Diagnosing Shelby Miller’s Troubles

Shelby Miller just pitched six innings and allowed just two runs on Saturday. For a pitcher who entered that game having given up 22 runs in 23.1 innings — while also recording as many walks as strikeouts and averaging under four innings per appearance — the start was definitely an encouraging one.

However, there are some caveats, as well. For one, it occurred against a terrible Braves offense. And Miller still gave up another home run. And he recorded two walks against just one strikeout. Miller is far from out of the woods at this point. His main problems so far this season have been pitch selection and lack of velocity. The former is easily fixable. The latter could be a source of trouble if he can’t find the lost velocity at some point — or, alternatively, if the lost velocity is the result of some physical problem that has prevented him from maintaining consistent mechanics.

In each of the past two offseasons, Shelby Miller’s teams have decided to move him. Depending on your narrative, that’s maybe a sign that two organizations gave up on a young pitcher. On the other hand, though, the Cardinals’ receipt of Jason Heyward and the Braves’ massive haul a year later both contradict that narrative: both receiving clubs gave up quite a bit for Miller. When the Cardinals gave up Miller, he was coming off a relatively disappointing 2014 season where his 17% strikeout rate, 10% walk rate, 3.74 ERA and 4.54 FIP were all worse than his promising 2013 season. While the season overall was underwhelming, there were reasons for optimism on Miller when the Braves trade for him, and he delivered on that optimism last season.

During the 2014 campaign, the Cardinals made a deadline deal for Justin Masterson, and while Masterson did not pitch well for St. Louis and has yet to recapture his old form, he did teach Miller a two-seamer grip that Miller was able to use the rest of the 2014 season. At the time of Masterson’s arrival, Miller had recorded a 4.14 ERA and 4.81 FIP, with a 16% strikeout rate and 11% walk rate. To that point in the season, Miller was throwing his four-seam fastball 68% of the time along with a two-seam fastball less than 5% of the time. The rest of the season, Miller threw his four-seamer 48% of the time while upping his two-seamer to 22%. The result? An increased strikeout rate, fewer walks, a better, but not great 4.00 FIP, plus a nice 2.95 ERA over his final 10 starts.

Miller carried that two-seamer to Atlanta, throwing it even more last season (34% of the time vs. 33% on the four-seamer). His walk and strikeout rates remained the same as his late-season run in 2014. Halving his home-run rate helped Miller to a 3.45 FIP and 3.02 ERA — and Miller’s best season as a professional. Arizona made Atlanta an offer it couldn’t refuse and Miller headed into the season hoping to continue last year’s success. 

Miller talked to David Laurila about his sinker and the aims he has with the pitch.

“When you’re only throwing a four-seamer, guys see it and see it and see it. I think you have to mix it up. A sinker is a great pitch. It looks like a four-seam fastball and at the last second it moves. It has a couple inches of sink, which can be the difference between a fly ball and a ground ball.

“The sinker allows me to give a hitter a different look. Everybody is different. Some hitters are better than others, and some people hit sinkers better than others. It’s really more about going in with a game plan. You’re not trying to overpower guys with sinkers. It’s more a pitch for double-play situations and early in the count when you’re trying to get ground balls. You have longer at bats and you have shorter at-bats, and my motto is, ‘Try to get guys out with three pitches or less.’”

Miller’s description of what the pitch does is accurate. On average, the two-seamer gets three more inches of arm-side run and also around three more inches of downward movement relative to the four-seamer, per Brooks Baseball. The pitch allows him to work inside to the righties for strikes and outside to lefties without as much concern that they’ll square up the pitch for damage.

The heat map from Baseball Savant depicts where Miller threw his four-seamer last year.

newplot (16)

Miller has a good four-seamer with a solid 13% whiff rate, per Brooks Baseball. The two-seamer, meanwhile, isn’t a swing-and-miss pitch, as Miller himself recognized, but it’s an offering that induces swings and limits damage for Miller. Here is where Miller threw his two-seamer last year:

newplot (17)

As we would expect, Miller places the two-seamer a little bit lower and more inside to righties. That pitch allows Miller to use both sides of the plate and prevents opposing hitters from sitting on the four-seamer. Miller doesn’t record a lot of swings and misses, but 63% of balls in play against him last season were on the ground or popped up. When the ball did get in the air, little damage was done: Miller conceded just a .093 ISO against.

The pitch that might have turned the season around in 2014 and helped him to unprecedented success in 2015 has so far been abandoned this season. Whether using the classifications at PITCHf/x or Brooks Baseball, Miller has thrown fewer than 5% two-seamers this season. He appeared to throw more two-seamers against Atlanta than he had in his other starts this season, but it’s still a paltry number when compared to last season. Whether his four-seamer is less effective due to the lack of two-seam use might be considered too large a leap, especially considering Miller has dropped about 1.5 mph on the pitch this season, but the pitch is no doubt worse, getting fewer than 8% whiffs and leading to a .189 ISO by opposing hitters.

It might be prudent to write this off as a small-sample blip and consider Miller’s start against Atlanta a step forward. However, it seems unusual that Miller would adopt a new (sinker-dominated) approach a year and a half ago, experience success with that new approach, only to then abandon it upon joining a new team. We can talk about pressure associated with a big trade, but that sort of talk is generally groundless. What might be instructive is to consider that Shelby Miller was good last year, but he has reasonable desires to be great. Miller’s four-seamer, even without last year’s velocity, is probably a better pitch than his two-seamer. He can get more whiffs, and he has used more changes this year, which is another pitch that could get more whiffs. A Shelby Miller with his old fastball and new good split-change could be a great pitcher. Unfortunately, his arsenal is not complying and he has devolved into something less than good.

There is also the issue of diminished velocity, which naturally leads to questions about a physical problem. A change in mechanics doesn’t necessarily mean injury, and any number of issues could lead to poor mechanics. The chart below shows Miller’s vertical release point on his four-seam fastball by game since the start of last season.

Brooksbaseball-Chart (15)

We are talking about fractions of inches — and Miller is a pitcher who has had changes between starts — but it looks like Miller is releasing the ball from a higher and higher point each start, with a little bit less last game. Now here is the horizontal release point from the same period:

Brooksbaseball-Chart (16)

Miller could still be finding his stuff after a long offseason, as he seems to be releasing the ball closer to his body than he did a season ago. If he’s hurt, the pitch mix discussion is mostly moot. Going back to the two-seam fastball might bring back at least a little bit of last year’s success. Miller’s two-seamer in tandem with his four-seamer and a decent cutter make him a good pitcher, assuming he can keep the ball in the yard. His lack of velocity is likely hurting him this season, and it’s possible there’s something physical of which we’re not yet aware, as his delivery has been inconsistent. Even if his home-run rate last season was the product of some luck, at some level, it could stand to increase a bit from last year’s mark and still remain a net positive for Miller. That strategy will probably not make him a great pitcher anytime soon, but it will probably make him better than he is right now, which is quite bad. There’s nothing wrong with trying to be the very best, but right now Miller is a high-risk pitcher reaping very little when it comes to a reward. Setting his sights just a bit lower could result in him getting a bit closer to the target when it comes to results.

We hoped you liked reading Diagnosing Shelby Miller’s Troubles by Craig Edwards!

Please support FanGraphs by becoming a member. We publish thousands of articles a year, host multiple podcasts, and have an ever growing database of baseball stats.

FanGraphs does not have a paywall. With your membership, we can continue to offer the content you've come to rely on and add to our unique baseball coverage.

Support FanGraphs




Craig Edwards can be found on twitter @craigjedwards.

newest oldest most voted
victorvran
Member
victorvran

What a terrible trade