On Not Seeing That Coming: Ian Kennedy
I wasn’t a giant fan of Ian Kennedy coming into 2013. Kennedy the pitcher, that is, not the person — after all, he might be as charming as the Dos Equis guy for all I know. But Ian Kennedy kind of registered on my radar as a respectable fourth, maybe fifth starter in a pretty good rotation. He ought to give you a decent ERA, acceptable WHIP, and perhaps a strikeout rate that flirts with about 22%.
The allure of Kennedy’s 2011 season was what likely drove his ADP up into the 10th and 11th rounds, well before the likes of Mike Minor, Matt Harvey, Lance Lynn, and a host of other cherry-picked examples of pitchers who have been far more useful. Indeed, Kennedy’s 21-4 season in 2011 came with nearly 200 strikeouts, 2.88 ERA, and 1.09 WHIP. The predictors didn’t scream fluke, but perhaps the 3.22 FIP was a whisper that he was getting a touch lucky.
If 2012 was a little ho-hum, then 2013 has just been awful. Kennedy has thrown 101.2 innings thus far, giving up 65 earned runs while walking 40 batters, good for a 5.42 ERA and 1.40 WHIP. Everything appears trending in the wrong direction for Kennedy in 2013. His swinging strike rate and strikeout rate are down, his walks are up, his contact rate is up and his line drive rate is up significantly. He’s at the point where he’s unusable in almost any format imaginable. And even if I wasn’t particularly high on him going into the season, I certainly didn’t expect this kind of debacle.
My knee jerk reaction is always to look at the velocity, and in Kennedy, the fastball actually has more zip in 2013 than it was in 2012:
| 2012 | 2013 | |
|---|---|---|
| April | 90.2 | 90.8 |
| May | 90.91 | 90.13 |
| June | 90.5 | 91.55 |
| July | 89.71 | 91.5 |
And in using his fastball roughly 60% of the time, it’s a particularly important pitch for him. But despite the increase in velocity, batters are finding it easier to square the ball up. Here’s his line drive rate per month on his fastball from 2012 to 2013:
| 2012 | 2013 | |
|---|---|---|
| Apr | 13.56% | 28.99% |
| May | 20% | 27.03% |
| Jun | 17.86% | 35.71% |
| Jul | 21.92% | 56.52% |
In 2012, versus Kennedy’s fastball, hitters averaged a .268 batting average and .468 slugging percentage. Not thrilling, but his change was effective enough to keep things respectable. In 2013, hitters are averaging .314 on the fastball and slugging .562. Obviously we can’t take a 56% line drive rate on his fastball in July to the bank because of the sample, but the trend is nevertheless concerning. Taken as a whole, April-June averaged about 30% line drives on his fastball. It’s not that he’s just failing to fool hitters, he’s been far too easy to hit.
In general, Kennedy has been been working behind in the count far more often than he did in 2012:
This might lead him to get more of the plate, and thus — more line drives and crummy outcomes. And things appear to go south quickly for Kennedy. Looking at his first 25 pitches from this year versus last, hitters are taking early advantage of him:
The sample is a little small for 2013, but that’s a pretty sweet slash line for hitters right out of the gate. I don’t know if this speaks to his preparation, but his first inning results produce the worst slash line (for him) out of any inning over the course of the current season.
You have to squint pretty hard to see good news for Kennedy. His BABIP is a tiny bit above his career average, but certainly doesn’t smack of bad luck at .298. His 67% strand rate isn’t particularly good and could creep up towards his career rate of 74% given some good fortune. His HR/FB is uncharacteristically high. But I’m not sure these things come close to mitigating what appears to be a train on the brink of derailment.
Michael was born in Massachusetts and grew up in the Seattle area but had nothing to do with the Heathcliff Slocumb trade although Boston fans are welcome to thank him. You can find him on twitter at @michaelcbarr.
Retrospective trades are always fun, but this one is just giddy to think about – effectively, the D-Backs gave up Max Scherzer to get Ian Kennedy.
Arizona’s pitchers in general have been pretty disappointing this year, outside of a surprising Patrick Corbin. The bullpen has been volatile and awful, and the starters inconsistent or ineffective. To think I drafted Wade Miley this year, expecting a step up in performance from his strong ROY-worthy 2012 campaign. Heh.
…and I traded Anibal Sanchez away for Kennedy this off-season. Yikes
Effectively the DBacks traded scherzer for Kennedy, Daniel Hudson, and David holmberg
However, Scherzer made serious mechanical adjustments when he went to Detroit and his walk rate subsequently lowered. Scherzer probably wouldn’t have become the pitcher he is now in Arizona.
If you’re looking to analyze a pitcher’s surprising performance, you might want to consider his command (or lack thereof) and consult and compare location heat maps.
I understand the notion that a guy HAS been bad, that his control is bad and the batted ball trends are no bueno. But at what point do we say he’s 28, has had success in the past, is healthy, and has sustained his velocity…all to say he’ll probably just pitch better going forward. Even if there are no direct signs of it.
Just because a guy hasn’t been unlucky doesn’t mean he won’t improve, considering his longer track record of being pretty good. Ian Kennedy’s control is historically a strength, not a weakness, and I find it hard to believe that that has been zapped away when health or age shouldn’t be factors.
IMO he is firmly in the camp of “Prove it first”. Two of the factors you mentioned (age, past success) are completely irrelevant going forward. Two of them (health, velocity) have clearly not been sufficient to get out of whatever hole he is digging now. And since he really hasn’t been affected much by the gods of evil randomness, I don’t see why he can’t continue to suck going forward. Or indeed, lose his job in the rotation.
What I have usually found in these sorts of situations is that when the season ends, people finally begin to tell the truth after having lied all season. Some minor injury, some “tell” in his mechanics, some family problems, some coaching conflict, something.
Anyone who’s watched a number of his starts can tell you, the guy just gets crushed in the first inning. Repeatedly. I’m not sure what if any adjustments have been made, but he just struggles to locate in the first. I mean, he’s struggled to locate all year long, but especially in the first.
He’s definitely worth keeping, though, for the Diamondbacks. He’ll probably accept a low number in arbitration this year and no real value to other organizations. I hope they can keep him around and iron him out, because he can be a workhorse when all goes right.
Is he worth keeping? That rotation is getting pretty crowded.
Corbin and Miley are going to be there, McCarthy is under contract for next year, Cahill is under contract for a few more years, Skaggs and Delgado appear ready, and Archie Bradley is going to be knocking on the door soon. There’s also the potential for Hudson to return from injury next year, although it’s not something I’d count on.
I guess it all depends on if he’s actually this bad. How many of those guys should we expect him to be better than? Because he wouldn’t be cheaper than any aside from maybe Cahill.
I wouldn’t say that Skaggs is ready yet
The problem is losing sight of where sabermetrics is good and not good. It’s great at *descriptors* of what happened, showing the true value of a performance that happened without the fog of bad stats like ERA, etc. Then it has *predictors* that can show given what’s happening in those descriptors, a guy is getting lucky/unlucky as compared to a baseline of people posting similar descriptors.
Both of these are useful tools but not the ultimate question. Where things go wrong is assuming those descriptors define “talent”; it’s assumed a player is going to automatically follow a career curve based on those descriptors because that must be his talent.
That misses the more important question to ask, “We know he did it, but *should* this guy be compiling these descriptors?”; because that question describes what we all want to know as baseball fans and in fantasy leagues that try to get a read on future performance: *talent*.
2011 Ian Kennedy was, by all descriptors, great, and as pointed out the predictors showed he wasn’t “lucky”. But for those willing to ask, it was obvious that would be his career year and he did not possess that level of talent. I argued at the time most of his stats were padded against awful teams and bad lineups. He only faced 5 winning teams outside the bad NL West and in those games was pretty bad. And bottom line, just watching his stuff, I wasn’t sold — so of course was called names and trolled when I argued so in 2011 when Kennedy was at his peak.
Hopefully a unified theory that addresses true talent will come someday in baseball research circles. In the meantime, I think a lot of the problem with writers “not seeing things coming” is they aren’t willing to go against the popular hype of the moment and ask if a guy is really this talented, or are these precise numbers that have so perfectly defined his performance a fluke?
Agreed, I think there is some hindsight bias in this article. Being lazy and looking at xFIP suggests he was somewhat “lucky” in 2011, similarly unlucky this year, and 2010 and 2012 are a wash.
Having watched him a few times this year, it seems his command hasn’t been where it was in the past, but again this can be letting the results color the impression.
I don’t know Ian personally, and I don’t know much about Ian’s pitching because I typically have little to zero interest in watching the Diamondbacks because their jerseys and overall aesthetic is like sandpaper on my eyes. That said, I am confident that if a study was conducted by a medical institution titled: “Is Ian Kennedy charming?” the results would strongly suggest that Ian is indeed NOT charming. However, looking at Ian’s positive qualities, if the strikezone were framed around batters’ heads, Ian would be the Cy Young frontrunner undoubtedly. I also think the league should not have punished Ian with a suspension, instead they might have considered forcing Ian to wear his first name on the back of his jersey for the rest of the season.