So the earlier entries have hit Maholm, Snell, Duke, Gorzelanny, and Ohlendorf in this series, which means that I've probably already covered the starting rotation. I've written in a few places that I thought that Jeff Karstens would get the final starting spot over Zach Duke, but the team has indicated that they're going keep Duke in the rotation, so I'm guessing Karstens is the odd man out. Is that a good decision? Let's find out.
The first thing that jumps out at me in Karsten's numbers from last year is that he got a decision in his first eight starts. That doesn't actually mean anything at all, it's just really weird. It's probably more important that those last six decisions were all losses and even in his two wins, including his near perfect game against Arizona in his second start with the Pirates, he didn't pitch as well as his 0.00 ERA showed. (I'm looking at his 6/5 K/BB ratio 15 innings).
Since we don't have a lot to go on with Karstens, let's take a look at his awesome start against the D'Backs and then a look at one of his uglier starts down the road to see if there's anything he did differently in the two starts. If you'd like to follow along at home, I'll be looking at data and stats from the August 6th start against the D'Backs and the August 17th start against the Mets (chosen because it was one of his longer poor outings, it's hard to learn anything from a 60 pitch shelling). As always, a hat-tip goes to Brooks Baseball for compiling this data.
The initial glance at the pitch charts (which, as the usual caveat goes, can be unreliable but work for the purpose of this exercise, etc. etc.) don't show much amiss. Pitch selection and pitch speed seem to be pretty similar in the two games, so there's not much to talk about there. And Karstens actually threw more strikes in his second start. I noticed, though, that the horizontal break numbers seemed weirdly different. Check out the overhead view of his pitches from the two starts (the first start against the D'Backs is on top):
Both his fastball and his changeup have a lot more horizontal break in the first start than they do in the second. There are a number of reasons for this. The algorithim that determines pitch type often gets confused between fastballs, sinkers, and two-seamers and has a hard time telling them apart and the same goes for the changeup, depending on the difference in speed. Karstens doesn't normally throw a sinker, though, and his changeup is at about 10 mph slower than his . Maybe the pitch had more movement on it, but it also looks like he was either throwing it from a slightly different arm slot or he moved to another spot on the rubber to throw it.
This certainly makes some sort of intuitive sense. If Karstens' fastball had more movement against Arizona, he probably had a harder time throwing it for strikes, but it would more effective, especially against righties, who the pitch was tailing away from on the day of his near perfect game. Then again, his slider seems to have much more break on it (look at where it starts and where it finishes on both graphs) in the second start.
Then again, Karstens only threw nine sliders in the 95 pitches he threw against the Mets, compared to 23 out of 115 against the D'Backs. Lower sample size means more variation in the pitches, so I don't think we can draw a conclusion from that. Instead, we'll stick with the fastballs. I went back in and looked at a few more Karstens starts. His fastball looked straight again on August 25 against the Cubs where he got shelled and again on September 7 where the offensively challenged Giants teed off on him for three innings.
It's hard to say anything conclusive from just looking at a few games' worth of data, but like I said above, this makes a lot of sense to me. Karstens doesn't really have any special qualities beyond the ability to throw strikes with a few different pitches and he can't really throw much past people. That makes him either (a) Josh Fogg redux or (b) the right-handed Zach Duke, whichever you prefer. A little bit of extra bend on his curveball and fastball could really throw a team off at the plate. Of course, the real questions are whether or not he knows how he did what he did against Arizona and how hitters would adjust to it if they saw it more often. I'm guessing Karstens will start the year out in the pen, so he'll have some time to play around with this sort of thing.