Kyle Barraclough’s flawed dominance

The one inherent flaw in regards to DC Comics’ luminary Superman is his lack of flaws. He has every power imaginable. In order to beat him, you would first have to get your hands on a precious stone from his home planet, which has long since been destroyed. Who’s got time for that? His perfection renders him unrelatable.

Batman, on the other hand, is a billionaire with a bunch of fun toys and expertise in hand-to-hand combat, but he’s still a mortal human like you and me. Spiderman can climb on walls and swing through the air, sure, but he’s also a teenager who just wants a normal life, only to have "great responsibility" thrust upon him. These are the characters we relate to, flawed in their greatness.

Three things are true of Marlins’ relief pitcher Kyle Barraclough (pronounced bear-claw); he has an 80-grade name, he’s one of the best relievers in baseball, and he has a gigantic flaw. Observe some notable statistics from last season and where his marks rank among relief pitchers who threw at least 40 innings in 2016. You’ll likely be impressed, save for one obvious problem — he walks everybody.

Of the 176 relievers with 40 or more innings, Barraclough was 174th in walk rate. Despite that poor showing, he was top-ten in both FIP and fWAR. With fWAR being a FIP-based calculation of WAR, and FIP being composed of only elements a pitcher can control — home runs, strikeouts, and walks — it was his exceptional strikeout rate alongside having only allowed one home run in 72 ²⁄₃ innings that explains those numbers.

Back in June, Jeff Sullivan of FanGraphs showed that Barraclough had the lowest rate of pitches put into play of any pitcher in baseball. The difficulty hitters had making contact would last, as according to PITCHf/x he would finish the season with a contact rate nearly 12 percentage points below the league average at 66.2 percent. At the time of that article, Barraclough had a walk rate of 17.7 percent, thanks in large part to a 22.6 percent walk rate in May. Here, you can see that his free passes had a big spike early in the year but did settle down a little bit over the course of the season:

FanGraphs

He still ended the season with a walk rate of 14.4 percent, well above the 9 percent league average for relief pitchers, so it’s not like he became a control artist after one bad month. Still, there is some reason to believe that he’s better than the 5.45 walks per nine that he showed in 2016. His walk rate settled down later in the year, and at only 26 years old it’s conceivable that the improvement will stick.

So what made Barraclough able to dominate hitters despite this failing in his skill set?

First and foremost, he throws hard, averaging 95.5 mph on the four-seam fastball that he uses 54.5 percent of the time. He compliments that with a dynamic slider that he throws 41.9 percent of the time, which checks in at 82.2 mph and generates a healthy 21.4 percent swinging strike rate. The difference in velocity between the four-seamer and slider is huge, made effective by the fact that Barraclough is able to throw from a relatively consistent release point for both pitches.

When he spoke to David Laurila of FanGraphs in August of last year, Barraclough echoed the importance of a consistent release point for the two pitches:

I think my slider has always been 10-11 mph [slower] than my fastball. Sometimes it might be more, but I don’t really pay a lot of attention to that. I’m more focused on the movement, and how much it mirrors my fastball out of the hand. I’m not conscious of, ‘Oh, I need to make it slower,’ or ‘I need to make it harder.’

It’s certainly nice to get a visual representation of his release point from Baseball Savant, but thanks to the exceptional new work on pitch tunnels done by Baseball Prospectus, we can really dig in to how these two offerings compliment each other.

First, let’s lay out the data for the pitch tunnel pairs of Barraclough’s four-seam fastball and slider. It should be noted that Barraclough does throw a changeup as well, but just 3.6 percent of the time, so we’re going to concentrate on how the fastball and slider work together, as they are the keys to his success.

Setting a minimum of 50 pitches for each pitch type pair, let’s parse these numbers.

Referring to the column “Release Diff,” we can check on our assumption that Barraclough maintains consistent release points between the two pitches. The lower the number there, the more similar the release point. He ranks 86th out of 205 pitchers going fastball-to-slider and 67th out of 203 pitchers when going slider-to-fastball. When pairing a fastball with another fastball Barraclough is fourth(!!) out of 431 pitchers, and when throwing back-to-back sliders he’s 28th out of 186 pitchers. There’s very little variation when he’s pumping fastballs or sliders back-to-back, and when he’s alternating fastball-slider or vice versa, his release consistency appears to be good, just not in the top tier of baseball.

The next noteworthy numbers for Barraclough are in regards to his “Post-tunnel Break.” This, to quote Baseball Prospectus, “tells us how much each spin-induced movement is generated on each pitch between the tunnel point and home plate.” So this isn’t measuring total difference in break, it’s measuring the difference in break from the tunnel point, which is 23.8 inches from home plate. In both alternating pitch type examples, when going fastball to slider or slider to fastball, Barraclough has the fourth-highest break differential in baseball — in both cases trailing Cam Bedrosian, Yu Darvish, and teammate AJ Ramos.

The excellent break differential is fantastic, but if the break starts too early, the amount of post-tunnel break will matter less because the batter has more time to identify the pitch. In the far right column on the pitch tunnel table you can see “Release:Tunnel.” This, again to quote Baseball Prospectus, “shows us the ratio of a pitcher’s release differential to their tunnel differential.” A lower number here means that the difference between the two pitches at the tunnel point is smaller, theoretically pointing to more deception from the pitcher, helping him to disguise pitch types at the point of release.

When going from his fastball to his slider Barraclough ranks 51st out of 205, and when going in the opposite direction he ranks 30th out of 203. Combining these above-average release-to-tunnel ratios, indicating good deception, with the elite post-tunnel break seems to explain why hitters have such trouble making contact against him.

If we take the macro view to include all pitchers and all pitch types (minimum 500 Pitch Pairs), out of 385 pitchers Barraclough ranks 15th in release-to-tunnel ratio and 46th in release difference. For pitchers who aim to vary their release points ranking, that high is not important, but for someone whose stated objective is that his slider “mirrors [his] fastball out of the hand,” it’s a strong showing and indicates that he’s succeeding in his goal.

Barraclough has one glaring weakness that would sink a lot of pitchers, but he's so good in so many other ways that it’s something he’s able to overcome. You can live with the high walk rate if you also have a high strikeout rate complimenting an exceptionally low contact rate, an above-average ground-ball rate combined with a below-average fly-ball rate, and a 95-mph fastball that shares a relatively consistent release point but large velocity differential with a wipeout slider.

Kyle Barraclough is already one of the best relief pitchers in baseball despite this significant blemish. Cutting even a couple percentage points off his walk rate would make him exceedingly more dangerous than he already is; if he’s able to do that, watch out — the sky’s the limit.