Shelby Miller, the Cutter, and Quality of Opposition

The past year has been a mixed bag for Shelby Miller. In these digital pages during 2014, Dave discussed him being broken during April, Eno postulated on whether he was fixed following the regular season, and then he was traded by the Cardinals in November for Jason Heyward. Now, fresh off a two-hit shutout that is the culmination of a fine early season run, he’s sitting on a microscopic ERA (1.33) and WHIP (0.83). Have the Braves really fixed Miller?

Let’s start with a statistic: .183. That’s Miller’s current BABIP. No other qualified starter in the major leagues has a figure under .215. Case closed, right? His FIP is 3.28 largely because of that BABIP, his high LOB%, and low HR/FB rate, and the regression is coming for him. Yes and no: even if we can expect less absurd batted ball numbers moving forward, there are still a few interesting changes Miller has made that warrant a closer look.

Miller’s turnaround from his general malaise in 2013 and 2014 (when his strikeout rate cratered and walk rate rose by almost 60%) started in September of last year, when his overall command improved, helping to reverse the trends in those strikeout and walk rates. He never had bad command in the minors, so perhaps there was always the potential for a return to better times with his ability to limit free passes.

Eno mentioned that he also started locating his fastball a little bit higher, leading to more whiffs, something Miller may or may not have been consciously doing. That, along with his curveball regaining some of its effectiveness, drove a small return to form at the end of last year.

This year, we’re seeing something entirely different out of Miller. His pitch usage has changed, moving directly to his cutter and sinker over his four-seamer. Take a look:

Season	FA%	FT%	FC%	CU%	CH%
2012	70.4%	–	0.5%	17.9%	9.7%
2013	73.6%	–	1.8%	18.6%	6.0%
2014	61.6%	10.3%	5.8%	19.5%	2.4%
2015	34.2%	33.1%	19.1%	12.6%	0.4%

That’s not a small change. Way less four-seamers, way more sinkers and cutters, and a slight drawdown in curveballs. The results from such a change are largely what you would expect when switching to a sinker-heavy repertoire: far more ground balls. His ground ball rate is up over 10% from last year, up to around 50%. His sinker, a fairly new pitch for him, yields above average ground balls and whiffs compared to league average sinkers. Ground balls are always good, but they’re especially good when you have Andrelton Simmons playing shortstop:

Atlanta doesn’t have a great defense (ranking in the bottom half of the league by both DRS and UZR/150), but their infield defense isn’t as bad: they’ve gotten positive contributions from both middle-infielders by both DRS and UZR/150, and their corner spots are at or just below average. They’re fairly well set up for a ground ball pitcher, especially up the middle, which could be a small part of Miller’s BABIP luck so far this season.

Now let’s focus a little more on that cutter. It’s now usurped his curveball as his main two-strike pitch other than his four-seam fastball, accounting for 30% of his strikeouts this year (the curve is down to 9.3% from 12% in 2014). His four-seam and sinker velocity is up by a tick this year while his cutter has held steady from last year, so that speed differential could also be contributing to the 4.5% increase in whiffs/swing on the cutter between last year (25.6%) and this year (30.1%). The movement profile is largely the same on the cutter from last season, so the addition of another pitch (the sinker) while maintaining the great velocity Miller has always had is most likely one of the reasons for the improved success on the cutter.

Finally, we should talk about the pessimistic angle to Miller’s resurgence: the quality of his opponents. Although he has shown a great willingness to radically change his repertoire a few years into his career and improve on pitches already in his arsenal, the offenses of the teams he’s pitched against have been been unimpressive. Let’s take a look at the current wRC+ of the teams Miller has faced from start to start (offenses change over time, of course, but this will give us the right idea):

We can assume a few points of wRC+ here or there, but we get the gist – outside of the Blue Jays, Miller has faced the same three teams at least twice, and all have a below average offense. The Phillies in particular are one of the worst offenses in baseball, and besides Giancarlo Stanton, the Marlins are not exactly terrifying either. This is not meant to discount what Miller has achieved so far this season; it’s simply to imply that we should be a little skeptical of him continuing this performance given both his batted ball luck and quality of opposition in 2015. After we see Miller’s batted ball numbers stabilize, as well as see him consistently successful against above average offenses, then we can really start to get an idea of where he might settle.

Times are looking up for Shelby Miller: his greater inclusion of the sinker in his repertoire has given him what amounts to a new fourth pitch, one that allows him to get ground balls and increase the effectiveness of his other offerings. His cutter has also had great success in 2015 despite no real change in velocity and movement. As we know, heavy reliance on cutters doesn’t always work out in the long run (many pitchers have “new cutters” that stick around for only a few months), so it remains to be seen how his new arsenal fares against top-tier offensive talent. Miller isn’t this good. Almost no one is. Given his raw stuff, the important thing is that he’s not broken anymore.