For the last several hours I have been cross-checking Daisuke Matsuzka's pitch f/x data (from Brooks Baseball) with choice games on Baseball-Reference.com between 2008 and 2010 looking for some pattern I could exploit and say "there it is! That is the reason he is so inconsistent!"
Really though, this may be a futile effort.
In terms of pitch f/x data, I have tried to pinpoint batter results by pitch release point, balls and strikes by release point, batter results and balls and strikes by strike zone plot against both right and left handers. The list goes on. There are a couple things that I will point out below but let me give you a little context first.
I was triggered to try and break down Matsuzaka once and for all after his last two starts have resulted in one of the oddest splits you will ever see in a pitcher. On May 22 against the Phillies in Philadelphia, Matsuzaka took a no-hitter into the eighth and finished the game with eight innings pitched, one hit, four waks and five strikeouts. His Bill James game score was a terrific 81, ranking the third highest of his career.
On Thursday, when everybody thought that Matsuzaka may have finally turned a corner. Then he comes out against the Royals (a team that has the highest batting average in the American League but has a very mediocre on-base percentage meaning they do not walk much) he goes 4.2 innings with one strikeout, three earned runs and eight walks. His game score for that clunker was 41 (which, considering the eight walks, was not that bad).
In both outings he threw 112 pitches, which makes the contextual difference between the outings more striking. Against the Phillies, Matsuzaka threw 73 strikes against 39 balls. Against the Royals it was 60 strikes to 52 balls.
Let's take a look at the strike zone plot from the Royals game by left and right handed batters, respectively. Note: Pitch f/x data is from the catcher's perspective.
What pops out with these strike zone plots are just exactly where the balls occur. If there is one thing consistent about Matsuzaka is that, for the majority of the time, he tries to pitch away from hitters and his balls tend to be focused in two zones. He is either missing high on the right side of the plate or low on the left side. Note how the plots tend to flow away from the hitter as Matsuzaka misses to lefties a lot high and outside and the chart stretches in that direction while he misses low and away to righties with the plot stretching the opposite in correlation.
What does this tell you? As a right-handed pitcher it means that his delivery is flying open on the right side of the plate and he is pushing the ball too far to the left side. This also has a lot to do with his release point. As Matsuzaka either lets his shoulder fly open too early or rotates too far, it is nearly impossible to fin a consistent release point for his pitches. Too many moving parts and the gears just fall off the machine.
Matsuzaka's release point against the Royals:
It is hard to put this in context except to note that the size of the overall cluster represents a large range of release points. Also note the fringes of green (balls) on the top left and lower right of the cluster, representing where he has either flown open too early or released too late. For a matter of comparison, here is the release point for Roy Halladay on April 11 of this year in a complete game shutout against the Astros with eight strikeouts, seven hits and no walks. Halladay is a good comparison to Matsuzaka because in terms of consistency of delivery, he is a machine.
Without breaking it down too far, it is easy to see that Halladay keeps his release point range smaller than does Matsuzaka, with a couple of fringe outliers drawing the cluster out.
So, whereas the scatter plots against the Royals were drawn out, Matsuzaka was much better within the strike zone against the Phillies, especially considering that Philadelphia is a team with a four dangerous left-handed or switch hitters (five when Jimmy Rollins is healthy, which he was not) in their everyday lineup. Matsuzaka was successful against the Phillies because he lived on the outer edge of the strike zone without deviating too far away from the plate, especially against left-handers.
Matsuzaka was good in the outside of the zone and keep the ball down without opening up too far. The overall strike plot for the game shows that he did tend to miss overall down and away to right-handers. His release point against the Phillies was also tighter though the difference was noticeable from where he was releasing the ball against the Royals verse where it was against Philadelphia.
You can see how the plot does is a little more center focused and does not have the rounded edge on the bottom left part of the cluster the way it does against the Royals. He still released late on balls to the lower left side of the zone, stretching the cluster out to the right but overall it was more consistent against the Phillies than Kansas City.
In each game Matsuzaka's primary pitch was his four-seam fastball. Against the Royals he threw 62 four-seamers and hit the strike zone with 32 of them (51.61 percent). A fair amount of those fastballs that missed were the ones that were placed high and on the outside of the plate. He did not throw any other pitch more than 16 (that being his slider, with nine strikes and then his two-seam fastball 15 times with eight strikes). Against Philadelphia his four-seamer was much more effective and he threw it a third less of the time with 41 pitches with a 63.41 strike percentage. His two-seamer was also on display that night with 29 pitches and 18 strikes.
Matsuzaka is more effective when he can establish the four-seamer and mix it with the two-seamer to give hitters two different looks at basically the same speed (92.93 MPH average with the four-seamer, 92.96 with the two-seamer). He needs to establish his four-seam fastball first though before throwing the two-seamer which sets up his slider and curve, which were extremely effective against Philadelphia with a combined 21 strikes on 30 pitches. Against the Royals he threw the curve ball 12 less times (19 to seven) and had 13 strikes on 23 sliders/curves.
The conclusion with Matsuzaka is that the only thing consistent with him is his inconsistency. As my former colleague Alex Speier at WEEI.com pointed out the other day, it is absolutely befuddling. From sore neck, back, shoulder to the difference of Japanese baseballs to American baseballs or hotel room beds. At this point, the difference in baseballs should not be a problem, especially from one start to the next. Matsuzaka did say that he had some "lower-body soreness" against the Royals and that could explain why he was flying open early. Really though, "lower-body soreness?" What is this, freaking hockey?
It seems that the one thing that Matsuzaka needs to do from here on out to climb back to the upper-echelon of pitchers is to stop worrying about his stuff, get out of his own head and pound the strike zone with his two effective fastballs, adding in his change-up the secondary pitches for balance.
Bottom line, Matsuzaka is not short on talent. He just needs to get out of his own way.