The Red Sox’s Barrel Blast

We are officially in the Statcast era of major league baseball. The Statcast tool offers unprecedented information regarding player movement and ball flight path and speed. The access to this information has undoubtedly changed the way we watch and think about the game. Highlights of home runs now come equipped with the requisite exit velocity and launch angle of the ball off the bat, and diving catches are now presented alongside the route efficiency of the fielder. Just watch a game on MLB.tv and between innings you will get to (repeatedly) see a rotation of three highlights augmented with Statcast information: a 504ft. Giancarlo Stanton home run in Coors Field, a 105 mph throw from deep left field by Aaron Hicks, and a Mike Trout home run to dead center. Seriously though, why, with all of the highlights available to the folks at MLB.tv, am I forced to watch those same three highlights interspersed among three mind-numbing segments from the Chatting Cage? In case you didn’t know, Nolan Arenado loved Remember the Titans. But I digress. While a lot of the Statcast information has been primarily descriptive in nature, Mike Petriello and his colleagues within the Statcast lab have been working to develop metrics that aim to infer something about player process.

One of the first metrics released by the Statcast lab team is Barrels. Barrels combines exit velocity and launch angle to determine high-value batted balls: those that were ‘barreled-up’. The basic idea is that if a ball is hit hard and within a certain range of angle it is likely to leave the yard, or at least go for extra bases. Either piece of information on its own is not enough. You can hit a ball hard (i.e., high exit velocity) but if you hit it too high or into the ground you likely won’t have much to show for it. Conversely, you can hit a ball at a typically strong launch angle, but if it is hit weakly it will probably find a defender’s glove. As determined by the Statcast lab members, a barrel is a batted ball whose exit velocity and launch angle have typically resulted in at least a .500 batting average and 1.500 slugging percentage. In Statcast terms these are balls that were struck with an exit velocity of at least 98 mph and a launch angle between 26 and 30 degrees (and expanding as the exit velocity increases). I recommend you check out the linked article above and the MLB.com glossary entry for barreled ball to see a visual of this idea.

As an example, this is what a barreled-up ball looks like:

Those are the details, but the take away note here is that barrels are what batters want. They are well-struck balls that typically end in good results. Those nice folks at the Statcast lab have even created leaderboards for us to find which batters are the best at barreling up the ball (and which pitchers are best at limiting barreled balls). This can be defined per batted-ball event (Brls/BBE) or per plate appearance (Brls/PA). This first limits things to when the batter puts the ball in play, while the second factors in things that might come with trying to crush balls every time in the batter’s box, namely more strikeouts and fewer walks. I played around with the leaderboards to see how Red Sox batters fared in barreling up the ball.

First, I limited the 2016 dataset to only include batters who had at least 60 batted ball events. I did this because 60 is a reasonable sample… ah who am I kidding, I did it because this is the highest minimum that includes Andrew Benintendi. Using the 60 BBE minimum leaves 439 players. The average Brl/BBE for this group was 6.38 percent (SD = 4.22). These barrels do not happen all that often; they are somewhat special events. Top barreler in this group last year was David Wright at 20.3 percent. I was as surprised as you are right now. Second best was Yankees’ hall of fame catcher Gary Sanchez at 18.8 percent. Shifting to things to a PA basis basically halves the average rate (3.74, SD = 2.30) and shakes up the top spots. Oakland’s Khris Davis was best at 10.7 percent and Sanchez dropped to third best (10.5 percent).

How about the Red Sox hitters? Interestingly, despite having the best offense in baseball last year, only four of the 14 Red Sox hitters who accumulated at least 100 PA had a Brl/BBE rate above the average (of my sample) and two of them are no longer with the team. David Ortiz (15.8 percent), Jackie Bradley Jr. (9.8), Hanley Ramirez (9.5), and Travis Shaw (8.8) were the above-average barrelers. Yet another measure showing how ridiculous David Ortiz was last year. Last year, old friend Jake Peavy was on the wrong end of Big Papi’s penchant for barreling baseballs:

While Shaw and Ortiz are gone, the good news is that their replacement, Mitch Moreland (10.7) was also an above-average per-BBE barreler last year. Putting things on a per PA rate adds Mookie Betts and Xander Bogaerts into above-average territory.

Generally, the order of the list falls in line with what we should expect. Power guys up top, weaker, slappy hitters at the bottom. However, there are a couple of players worth noting. I was surprised to see Mookie Betts be so far down the list (and below average in Brls/BBE). I mean, he hit 31 home runs last year, but evidently they were not all barreled. Perhaps there is some Green Monster magic afoot here. Also, Mookie hit a lot of doubles and many of them are likely a result of his speed more than his barreling up the ball. Referring back to the Statcast article there is a plot of slugging percentage on batted balls shown as a function of exit velocity and launch angle that has a large, sweeping area of decent slugging just below the barrel area. Petriello says this area has “…some extra-base hits, but mostly the kind of doubles and triples that are bloops, well-placed or generated by speed…”. I suspect Mookie has a bunch of contact in that region. Also, he fared well on this measure when considering the rate per plate appearance. Dustin Pedroia showing up toward the bottom of the list might be a surprise given his great year last year, but his power numbers were down, and he reported making a conscious effort to not focus on strength during last offseason. So it makes some sense. Finally, at the bottom is poor Ryan Hanigan, who went barrelless last year; a rough end to his time in Boston.

It is important to note that this measure, barrels, is supposed to be a process metric. So it is great that a player gets a barrel here and there – sorry, Ryan – but we care more that it reveals the players who have a swing (and approach) that consistently generates velocity and balls in the air. For now the Statcast data are only available for the 2016 and 2015 seasons so we are limited in seeing consistency over many years, but I found that the year-to-year correlation in Brls/BBE was 0.747 and 0.742 for Brls/PA (for players who had at least 60 batted ball events in each of 2016 and 2015). Just a quick note that Giancarlo Stanton led the way in 2015 with almost a third of his batted balls getting classified as barrels. Maybe MLB.tv has a few more highlights of him lying around on the cutting room floor. The magnitude of the year-to-year correlation suggests this measure is picking up on something more than randomness or luck, which I suppose is not all that surprising given that characteristics of a player’s swing (plane, strength) is not likely to change dramatically from one year to the next.

The year-to-year consistency was evident for Red Sox players (positive numbers mean the rate was higher in 2016):

Some reasonably large changes at the extremes, but all centered around a team-level change of about 0, which was also true for the full sample.

All told, while the 2016 Red Sox did not exactly light the world on-fire with their ability to barrel up the ball (at least according to this Statcast metric), they were still a dominant offense. What they perhaps lacked in barreled-up triples and dingers, they made up for with a steady barrage of doubles, singles and walks. David Ortiz and his barreling ability will certainly be missed, but Mitch Moreland will help pick up some of the slack with his strong ability in getting the barrel to the ball. Also, Pablo Sandoval, who actually managed to barrel-up fairly often (6.3 percent of BBE, 4.20 percent of PA) in his disastrous 2015 season, should give the team an offensive boost from the third base position. On the whole, lots of barrels or few barrels, I expect the Red Sox will have a top-five offense again in 2017.

Photo by Kim Klement/USA Today Sports Images