The Pitch f/x View of Spring Training
We’ve talked about the cliches you hear every spring training – player X is in the best shape of his life, player Y has new-found focus, etc. One that’s actually useful (if only occasionally) is hearing that a pitcher’s learning a new pitch. Jeff Sullivan gives a quick run-down on all of the AL West pitchers trying something new here, and I thought it might be useful to take a look at some of the new pitches Mariner hurlers are testing out in Peoria through pitch fx. It’s all well and good to hear that a pitcher’s trying something new, but it’s much more interesting when you can see if the new pitch actually moves differently.
There are a number of huge caveats here. First, pitch fx is only in use in two ballparks in Arizona, which takes the already minuscule samples of spring baseball and then halves them. Second, this doesn’t really help us get to results, which is ultimately what matters; the samples can only tell us about movement, not if the pitches are actually, you know, good. Third, there’s an awful lot of guesswork involved here. No pitcher throws each pitch consistently; each pitch has a range of possible breaks, so picking out the pitch types isn’t a foolproof process. Fourth, pitch fx has been a little…weird at times. It may not be calibrated quite right, but personally, I’ll take what I can get.
First up is Ryan Rowland-Smith:
Shannon Drayer tips us off that the Aussie’s working on a cutter to help get righties out (his xFIP versus righties was nearly a full run higher in 2009 than his xFIP versus lefties, which makes sense given that his best pitches are a curve and slider). How’s it going? Well, this is tougher – RRS’s fastball movement has been a bit more variable, and that makes separating the cutters from FBs (the velocities are quite similar) more art than science. Basically, I’m looking for horizontal movement that’s *significantly* different from his FB’s horizontal movement – that’s the point of the pitch after all. In this case, I’m looking at pitches with negative horizontal movement, meaning a ball that breaks in to a righty. His normal fastball tails a bit away from righties (most all pitchers’ fastball tail a bit away from same-handed batters). You can visualize this by looking at the first graph here; from the top view, the cutter should move like his slider, albeit with fastball-like vertical break.
Well, so far, he’s thrown a handful of cutters to show that he can generate that slider-like horizontal movement. Unfortunately, his FB movement’s all over the place, so I could definitely be excluding a lot of less-successful cutters. Still, it’s nice to see that he’s actually able to generate distinct movement with the pitch without sacrificing a lot of velocity – and though the sample is pretty meaningless, his results in that ugly start against Colorado last Thursday were comparatively solid. As the samples increase, it’ll be great to see if his platoon splits improve. We’ll get a bit more data today, as Rowland-Smith gets the start for the M’s against the Angels.
As an aside, RRS was clearly working on his cutter and slider in that start against the Rockies, as he essentially abandoned his curve ball. He threw a grand total of two of them in his 2 and 2/3 IP, 66-pitch outing. *This* is why people say that spring training stats don’t matter. The curve has been Rowland-Smith’s best pitch (according to fangraphs pitch-type linear weights), and he just didn’t use it – that colors the way I look at his line. And in this case, it’s much less likely that pitch fx botched the pitch classification – RRS’s curve looks totally unlike his other pitches from a velocity or break standpoint.