Reviewing Nathan Eovaldi's Changeup

Eovaldi vs LAA The word that's been attached to Nathan Eovaldi the most since he was acquired by the Yankees, at least as far as I've seen, is "potential".  It's not hard to understand why when you look at the basic package: solid frame (6'2"/215), very good fastball velocity, 25 years old with multiple years of MLB experience already.  With that makeup, there's no reason Eovaldi can't be a top-of-the-rotation starter in either league.  He has all the potential in the world, potential that he's slowly started to show in the form of his improving FIP and K/BB numbers over each of his 4 MLB seasons.  He also has the potential to make the leap to upper-echelon starter this year if he can work out some of the kinks in his game and smooth over some of the rough edges.

The roughest of his edges might be Eovaldi's glaring lack of a reliable third pitch.  He came up as a heavy fastball-slider guy who threw a few cutters and a few curveballs.  Last year he was still primarily fastball-slider with a few more curveballs, a handful of 2-seamers, and a smattering of changeups.  The changeup is believed to be the secret to unlocking Eovaldi's potential.  It's a pitch he's struggled with in the past and a pitch he's admittedly spent a lot of time working on going back to late in the 2014 regular season.  Improving Eovaldi's changeup is going to be the top priority for Larry Rothschild this year, so I wanted to take a look at what he has to work with and what needs to happen to make it better.

In general, the changeup has never been anything better than an occasional 4th pitch for Eovaldi.  He looked like he was starting to work it into his repertoire when he went from using it 1.6% of the time in his rookie year (10 out of 595 pitches) to 5.1% of the time in 2012.  Since then, however, his usage has dipped back below 2% in each of the last 2 seasons.  Career to date, Eovaldi has thrown anywhere from 189 to 193 changeups out of 7,555 total pitches (about 2.5% usage).

Being positive and giving him more credit, let's use the 193 number, which is according to Texas Leaguers' PITCHf/x.  Those 193 changeups break down to 104 thrown for strikes, 80 that were swung at, and 14 that were swung at and missed.  Therein lies the big problem with Eovaldi's changeup, it doesn't miss bats.  Even worse than that, it gets smacked around.  Batters have a .440 average and .660 slugging percentage against it, which may explain why Eovaldi has had trouble establishing it as a useful pitch.  He needs to throw it more for it to become effective, but he doesn't throw it a lot in games because when he does, he gives up hits and runs with it.

Focusing in on the 2 largest yearly samples, it's easy to see why the pitch has been such a problem for Eovaldi.  The 107 he threw in 2012 plot out like this:

Eovaldi Changeup Plot 2012

That's too many balls left up in the zone.  It would be one thing if Eovaldi was throwing a pitch that had some late break or drop to it that finished it in those spots, but that's not what he was doing.  Most of his 2012 changes were slow, lazy pitches that floated into those spots, making them easier to identify than his fastball and easier to hit.  Mike Axisa had a gif of one of Eovaldi's 2012 changeups in a post he wrote on this topic back before Christmas and you can see exactly what I mean if you watch it.  That ball is coming in slow and straight.  With the overwhelming majority of these changes being thrown to lefty hitters, it doesn't take a baseball genius to see why so many of them have been hit and hit hard.

If Eovaldi was going to be successful with a changeup like that, he needed to be burying it in the lower left corner of that strike zone plot or even lower than that to generate swings and misses.  He wasn't able to do that and as a result he mostly scrapped the pitch over the last 2 years.

Eovaldi Changeup Plot 2014

There's the breakdown of the 57 changeups he threw last season and it's even worse.  Too many pitches landing in the middle of the strike zone, too many close to the corners where hitters can at least make contact, and a lower swing rate (42.1%) and whiff rate (5.3%) than he had in 2012.

It's clear that Eovaldi needs to work on the command and location of his changeup to make it more effective, and he may already be on the way to doing that with what he was working on at the end of last season.  His September changeups were in the mid 80s-low 90s velocity range, noticeably faster than the low 80s he threw a few years earlier, and they had some of that down-and-away movement from left-hander hitters that you want to see (again, the Axisa gif).

So it's not a completely clean slate that Eovaldi and Rothschild have to start from to build this pitch into a usable one, and given Rothschild's track record and Brian McCann's overall game calling and pitching framing skills, I'm confident that positive strides can and will be made this year.  Eovaldi already has his fastball-slider combo working well for him against right-handed hitters (.244/.309/.369 tripleslash against).  He needs to make the changeup work for him to be better against lefties (.288/.350/.421), and any improvement with that pitch should translate to more success across the board.  A starter with 3 useful pitches is almost always better than one with 2.  A starter with high-90s heat and 2 reliable offspeed pitches can be better than most.

(All stats and plots courtesy of FanGraphs and TL)