he should handle each type of pitcher. Using Whelk's tool we see that you should expect BenZo to have a wOBA of .346 vs. righties and .366 vs. lefties. This quite literally begs the question of, "Well what has he done lately?" Well shut up for a minute and I'll tell you you filthy seeker of alms. Zobrist has hit lefties at a .317 wOBA while mashing righties at a .365 clip. Say whaaaaaaat?
Using Pitch F/x courtesy of Joe Lefkowitz (note that some games are missing, feel free to overlook this note as much as I have overlooked the gaps) we see that he has faced nearly twice as many pitches from righties so it's safe to say that roughly one third of all of his pitches have come from southpaws. What's incredibly intriguing here is how he has seen virtually the same pitch mix regardless of pitcher handedness. As in the past, we can use Run Values (RV) to get a great idea of how he is faring on each pitch type. We can then adjust these sums to put them on an even keel aka per 100 pitches.
Against lefties he has struggled a bit with the breaking ball (BB) and a bit more with the fastball (FA), but has done well against the change up (CH). Bear in mind that this adjusts for count whether the pitch was a ball or strike and if put in play whether the result was a single, double, triple, homer, or out. Against righties we see that he has struggled quite a bit with the BB, but has had great success against the CH and a bit less, but still performing quite well, against the FA.
I like to compare this to wOBAcon which only looks at results on balls in play. We can infer that if both numbers are bad/good then it's due to taking balls for balls and having success on hitting strikes. When these numbers are in discord it is due to 1) discerning balls and strikes well, but lacking success on balls in play (good RV/bad wOBAcon or 2) the opposite of bad discipline, but good results (bad RV/good wOBAcon). You can think of the RV numbers as a better indicator of process while wOBAcon stands as a better tool for measuring results. We see that his struggles with the BB from righties and the FA from lefties are confirmed. The grayer areas involve the BB and CH from lefties showing a bit of bad process, relatively good results, and we see the CH and FA from righties showing better process than results, thus far. This is a longwinded way of saying that you'd expect to see better results from him going forward against the fastball and slowball from righties going forward with a bit of a downturn on the breaking ball and change from lefties, assuming his process stays the same.
The overall RV lines show that, so far, he's had a better process against righties than lefties while having nearly the same exact wOBAcon against each. This, again, would seem to indicate that we should expect him to handle righties better going forward and more inline with the expectations laid out by the regression tool. Let's also take a look at these RV numbers by pitch location:
We see that versus righties he's doing a pretty good job of laying off pitches on the corners out of the strike zone and inside. He's also excelling on pitches up and down over the middle of the plate while showing better than average numbers on pitches middle-middle and middle-in. Where he is really struggling is on pitches up/down and in and everything on the outer third of the plate. To see the location figures above a bit better I think this chart can be useful:
Zobie is not having much success on the outer third and that's exactly where he's being pitched. A LOT. You see barely any pitches inside with almost no CH as pitchers are only throwing FA and BB there. This is an area where he's going to have to adjust to take this away from pitchers until they start attacking him inside more. That means going the other way and not trying to pull everything. Once word gets around that he's covering that part of the plate better then surely pitchers will adjust, but in the meantime he's got to be looking fastball away as that's the predominant pitch that he's seen. Let's take a look at how lefties are throwing to him:
Recall that lefties are the pitchers he's struggling with more this year despite showing that he's better against them throughout his career. He's doing well on everything off the plate and showing great success on pitches up and over the middle, and inside/away-middle. He's really struggling on pitches down on the corners of the zone and in what can only be considered an upset he's struggling mightily with pitches right down the pipe. That seems incredibly odd thus far so let's see what the location chart has to tell us:
We see a ton of breaking balls in and down both over the plate and further inside while getting most of his CH down and fastballs all over the edges and middle of the zone. We see that their pitching him away a bit more than inside, but not as noticeably as righties were staying away from his power. He's getting almost no pitches up. Because he's not seeing such a concentration of pitches from lefties it would be foolish to attempt to sit zone. He needs to continue to take balls that are called as it seems he's doing a good job there, but he's also going to need to see better results on those pitches down within the zone. It looks like pitchers show real fear about leaving anything up and maybe this is a point where some of those low pitches start turning into ground ball singles instead of ground ball outs, but the one area he should be mashing is that middle-middle portion where he's seen a lot of pitches and hasn't done a whole lot with them yet.
Between some regression and making a couple of adjustments it is likely that Ben Zobrist will see his final splits lines approach expectations of hitting lefties better than righties, and both very well. Only time will tell if this proves to be the case.