If there's one thing that's really bugged me about the Pirates this year (OK, actually there's like 42 things, but this is #1), it's the way Ian Snell's pitched. He and Gorzo have been equally bad, but I kind of expected it out of Gorzo (mostly because I figured he'd be hurt this year). Snell's ugly start gnaws at me because I can't figure out why he's been so bad and there's no real obvious reason for his dropff. I'm not someone to sit around and let something bug me when I think I can figure out the answer, so I've been digging through Dan Brooks' fantastic PitchFX tool for Snell's starts to see if I can figure out what's different this year. This is the first year that the PitchFX data for every start are available to the public and the numbers I can find from last year are very incomplete. Instead, I'm going to compare the numbers from this year to what I found in the Bill James Handbook 2008 for Snell. He doesn't explicitly say what system he used to chart the pitches, but I'm guessing his Baseball Info Solutions company uses either the same PitchFX system or something similar, because most of the stats he gives are the same or similar to PitchFX.
My initial suspicion, shared by several of you, is that Snell hasn't been throwing as hard this year as in the past. Last year, according to the James, Snell was 6th in the NL with a 92.4 mph heater. Using PitchFX, here's his average fastball speed from each of his starts, in reverse order, with a link to the boxscore from that night:
Can we conclude anything from this? He's definitely not quite throwing as hard as he did last year (only once topping his fastball average from the entire season in any one start), but he hasn't lost
that much speed and his average fastball on his best start of the year, April 6th against the Marlins (the 10 strikeout game), was 91.59. Maybe he's lost a little bit of zip, but it's nothing Barry Zito-like and I don't think we can blame that for all of his problems.
The next question is pitch selection. Last year (again, according to James), Snell threw a fastball 50% of the time (on of the lowest percentages in the NL and a slider 35.5% of the time (tops in the NL). Since I have the fastball data on hand, let's look at his fastball percentage from the same starts:
Ah-hah! I believe we're on to something here. Snell's throwing a lot more fastballs this year than he did last year. I thought maybe this was because his control has been worse this year, but it really hasn't. This year he's walking 2.92 batters per 9 innings and last year he walked 2.94/9. So if his control hasn't changed, why is he throwing more fastballs? Today in the Q&A,
Dejan indicated that he's recently dropped his sinker in terms of ironing out what's wrong with his fastball, but I'm fairly sure that that the PitchFX (and also, the Baseball Info Solutions method used by James) charts sinkers as two-seam fastballs and the actual data output just gives "fastball," so throwing less sinkers in favor of fastballs shouldn't change his fastball percentage at all.
I can't pretend to know what's going through Snell's head. There has been a slight drop in speed in Snell's fastball this year, but some of that may be due to the fact that he's throwing it more. For whatever reason, though, the percentage of his pitches that are fastballs is significantly higher this year. Has he lost faith in the slider (or another breaking pitch)? Is his fastball flatter than last year? Does the slight lack in velocity bug him to the point that he's just whaling away with fastballs, hoping to get it back? I don't know (and don't really have the data to find out). What I do know is that more fastballs means less guessing by the hitters, which is as good an explanation as any for his dropping strikeout rates and increased hittability.