From April through mid-August, Hiroki Kuroda was one of the best pitchers in baseball. After throwing eight shutout innings against the Angels on August 12th, Kuroda's ERA was a sparkling 2.33. Things went downhill from there, though. Over his next eight starts, he went 0-7 with a 6.56 ERA before the Yankees decided to shut him down instead of allowing him to make his final start. Kuroda is a free agent this year and the Yankees have offered him a one-year qualifying offer worth $14.1 million, which he will probably decline. If he does end up hitting the open market, the Yankees will have to decide how much they're willing to pony up retain his services; and how heavily to weigh his last eight starts. So what the heck happened to Kuroda? How did he go from being an excellent, maybe even elite, pitcher to a complete albatross?
Looking past the differences in ERA, there weren't many drastic changes between Kuroda's first 24 starts and his last eight. His strikeout and walk rates remained roughly the same and he continued to induce a good amount of ground balls. What really stands out are his batting average on balls in play (BABIP) and home run/fly ball rate (HR/FB%), which are both very luck driven in small samples.
For comparison, Kuroda's career HR/FB% and BABIP are 10% and .282.
Russell Carleton of Baseball Prospectus did some awesome research that shows how quickly certain stats become "reliable." According to his research, rates associated with strikeouts, walks, grounders, and fly balls become reliable very quickly -- within just a few starts; but it takes multiple seasons for HR/FB% and BABIP to stabilize.
Based on eight starts, we can say a lot about Kuroda's ability to strike guys out, avoid walks, and induce grounders; but hardly anything about his ability to control what happens to the ball after that -- like where the balls are hit in relation to where fielders are standing.
Still, BABIP and HR/FB% aren't entirely out of a pitcher's control. Bad pitches tend to get hit hard, and hard hit balls tend to find holes (or turn into home runs), so maybe Kuroda was just throwing bad pitches? It certainly felt like his stuff was not quite as sharp late in the season, but its hard to say how much of that feeling was because he was allowing so many runs. Plenty of pitchers become fatigued and lose some of the pop on their fastball late in the season, so maybe that's what happened to Kuroda?
Velocity loss wasn't the issue. If anything, his velocity actually improved late in the season, according to pitch F/X data -- 91.3 MPH vs. 92.1 MPH for his fastball. Although his velocity checks out, it's possible that he just lost command of his pitches over this period and left too many over the heart of the plate. But if Kuroda was having any sort of chronic command problems, he probably would have issued more walks, which he didn't.
All signs point to Kuroda's "breakdown" being nothing more than bad luck. So should the Yankees be hesitant to bring him back next year? Well yeah. He'll be a 39-year-old pitcher which comes with some risk, but his last eight starts shouldn't be what scares them away. Kuroda was worth 3.7 and 3.8 fWAR is 2012 and 2013, respectively. Assuming Kuroda is able to stay healthy, which he always has, I have little reason to believe he won't eclipse 3 WAR mark again in 2014.