Ian Kennedy’s Regressed Fastball-Off the Radar

Jul 30, 2013; St. Petersburg, FL, USA; Arizona Diamondbacks starting pitcher Ian Kennedy (31) looks up as the lights go out during the game and under delay at Tropicana Field. Mandatory Credit: Kim Klement-USA TODAY Sports

Ian Kennedy was traded from the Arizona Diamondbacks to the San Diego Padres for Joe Thatcher, Matt Stites, and a draft pick. In 2011, Ian Kennedy was worth five wins over a replacement pitcher according to FanGraphs and had seemingly become one of the best pitchers in baseball. In 2012, he regressed somewhat, but still was a slightly better than average pitcher and threw over 200 innings again. However, 2013 has been less kind, while he is still eating up innings, but at a well below average rate.

The regression lies heavily with his fastballs, which have become steadily less effective. According to FanGraphs wRC +, Kennedy’s fastball went from about 29 percent better than league average in 2011, his best year, to this year it has been about two times as bad as a league average pitch (regardless of pitch type). Basically, Kennedy’s fastball went from a very good pitch to a terrible pitch. Especially when it comes to fastballs, the first and most obvious thing to look at is velocity. This is clearly not the case with Kennedy, as his velo has actually increased slightly, averaging over 90 MPH with both his four seam and his two seam fastballs this season, something that he hadn’t done in previous years. So even though he is throwing harder, something else is wrong or different.

The next obvious thing is release point. Kennedy did have a release point change from 2010 to 2011. He lowered it by about 1/3rd of a foot on average and had more success. So far this year, it has been a little higher, but not as high as it was in 2010. His horizontal release point is where it was in 2010, a closer release point to the middle of the plate (only about 1/5th of a foot though) than it was in 2011. Release point could have played a role in his struggles, as he reverted back to an older version of himself. However, I wanted to keep looking at his fastballs, so I looked at all the fastball types that he threw as a Diamondback.

The first and most obvious thing is that he has added, or at least MLBAM has begun giving him, a cutter. Before 2012, he didn’t throw a pitch that was labelled as a FC as a Diamondback. However, the tags insist that he threw 40 of them in 2012, all in August or later. In 2013, he threw it from start one and has thrown it 137 times according to the automatic tags (for what it worth, Brooks Baseball’s manual tags have him throwing cutters all along, but the movement between 2010-2011 and 2012-2013 is different between the labelled cutters. They also call the 2-seamer a sinker). Unfortunately for Kennedy, it has been very ineffective according to the FanGraphs measure.

The cutter usage doesn’t explain why the two seamer and the four seamer have regressed however. Since we ruled out velocity, and release point may play an undefinable role, I wanted to look at movement. Are his fastballs moving the same way they were when he was challenging for the Cy Young? Here is his average movement data by pitch type and year on a graph

As you can see, the biggest difference is the cutter movement, with the 2012 cutter having very strange movement. It seems that his worst horizontal 4-seamer was in 2010, still a good year for him, and it was also his best vertical movement year. This year, he is getting it to move more vertically than it had the year before. The 2-seam has a slight drop in horizontal movement this year, but not a big one. For another look, the following has all of the fastballs he has thrown as a Diamondback on the movement chart, with me emphasizing about where the cutters are

With movement not helping a lot, the next logical step is location. Here are where he has located his individual fastballs over his time with Arizona:

Despite the movement difference on his cutters, at least in measurement, he has located it basically the same. Both the 4-seamer and the 2-seamer seemed to creep up, which could be a plausible explanation for the struggles. Pitchers are more likely to get whiffs when they throw fastballs up, but they are also more likely to give up homers. Kennedy has never been a groundball pitcher, but his home run rates have gotten worse over the past couple of years. Moving to San Diego and pitching more games in that park will definitely help his home run rate, but how whether or not his fastball improves will determine if Kennedy is just an innings eater for the Padres, or an effective above average starter again.