Mood Music - Heart of the Unicorn by Gamma Ray
With the well publicized lack of depth in the New York Yankees starting rotation, the Yankees are going to need big things from the perceived #2 starter, Phil Hughes. After winning a spring training competition in 2010 and returning to the rotation, Hughes gave us glimpses of why he was at one point the top pitching prospect in baseball, but also reminded us at times that he was a 24 year old in his first full season of starting in the major leagues.
As encouraging and successful as Hughes was before the All Star Break, he was equally depressing in the second half. While before the break he pitched like the #2 that we are all hoping that he can become in 2011, after the break, he was pretty close to replacement level.
Pre ASB - 3.65 ERA, 3.56 FIP, 1.18 WHIP, 8.11 K/9
Post ASB - 4.90 ERA, 5.19 FIP, 1.34 WHIP, 6.57 K/9
Most troubling of all, after excellent strikeout rates in April and May (9.00 K/9, 9.08 K/9), he simply stopped missing bats and developed a serious problem in finishing batters with two strikes. According to Baseball Reference, Hughes posted a 115 sOPS+ with two strikes, meaning that in two-strike counts he was 15% worse than the league average split.
I'm firmly of the belief that when the strikeout pitch abandoned Hughes, the two-strike woes began. More depth and lots of Joe Lefkowitz's Pitch F/X tool after the jump.
Here is some Pitch F/X data by month for Hughes' main pitches: the four seam fastball, the cutter, and the curveball -
(When you take a sample of one month worth of only one pitch, you're definitely dealing with the small sample size monster, and any analysis that I make from here on out is intended to be taken with a grain or two of salt.)
Well, now that we have accrued a huge amount of data, let's see if we can wade through and draw some reasonable conclusions. Firstly, I think you have to consider the fact that Hughes threw 176.1 major league innings in 2010 after throwing only 86 in 2009 (plus 19.1 at AAA). It is likely that Hughes battled fatigue during a 162 game marathon, and there is evidence both for and against that to be found in his Pitch F/X data.
It is remarkable how consistent Hughes was with his four-seam fastball. In both velocity and life, Hughes featured a nearly identical fastball every month, which opposes the notion that he was dealing with a tired arm. His off-speed deliveries, especially the curveball, were what deteriorated as the season progressed, leading to a sharp increase in contact percent.
Now that we have (sort of) figured out why Hughes stopped striking guys out (contact on the off-speed instead of swings and misses), I think that the more important question is why. Why did hitters figure out how to make contact with breaking balls as the season went along? I have several theories:
- Exposure - The more tape that gets out on Hughes, the more that hitters can study his pitch tendencies and the break on his pitches.
- Regression to the mean - Hughes' breaking pitchers are not as nasty as they were in April and May, nor are they the beach balls that they were in the months after.
- Pitch Selection - Hughes may have fallen into predictable patters with his pitches, making hitters less likely to be fooled and chase a breaking ball out of the zone. Also, Hughes' curveball seems much more effective when it's thrown at 76+ MPH, and less effective when thrown around 74 MPH. To me, this confirms what my eyes tell me: Hughes seems to throw two different types of curves, one that is faster with less break, and a big loopy slow one with lots of break, and the faster one is a more effective pitch.
- Fatigue - While Hughes may have been able to maintain the velocity on his fastball, he may have lost the arm strength to get the snap that he needed on his breaking pitches during the grind of the season. (Maybe some of you former pitchers could tell me if this holds water.)