Daisuke Matsuzaka had a great 2008 season - we can all agree on that. Most of us also agree that he was assisted by Lady Luck. His ERA was 1.13 below his FIP; the only pitcher with a larger difference was Armando Galarraga (-1.16 E-F). His BABIP was pretty low at .267, and he was among the league leaders in HR/9 at 0.64. He was also among the league leaders LOB% at 80.6%, but not for the same reasons as guys like Johan Santana or Jake Peavy. While the last two guys had LOB%'s of 82.6 and 82.2, respectively, they also had WHIP's of 1.15 and 1.18, respectively. Dice-K's WHIP has 1.32, meaning he allowed an extra base runner every 6 innings, roughly; he was allowing base runners at the same rate as Paul Byrd, Javier Vazquez, and Gil Meche - Braden Looper was even 0.01 better than Dice-K.

As you can see, he was over performing last year, and we could've anticipated some sort of regression to his true talent. Unfortunately, his true talent seems to have decided to return to Japan, or something to that effect. His numbers this season are nowhere close to his performance from last year that earned him 3rd place in the AL Cy Young voting. Let's take a look at his PitchFX graphs to see if we can find any sort of clues to what caused his FIP to jump to 5.40 this season.

The first thing I'd be interested in seeing his pitch usage. The data comes from FanGraphs:

.

Yeah, I know, broken axis and all that stuff. I'm pretty sure everyone can tell that I chopped off the first 40% because it all fastball (for abbreviation key, click the FanGraphs link). So, just imagine another big chunk of red below the graph, and you'll be cool; I wanted to show the detail at the top of the graph, so I sacrificed the obvious part at the bottom.

Okay, now to analyze my rainbow graph: Dice-K's throwing a lot more cut fastballs and two seam fastballs now, and a lot less curveballs and split fastballs. I'm pretty sure that PitchFX identified these pitches correctly, since a cutter and a two seam can move the same way, as can a curve and a splitter; the identification is supposed to based on spin axis, though, so I'm not completely certain.

[Not sure if this is related, but I remember hearing a story about how the two seam fastball was a new phenomenon to either Ichiro or Hideki Matsui (not sure which), because it wasn't thrown all that much, if at all, in Japan. Maybe Dice-K saw the American pitchers throwing it and decided to pick it up?]

Moving along, let's see if his stuff has changed, velocity wise.

• He's never has a blazing heater, but he's been in the low-to-mid 90's consistently for his entire American career (Link).
• Slider hasn't changed much either, maybe a slight drop-off (Link).
• Cutter's been a bit erratic, but no more so than it was in 07 (Link). 
• His curve isn't dropping into the mid 70's as much as it had (Link).
• His change-up is actually slowed down a bit, but it's still in the low 80's, when he actually throws it (Link).
• His splitter's gone, but he never really threw it consistently to begin with (Link). 
• His two seam isn't exactly his go-to, most likely because it's been almost the same speed as his four seam (Link).

So, if the problem doesn't appear to be with the velocity of all his pitches, maybe he's showing some kind of injury through a change in release point. From his last start, all of 4IP and 91 pitches (Brooks Baseball, since FanGraphs doesn't have it up yet):

Now, compare that release point to his best start from last year [by WPA: .397], July 7th against the Twinkies, when he pitched for 7.1IP and 108 pitches:

The grouping is about 6 inches lower, which could be a major difference. His arm isn't dropping as much - could that be causing less movement on his pitches? Let's check.

From his most recent start:

Now, again, compare that to his best start last season:

  [Apparently the abbreviations changed. FA=FF, and the rest should match up.]

It appears his fastball has been tailing on him more, and his slider is dropping more. But, aside from that, he's about the same, movement wise. Is it a control problem? Again, to the graphs:

Most recent start:

And, again, compare it to that really good outing:

Remember, he threw a more pitches in the good outing, but not a lot more. The main thing that jumps out at me here is that Dice-K isn't throwing in a smaller area, which, when you consider that his stuff is moving slightly more, means that he appears to be trying to throw more in the zone. To me, the problem appears to be that, for all we complain about him nibbling and not attacking the zone, he was much better when he didn't try to keep the stuff in the zone. What do you think?