The Emergence of Shelby Miller

It’s no secret that Atlanta Braves starter Shelby Miller has pitched well to open up this 2015 season.

Miller has been impressive thus far, posting a 1.60 ERA and striking out 39 across 45 innings. In fact Miller, who was acquired last offseason from the St. Louis Cardinals, has already been worth more this season (0.8 fWAR) than he was all of 2014 (0.5 fWAR). Miller’s Deserved Run Average (DRA) of 3.18—Deserved Run Average is a new pitching metric developed by Baseball Prospectus, which you should read about here—is even the third lowest in all of Major League Baseball. It’s obvious the former 19th overall pick has had a fair amount of success this year. However, the question remains:

What is the reason for Millers’ strong start, and is it sustainable?

Well, one of the main reasons the 24-year old has been so dominant thus far is the use of his three fastballs.

What’s odd about the young right hander is that he does not often use an off-speed pitch. As a matter of fact, 86.27% of all of his pitches this season have been one of the three variations of fastball he has in his repertoire—a four-seam fastball, a sinker, and a cutter.

Miller has used these pitches almost perfectly this season, holding opponents to a .176 AVG (25 for 142) and striking out 36 of the 158 batters he has faced. However, let’s dig deeper into each pitch.

The sinker seems to be Miller’s personal favorite, as he himself has cited this pitch as being one of the main reasons for his success this year. In a piece written by David O’Brien of the Atlanta Journal-Constitution earlier this month, Miller described how he felt the sinker took him to the next level of pitching. While Miller himself calls it a sinker—saying he picked the pitch up by watching Justin Masterson pitch when they were both in St. Louis last season—it could also be considered a two-seamer, but that’s neither here nor there.

Miller has pitched very well with his sinker since he began using it, using it this season to primarily get ahead in counts—especially against lefties. On average, he actually throws it about as hard as his four-seam fastball. His sinker, which averages 94.45 mph, is actually one of the harder thrown sinkers in the league.

When we look at his sinker’s movement, it has actually changed a little from last season. As a matter of fact, Miller’s sinker has actually been dropping less than it did in 2014—but moving more inside on a right hander and away from a lefty.

Don’t believe me? See for yourself:


Like most sinkers, Miller uses his primarily on the inside portion of the plate for a righty—resulting in the high number of groundouts it produces. Although opponents are batting .250 off of his sinker, what it really helps to do is set up his other pitches. Miller’s three fastballs complement each other nicely because they can help disguise one another, as they all start from about the same release point and are generally around the same velocity—with his cutter being the exception.

Now for Miller’s four-seam fastball.

Owning one of the leagues’ lowest BAA’s on four-seams, .179 BAA (10 for 56), Miller has dominated hitters by just simply attacking them. Out of the 242 four-seam fastballs Miller has thrown this season, 67.76% have been for strikes. Combine that with the fact that Miller throws this pitch for a strike ≈65% of the time when he uses it for the first pitch of an at-bat.

As these heat charts allude, Miller prefers to use his four-seamer to come inside on left-handers. As for righties, he prefers to still stay on the right-half of the plate while also elevating his fastball when need be.

Shelby  Miller                  (2)

Shelby Miller FF vs. RHB

Shelby  Miller                  (3)

Shelby Miller FF vs. LHB

Shelby  Miller                  (1)

Shelby Miller FF overall

For a pitcher, I’d make an educated guess that it’s easier to go after hitters if you have some heat in your back pocket. Originally from Texas where—believe me—heat is abundant, Miller’s fastball this season has sat typically in the mid to upper 90’s. And while a fast fastball does not always lead to success, it often will when it’s whiffed on a good amount.

In fact, Miller’s Whiff per Swing percentage on four-seam fastballs is 24.19%. That’s good enough for 18th out of the 116 pitchers who have thrown that pitch at least 100 times this year. It’s also likely a contributing factor in the 32.14% strikeout percentage he has on the pitch.

Now onto Miller’s most effective pitch, the cutter.

One of the more fascinating things about this pitch for Miller is that it isn’t a flashy pitch with a great amount of movement or velocity. Here’s how his cutter breaks down:


What that means is that the average movement and velocity for Miller’s cutter is, well, average. However, just because the pitch itself doesn’t stand out when compared to other cutters around the league doesn’t mean it won’t have a great result.

Statistically Miller might actually have one of the best cutters in the game right now, if not the best. Opponents are batting an abysmal .079 (3 for 38) with 14 K’s off of the aforementioned pitch. It’s been an especially effective pitch against lefties, who are 0 for 16 on the pitch so far this season.

This cutter has been a groundball machine for Miller, producing a groundball 54.17% of the time it’s been put in play. Combine that with the fact that Miller’s cutter is also a whiff machine, as it carries a 31.34% Whiff/Swing percentage, and you come to one conclusion:

If Miller’s cutter itself doesn’t have anything overly special as far as movement or velocity go, how is it so effective?

The answer to that actually lies in how he intermingles all of his pitches. Take a look at the release point for all three of his fastballs, as well as his curveball:

Brooksbaseball-Chart (2)

All four pitches are released in the same general area, making it tough for a hitter to read what’s coming. Think of it this way, you have three fastballs that are all generally around the same speed—with the exception being the cutter, which is understandably slower—which are all released from the same area, and all move in different directions. That’s tough for any hitter to read, especially when Miller locates them as well as he has in 2015.

Therefore, the effectiveness of the movement and velocity of Miller’s cutter lies in how well he is able to disguise it with his other pitches.

Combine that with Miller’s curveball—which has a BAA of .067 (3 for 15) and is one of the harder ones velocity-wise the league has to offer—and you have a very nasty repertoire.

So each pitch has been effective thus far, but what does this mean as far as balls in play?

Well, in the words of Father Pat from the Will Ferrell comedy Semi-Pro, “Groundballs, Jackie. Groundballs, for all these people.”

OK, so maybe I’m changing that line up a bit, but the point remains.

The righty is inducing worm-killers at an impressive level this year—especially given his past track record. Last season Miller’s GB% stood at a pedestrian 39.9%, however this season he appears to have reversed his fortune—raising that his GB% to 48.7%. The almost 10% increase in groundballs has caused his FB% to virtually decrease equally, while the percentage of line-drives he gives up have essentially stayed the same.


Just because Miller is giving up groundballs at a much higher rate, that doesn’t necessarily mean he is pitching better. In fact, groundball pitchers typically give up more overall hits than flyball pitchers, they just don’t go for extra-base hits as often.But what does this mean?

The reason a heightened GB% is a boon for the young starter is because of what happens when the balls hit off of him get into play. Basically, Miller’s ridiculously low opponents’ batting average against, as well as his BABIP (Batting Average on Balls In Play), are indicating that the majority of the contact his gives up is finding one of his fielders.

Take a look for yourself. Here’s those same stats I just mentioned from 2013 until his start Monday:
Picture4The oddity here is that while, one might assume, a lower opponents’ average and/or BABIP for a pitcher might point to the strength of that same teams’ defense that is not the case here. The Atlanta Braves rank near the bottom of the National League in UZR (-4.4) and UZR/150 (-5.1), so it’s not as if there’s an amazing show of defensive prowess going on behind him—other than the one that
Andrelton Simmons puts on nightly.

As common-sense follows, groundballs are typically most effective when they are hit at someone—which is what the grounders hit off of Miller have done this season.

Shelby  Miller

Is this type of high performance sustainable?

As most people know, BABIP and Left On-Base% (LOB%) are very good indicators of whether a pitcher will sustain their success, or if they have just been lucky. If a pitcher has a BABIP or LOB% well deviated from their career average—or the league average for that matter—they will typically see those numbers return to the mean. What this means is that if a pitcher is having a great year, but a low BABIP and/or high strand rate he will typically see the opposite in the future—and the same goes for the other side of a coin. Essentially, it’s a helpful way to tell if a pitcher is over or under performing.

Using this thought, it might be tough for Miller—who is team-controlled through 2018—to sustain the type of success he’s seen so far this season. This is likely because Miller’s BABIP is very low compared to his career .261 AVG and his strand rate is extremely high compared to an already high 79.5% career strand rate. Combine that with a 3.41 FIP and 3.68 xFIP, and we could very well be seeing a pitcher that is currently over-performing.

While Shelby Miller’s 415 career innings are relatively low to base a career average off of, it’s the best possible baseline that can be used. So, is Miller’s performance thus far sustainable? I’d say we might see a little bit of regression as the season goes on.

One thing that has helped Miller thus far is that he has faced some struggling line-ups. Thus far, Miller has faced the Miami Marlins twice, the Philadelphia Phillies twice, the Cincinnati Reds twice, and the Toronto Blue Jays once. The Marlins (91), Phillies (69) and Reds (93) all rank near the bottom of the league in the number in parenthesis, which is wRC+ (an indicator of how well an offense has created runs).

Coming up, Miller’s next five projected starts will be against the Milwaukee Brewers (77), San Francisco Giants (98), Arizona Diamondbacks (99), San Diego Padres (100), and New York Mets (85). It’s a mixture of good and struggling offense that he will face coming up, and it’s an area in his season in which he could really begin to define himself.

So, I’ve said all of that to say this:

While Miller’s individual pitches can lead to an understanding of how he’s had such a good start, he’s also being aided greatly by the groundballs being hit off of him. Those groundballs, which are finding a gloves more often than not, are contributing factors to a low BABIP and a high LOB% that can signal a pitcher who might be outperforming and could regress to their mean in the future.

Although Miller might be prone to a regression, that doesn’t necessarily mean he has to—as Miller still has a lot of time this season to solidify an already strong performance. After all, we are only two months in to this 2015 season.

–Stats taken from Shelby Miller’s profile on, as well as from
–Heat-maps and Spray Chart taken from

Leave a Reply