PitchFX Breakdown: Ian Kennedy vs. San Diego & Ian Kennedy vs. Miami

Ian Kennedy - seen here doing his best Mike Zagurski impression - didn't seem to have his typically-sharp command last Saturday against Miami. Does Brooks Baseball's PitchFX data agree?

I'm going to try my hand at a new trick, so this is a bit of analysis on the fly - as such, forgive the potential sloppiness of it all. However, in watching Saturday's start from Ian Kennedy against the Miami Marlins, you couldn't help but notice that his command was a bit off from the get-go and that he was getting by on being effectively wild, a far cry from what we're used to from Kennedy. With PitchFX and the wonderful resource of Brooks Baseball becoming more and more a part of how the game is analyzed - this is stuff the professionals look at, too - I decided to provide a primer on some of the tools the site provides while fitting it into the context of one of our favorite pitchers. Hopefully this post provides fans with a good idea of just how much information Brooks provides (it's truly incredible - they track every pitch thrown in the majors).

I'll be comparing Kennedy's start from last Saturday - he finished with one run and seven hits allowed across 6.1 innings, striking out five and walking three - to his outing from April 12 against the Padres - final line of 6 innings, one run (earned), five hits, nine strikeouts, and one walk. The innings pitched and runs allowed from each outing are similar, but can PitchFX explain the difference in peripheral results and, perhaps, shed some light on which outing Kennedy was sharper in? (Hint: the answer won't be surprising.)

The first and most obvious question is, simply, is the perception that Kennedy's command was off on Saturday founded legitimate? A quick look at Brooks' strikezone plots reveals the details:

April 12 vs. San Diego:

Location_php_medium

via www.brooksbaseball.net

April 28 vs. Miami:

Location_php_medium

via www.brooksbaseball.net

Clearly, yes, Kennedy has issues locating in the lower part of the strike zone. Unless he's been taking effective velocity lessons from Trevor Bauer - and I certainly hope not, given his career average fastball velocity of 89.7 mph, per FanGraphs - Ian was clearly missing his spots on Saturday, and may have been fortunate to allow just one run in that outing. It isn't even fair to say that he was pitching to the ballpark, as Petco may be the only park more pitcher-friendly than the new Miami Carnival/Gentleman's Club. There's clearly something to observe here.

Now that we've established that Kennedy's command was indeed off, one of the first questions we ask when a pitcher is struggling with his command - and a question asked by the booth on Saturday - is whether or not the pitcher has lost his comfortable release point. If a pitcher can't consistently release the ball from the same position - whether because of an inability to find a consistent arm angle or due to releasing the ball too soon or too late in the delivery - the result is often missing the strike zone altogether or, even worse, putting the ball right in the middle of the zone where it can be crushed. Was Kennedy's release point on Saturday different than his outing on the 12th?

April 12 vs. San Diego:

Release_php_medium

via www.brooksbaseball.net

April 28 vs. Miami:

Release_php_medium

via www.brooksbaseball.net

By what the charts tells us, there's very little difference between the release point clusters of the two outings, especially when you consider the cluster of pitches on the upper-left edge of the San Diego chart are largely swinging strikes (and, oddly enough, mostly cutters), suggesting that those pitches were very in-sync with his typical delivery. Perhaps a couple of the pitches that appear to have been released a bit to the left (away from Kennedy's body) of the mass on the Miami chart were a bit off. Nonetheless, it didn't seem to be a consistent problem for Kennedy on the night - mainly just a pitch or two - and Kennedy's command certainly seemed to be off for more than just a couple of pitches.

Moving past release point, how about the movement of his pitches? Kennedy may have not been struggling with his release points, but if the ball wasn't moving as he would typically expect it to, then - obviously - it could have had a bat habit of not going where he wanted it to go. Why might this happen? Sadly, it's hard to pinpoint a particular reason that would be the main culprit - struggling with his feel for any of his grips, mound conditions, etc. Additionally, a change in pitch movement could have even been the result of the initial trajectory of the pitch - rather than the other way around - making it even more challenging to find out what caused his command struggles in the first place. Brooks offers both horizontal and vertical movement charts that sort each pitch by velocity, allowing us to identify the pitches in clusters as fastballs, cutters, changeups, and curveballs. Here are the charts, beginning with horizontal movement:

April 12 vs. San Diego:

Horzspeed_php_medium

via www.brooksbaseball.net

April 28 vs. Miami:

Horzspeed_php_medium

via www.brooksbaseball.net

For those curious, the large cluster of pitches in the upper-left corner is, of course, fastballs (the most frequently-used of his offerings), the smaller cluster just below and to the right of that is of cutters, the bottom-left group is changeups, and the bottom-right group is of curveballs.

I see a couple of things to glean from these charts (keeping in mind that the scales of vertical axes are different on the charts). First, Ian had much more control of the horizontal movement and tightness on his curveball in the San Diego outing than in the Miami outing. Second - a point that actually favors the Miami outing - is that he was able to consistently get his cutter to cut against the Marlins, whereas he threw a few mid-80's BP fastballs (i.e. cutters that don't cut) against the Padres.

Let's see what the vertical movement charts have to say:

April 12 vs. Padres:

Vertspeed_php_medium

via www.brooksbaseball.net

April 28 vs. Marlins:

Vertspeed_php_medium

via www.brooksbaseball.net

Here's where we get to what I think is the hugely telling part of this analysis. In the top chart - from the Padres start - it's very easy to see the different clusters - fastballs in the top-right, the small group of cutters just to the left of the fastballs, changeups in the upper-left, and curveballs in the bottom-left. In the bottom chart - the Marlins start - Kennedy's vertical movement is all over the place. With a cursory glance, the bottom chart appears generally more sloppy. Further, several of Kennedy's curveballs showed minimal vertical movement - i.e. he hung them - and his fastballs are very highly concentrated in one area, which means he may have been slightly less effective at mixing speeds and keeping hitters off-balance with his fastball (though I'm very willing to admit that this could be an issue of different scales on the horizontal axis and my inability to adjust the charts for them properly).

The most important - and easy-to-see - thing, though, is the difference in how many pitches had between -5 and +5 inches of vertical movement between each outing. In the outing against San Diego, Kennedy rarely threw a pitch in that particular vertical movement range, with four cutters, two fastballs, about half of his changeups, and zero curveballs residing in the -5 to +5 range of vertical movement. In the outing against Miami, on the other hand, Kennedy absolutely filled up the -5 to +5 range. From what I can gather, there appears to be five cutters, about five fastballs (with several more sporting less than +6 inches of vertical movement), all but two of his changeups, and three or four curveballs in that -5 to +5 range of vertical movement.

The important question to ask from this is, of course, whether this is a cause of Kennedy's location issues or a symptom of it. Was his curveball not diving on occasion what caused him to hang it, or was it not diving because he had occasionally hung it in the first place? Was his fastball sinking more because he had left it up, or was he trying to leave it up above the strike zone only to have it unintentionally dive down? I think a look at his strikezone charts by pitch type answer this question:

April 12 vs. San Diego:

Location_php_medium

via www.brooksbaseball.net

April 28 vs. Miami:

Location_php_medium

via www.brooksbaseball.net

(Random note: I don't think Kennedy really threw that many two-seam fastballs on Saturday - I think those are misidentified four-seamers stemming from the unexpected sinking of the four-seamer caught by the algorithm after the pitch was left up in the zone and had exaggerated sink.)

I think the most important pitch to look at here are the changeups, because a pitcher never wants to throw a changeup up in the zone, which can help establish the proper relationship of causation. The changeup is a bottom-edge-only offering, and a pitcher leaving it up a) causes the pitch to register as having "more sink" (i.e. flatten out relative to a spinless ball) than it otherwise would, and b) will likely result in something bad happening for pitcher. The frequency of changeups up in the zone in the Miami game - particularly in contrast to the impeccably-located changeups from the San Diego game - suggests to me that the evening-out in vertical movement on the changeup cluster in the Miami game is likely the result of the misplacement, not the cause of it. Similarly, I would say that it can be deduced from this that the changes in vertical movement are a byproduct, not a cause, of Kennedy's lack of command in the Miami outing.

So, what did we learn? First, that our eyes did not lie - Ian Kennedy did not, in fact, have his typically-phenomenal command last Saturday against Miami. Second, it doesn't appear to be a mechanically-related issue, as Kennedy's release points appear in-line with those he has shown in his more dominant performances, and I would venture to guess that any mechanical complication that would be significant enough to throw off his command this much would reflect somehow in his release point. Thus, it seems that, in spite of having his release point in line, Kennedy simply wasn't able to lock into the edges of the strike zone on Saturday.

Finally, to bring us back to the intro, it's important to recognize that, somehow, someway, Ian Kennedy managed to work 6.1 innings and allow one earned run with noticeably spotty command of his pitches. For someone who doesn't throw particularly hard, that's an incredibly impressive display of pitching acumen, not to mention, perhaps, a reflection of the magnitude of the Marlins' offensive struggles thus far in 2012. Kennedy was absolutely filling up the upper half of the strike zone with fastballs and changeups, and managed to mix speeds enough to leave the game in the seventh with a 2-1 lead. Much as this post might be interpreted as bashing Kennedy's outing last Saturday, the results are what matter at the end of the day, and the results were incredibly impressive.

SB Nation Featured Video
X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join AZ Snake Pit

You must be a member of AZ Snake Pit to participate.

We have our own Community Guidelines at AZ Snake Pit. You should read them.

Join AZ Snake Pit

You must be a member of AZ Snake Pit to participate.

We have our own Community Guidelines at AZ Snake Pit. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker