Did Jose Fernandez Get More Dangerous?

Let’s accept that, after four starts this season, there are few conclusions that can be reached about Jose Fernandez. After all, we need bigger samples of data, and even then, conclusions mostly have to be pretty soft. After four starts, we know very little. But we’re always allowed to make observations. We can identify hints of things, things that might be true, and so I ask, has Fernandez become more dangerous on the other side of his elbow surgery?

You probably haven’t missed it, but just in case you did, Fernandez rejoined the Miami Marlins not long ago, and one of the Internet’s most favorite pitchers has 32 strikeouts in 26 innings. He’s also issued just three walks, throwing almost three-quarters of his pitches for strikes. And it doesn’t seem like he’s lost any movement or zip. Set a low enough minimum, and Fernandez leads all starting pitchers in out-of-zone swing rate. It’s here that I want to linger. I want to talk about that number, and I want to talk about Fernandez’s best pitch.

When you look at a pitcher’s O-Swing%, a lot goes into it. Most obviously, pitchers throw multiple pitches. It’s pretty clear, though, what’s driving Fernandez’s elevated chase rate. Here are the chase rates on his curveballs, for each season, acknowledging that this season is made up of just four games:

• 2013: 43% chase rate
• 2014: 47%
• 2015: 60%

The pitch is a famous pitch. It’s Fernandez’s best pitch, and it has its own nickname. It’s the pitch that most allowed Fernandez to be a superstar before getting hurt. This pitch has always been outstanding, exceptional, virtually unparalleled. But the 2015 increase catches your eye. It caught my eye. As far as I can tell, no starting pitcher right now has a higher chase rate on his breaking ball. Fernandez’s rate was always high, but now this is extreme.

Changes in the numbers make me wonder about other changes. Changes that come before the numbers, changes responsible for the changed numbers. With pitches, I like to look at pitch locations, and so you may behold this .gif that was made possible by Baseball Savant. Heat maps of Fernandez’s breaking balls:

I think there’s a shift here. One that’s relatively pronounced. The breaking balls go glove-side more often, they stay more out of the zone and they drop. They’re lower — at the knees or below them. There’s a cluster in the borderline area, where it’s almost impossible to expect a hitter to lay off. Where the pitches might be balls, but they’re too close to watch.

That’s the visual approach, but numbers can be applied. On the breaking ball, Fernandez’s strike rate is higher than ever. Yet, his zone rates, based on PITCHf/x:

• 2013: 52% curveball zone rate
• 2014: 49%
• 2015: 34%

His rates of curves down the middle, from Brooks Baseball:

• 2013: 7% grooved curveball rate
• 2014: 5%
• 2015: 2%

Tellingly, Brooks Baseball also lets us examine average pitch height. I mentioned it looks like Fernandez’s breaking balls have gone lower this season. Here’s confirmation. Average pitch height, as reported relative to the vertical middle of the strike zone:

• 2013: -7 inches from middle of zone
• 2014: -8
• 2015: -13

Let’s bring this all together: Fernandez’s average curveball, this season, is five inches lower than his average curveball last season. He’s throwing fewer curves over the plate, and fewer curves in the middle of the plate, but hitters have been swinging aggressively, far more than they did in the past. So the pitch has actually gotten Fernandez more strikes. The path leads to this:

• 2013: 16% curveball whiff rate
• 2014: 22%
• 2015: 30%

The pitch isn’t unhittable, but it’s closer to unhittable than it used to be. The small sample means we can’t know very much for sure, but it’s encouraging. Yes, one of the starts came against the Philadelphia Phillies. You almost want to just toss out any Phillies-related data, but based on a quick analysis, the hitters Fernandez has seen so far have actually been a little better than average. So it’s not just a soft landing. It’s Fernandez doing good things, causing hitters to do bad things.

Here, for your enjoyment, is a swinging strike:

Here’s a swinging strike, in slower motion, for greater enjoyment:

And sometimes there are called strikes. So you can’t just identify the curveball and automatically lay off. As if it would be that easy, anyway.

The working theory: Jose Fernandez has developed better command of his breaking ball. He seems also to trust his changeup more, which wouldn’t be fair. But this is about the curve, and hints that Fernandez has gotten better about locating it where he wants — down and mostly glove-side. He probably didn’t want to throw as many in the zone as he used to. To his credit, it’s a good pitch anywhere. But Fernandez said during the course of his rehab he wanted to make sure to stay on top of his breaking ball more, so as not to drop down and put additional strain on his elbow. Dropping down would cause the pitch to flatten out. Staying on top would aid him in spotting the pitch somewhere low. One of the most challenging things to analyze is mechanical consistency, but Fernandez seems to have it. It also would make sense he might emerge from rehab better off, having spent so long paying attention to all his moving parts.

The conclusion is it’s too early for a conclusion. It’s not too early for an idea. It’s worth wondering: Is Jose Fernandez more dangerous? Has he gotten even better throwing breaking balls that induce do-nothing swings? If you can throw a pitch out of the zone, and still get guys to swing at it, you’ve got almost zero risk of downside. It doesn’t seem fair, but fairness doesn’t have anything to do with it.