We all know Trevor Hoffman has been having some problems this season, and much of the discussion has focused on the fact that he's not throwing as many change-ups as he usually does. A quick glance at the data bears this out; from 2007 to the present (the pitch-f/x era), Hoffman has thrown his changeup 31% of the time, but he's only thrown it 19.4% of the time this season. Clearly this is a discrepancy, but Hoffman's postgame quote suggests there may be more to this:
"I’m pigeon-holing myself into situations where the hitter can be a little more patient. He doesn’t have to offer at (the changeup). I’m pitching behind in the count. You do that in the big leagues, the numbers will indicate that. It’s more of an ‘out’ pitch than a ‘getting back in the count’ pitch. That’s the dynamic of that pitch."
In other words, Hoffman seems to be saying that he's throwing the changeup less often because he's not finding himself in changeup counts as often as he usually has in his career. As luck would have it, just today Fangraphs introduced pitch-type splits based on the count. The data are intriguing, to say the least:
The first table has Hoffman's career (since 2002, anyway) changeup splits, and the second table has his 2010 splits. Some patterns emerge. Obviously, Hoffman hasn't thrown any changeups with no strikes (except for one with an 0-0 count). He hasn't thrown very many of those throughout his career, though, so that isn't too strange. This year's numbers in 1-strike counts are also quite similar. However, for some reason, although the change up has been his bread-and-butter in two strike counts throughout his career, he's really not been throwing it in those counts this year. (It's not pictured here, but this doesn't seem to be a Pitch-f/x classification error; the decrease in change-up% is accompanied by an increase in fastball%). So, quite strangely, despite Hoffman blaming the lack of change-ups on the fact that he hasn't been in counts where he can use it as an outpitch, for some reason he's actually been throwing it less in those very counts. In particular, when he has an 0-2 count (an outpitch count if there ever was one), he's throwing fastballs 100% of the time.
Also, perhaps worthy of note, 100% of his pitches on 3-0 counts have been unclassified.
Anyway, about the only conclusion I can draw here is that, despite Trevor's protestations, it's not just a matter of not being in changeup counts; he's undeniably throwing the pitch less, even than when he usually would. I don't know why this is, but I do not think it bodes well.