Sabermetrics

The Nationals Secret Weapon

When the season began two months ago, Nationals general manager exclaimed his excitement about his team’s bullpen in an interview with Buster Olney on the Baseball Tonight Podcast:

“We like our bullpen. It’s something that we confident about going into the ‘18 season and we’re confident that this is a group that can finish off games for us when our starters give us a quality start like they often do.”

In the early going, the Nationals pen validated their GM’s confidence as the group has ranked in the top 10 in the MLB in ERA-, FIP, K%, and BB%. That was before they acquired Kelvin Herrera’s 1.03 ERA earlier this week. While Sean Doolittle, Ryan Madson, and Sammy Solis have all been solid for Washington, it is a non-roster invitee that is anchoring their much-improved bullpen.

On May 24th, the Nationals selected the contract of 30-year-old journeyman reliever Justin Miller from Triple-A Syracuse. Miller’s contract not only included a June 15th opt-out provision, but he was also dominating hitters in the upper minors. With both of these considerations working in tandem, it was the perfect time for the Nationals to give Miller a shot.

At the time of Miller’s call-up, he had pitched 13.2 scoreless innings at Triple-A, striking out 23 while walking just 3. Given Miller’s lifetime 4.99 ERA and his recent struggles in upper minors for the Angels organization, it is safe to say that the expectations for Miller were low. What Miller has done since then, however, has been nothing short of remarkable.

In 11 appearances, Miller has struck out 42.3% of the batters he has faced while walking just 3.9%. That 38.5% K-BB% ratio leads the MLB since he joined the Nationals. Small sample size, luck-related success does not explain Miller’s renaissance as his 1.84 ERA is significantly higher than his 1.39 FIP thus far. For those keener on Statcast as their source for predictive statistics, Miller checks out here as well. Miller’s .188 wOBA against is just slightly lower than his expected wOBA (xwOBA) of .202.

While all the traditional indicators seem to suggest that Miller will not completely fall apart after his hot start, it is imperative that we find a reason for this success, an adjustment to justify better results.

One of the least quantifiable attributes a pitcher can have is his deception. This is why despite continuing to employ a standard fastball/slider two-pitch mix at his career-average velocities, Miller has gotten drastically different results.

For Miller, this meant taking an already semi-closed stance and pushing it to the extreme. As you will see below, Miller’s left foot is much closer to third base than it was the last time we saw him pitch in the big leagues.

Compare the following video clips, the first being from Miller’s previous major league stint back in 2016.

Justin Miller (2016):



Justin Miller (2018):


While Miller has always had a closed stance,  by extending his front foot even further towards third base, he is able to hide the ball longer. Miller spoke about this added deception in a recent interview:

“I kind of wanted to see it from a hitter’s angle, what he does, he’s got his back turned toward you. As a right-handed hitter, the ball comes out of nowhere.”

By throwing from the third base side of the mound, the ball appears as if it is coming from behind a right-hitters when released. This has made righties noticeably uncomfortable against Miller, as evidenced by an abysmal .152 / .200 / .219 slash line against him so far this season. But while Miller has tailored his motion to dominate righties, which may prove vital for situational matchups in October, this subtle mechanical tweak has also granted him another intangible benefit: confidence.

Of all pitchers who have thrown at least 100 pitches this season, Miller’s 59.5% strike rate leads all of baseball. Batters are also swinging at those pitches less frequently (just 58% compared to 70% in 2016). Naturally, this means Miller has been receiving more called strikes, which helps to explains his career-high F-Strike% (first pitch strike percentage) of 63.5%.

First Pitch Strike Locations:

Justin Miller (First Pitch)
This heat map displays Justin Miller’s location on the first pitch of at-bats.

What is notable here is not that he is able to locate first-pitch strikes, but that he is able to do so while mostly avoiding the heart of the zone. Miller not only locates consistency in the lower half of the zone, he does so with an unpredictable pitch mix. Despite only having two pitches, he is able to upset the timing of hitters by consistently locating both pitches in any count. In fact, Miller has thrown his slider more frequently on the first pitch of at-bats this season than his fastball. When opposing hitters do offer at first pitch sliders from Miller, they have a wOBA of exactly zero, in large part because they are too frozen to swing at all.

While Miller has shown an ability to throw quality strikes early in the count, it is his unpredictability, namely a unique tendency to pitch backward, that has kept hitters off balance this season. Once ahead Miller gains the advantage in the count, he will work at the edge of the zone with both his fastball and slider.

Fastball and Slider Locations When Ahead in the Count)

Justin Miller (Putaway Pitches)
This heat map illustrates Miller’s ability to work the edges of the zone once he gets ahead in the count.

By getting ahead in the count, Miller forces opposing hitters to get more protective and offer at pitches that they otherwise would lay off of if the count was in their favor. This explains why opposing hitters are swinging at pitches outside the zone 35.7% of the time against Miller, a rate roughly 10% higher than Miller’s previous career-high. When hitters swing at pitches that are outside the zone, they rarely have success, but this is particularly evident when hitters flail at Miller’s out-of-the-zone offerings. Hitters facing Miller make contact on pitches outside the zone just 51.2% of the time and are managing a meager .038 wOBA when behind in the count.

Justin Miller’s success does not come without concerns. For example, his heavy flyball rate and low groundball rate comprises a batted ball profile that generally portends regression. But just because the baseball world applies the black-and-white, no-nuanced thinking that flyballs are bad and groundballs are good, does not mean that there aren’t exceptions.

For one, not all flyballs are the same and the fact that the average flyball/line drive exit velocity is just 88.7 (14th lowest in the MLB), indicates that his low HR/FB ratio may not regress to the mean as much as many might think.

I am not here to say that Justin Miller is an elite relief pitcher that is going to anchor the Nationals bullpen come October. What I am saying is that the Nationals, at the very least, may have found themselves one of the best relievers in the game when it comes to retiring right-handed hitters.

While the numbers support the notion that Miller’s career revival may be sustainable, Miller’s incredible statistics is not what attracts readers to these stories. Instead, it is the perseverance of a man hanging on by a thread that could very likely be pitching high leverage innings in October.

Advertisements
Tagged , , , , , ,

About Traven Tapson

I am a recent graduate from Claremont McKenna College pursuing a career in baseball operations for an MLB team. I am fascinated by the analytical side of baseball and use this blog as a platform to share my insights and knowledge with those who share my curiosities.
View all posts by Traven Tapson →