Identifying the most- and least-shiftable teams
When the Giants signed Nori Aoki to a small contract, it looked like a sound investment even without digging too deep. The Giants wanted to add another regular outfielder. They found one for cheap, who can play adequate defense while getting on base more often than average. While it’s true that Aoki doesn’t hit for much power, he’s valuable despite that, just as David Ortiz is valuable even though he doesn’t steal bases. All right, so it’s not the exact same thing, but you see where I’m going. Focus on what a player can do, and so on and so forth.
Yet there’s also interest in the details. Aoki makes a lot of contact, which is one of the many ways he’s different from, say, Michael Morse. He puts the ball in play, just like Casey McGehee puts the ball in play, and the Giants have put together a higher-contact offense as opposed to a higher-power offense. But there’s also another thing about Aoki: he might be the least-shiftable hitter in the major leagues. He sprays the ball all over the place, unpredictably, which makes him tricky to defend. Dividing the field into thirds (instead of halves), the average hitter pulls about 54 percent of his groundballs. Aoki has pulled just 34 percent of his groundballs, meaning there’s no sense in moving your infielders around. He’ll hit the ball where he hits the ball.
As you certainly know, teams are shifting more and more often. Maybe you’re familiar with the numbers; maybe you’ve just had a sense. Here are some numbers, if you want them. As recently as 2011, there were just shy of 2,500 balls put in play with a shift on. The next year, that number went up 94 percent. Then that number went up 79 percent. Then that number went up 63 percent. There were 564 percent as many shifts in 2014 as there were in 2011. Shifts are even rising dramatically against right-handed hitters, which is particularly unconventional.
Two points. One, shifts have gone way up. Two, they’re going to continue to do that. Why wouldn’t they? Shifts make sense. Put people where the ball goes and the ball won’t get through as much.
So there’s been talk about counter-shifts. The obvious maneuver is to put down a bunt, but for whatever reason, hitters have been reluctant to do this very often. And then there’s the simple idea of just using the whole field. The problem being, a hitter can’t just start doing that. A hitter’s swing tends to be a hitter’s swing.
I should work to get closer to the point. Jumping ahead: I got curious to know which teams, for 2015, look to be the most- and least-shiftable. Now, I don’t think teams are built specifically to counter the shift, but that doesn’t mean we can’t pay attention. The 2015 offenses have mostly been built. Who looks the most susceptible to the infield shift? Who, on the other hand, looks the least? Where will there and won’t there be extra hits prevented?
At FanGraphs, we have team-by-team depth charts, maintained by a number of the site’s authors. Using those charts, we can get an idea of which players will bat, and how often. Though they aren’t perfect, of course, and some situations will change between now and opening day, such changes should be relatively minor. Now then, the other part: I grabbed data from 2012 through 2014 and calculated, for every hitter, his rate of pulled groundballs. Then it was just a matter of putting those rates together for each team and weighting them by projected plate appearances. This yields an expected team rate of pulled groundballs.
In a handful of cases, I didn’t have enough data to work with. For example, Rusney Castillo and Joc Pederson are projected to play very often in 2015, but they’ve hardly played yet in the majors, so I can’t examine their respective performance records. In cases like this, I just had to eliminate the player from consideration, but this should make little difference. We’ll do what we can with the numbers we’ve got.
The result of everything:
|Team||Projected Pulled GB%|
There’s not a ton of variation. The highest number is about 61 percent, the lowest about 50 percent. But somewhere in there is a tipping point, beyond which a shift will be deployed. You can see the Marlins as the least-shiftable team, by this method. At the top, two members of the AL East.
Regarding the Rays, what I don’t have is good data for Steven Souza, who’s likely to be a regular outfielder. From what I can tell, Souza doesn’t appear to have exceptional pull or spray tendencies. Eight of the remaining 11 players expected to hit most often have posted pulled grounder rates of at least 60 percent. Only James Loney comes in a little below the league average. Evan Longoria leads the way, at 67 percent. Barely behind him, new acquisitions Rene Rivera and Asdrubal Cabrera.
The Yankees aren’t a surprise at all. The AL East has developed a reputation for being shift-heavy, partly because there are just some shiftable teams. Credit to Brett Gardner and Didi Gregorius: they’ll hit the ball the other way. But, all right, let’s say this: one standard deviation above the average, for pulled grounder rate, is just under 63 percent. The Yankees have seven regular or semi-regular players north of that line. Both Chris Young and Mark Teixeira are over 70 percent. You know about the problems Brian McCann had. The Yankees have put together a flexible offense that can swing a lot of left-handed bats in home games. This is one of the consequences. Those lefty pull hitters will also lose singles to shifted infielders.
As for the Indians, Carlos Santana and Brandon Moss are extremely shiftable. Nick Swisher only a little less so. On the Astros, Chris Carter is almost an automatic shift. Several players come in above 60 percent. The Blue Jays are going to watch opponents stack the infield against Edwin Encarnacion and Justin Smoak.
At the other side, we have a different picture. As expected, there are the Giants, fourth from the bottom. Brandon Belt is the only realistic shift candidate, and guys like Aoki, McGehee, and Joe Panik will just shoot the ball anywhere. Below the Giants, the Nationals don’t have a single regular at or above 60 percent. The highest would be Jayson Werth’s 57 percent, and then Anthony Rendon, at the MLB average.
Lower still are the Brewers. Sometimes it does make sense to shift against Aramis Ramirez. Against the rest of the lineup, not so much. Even Ryan Braun has learned more about the opposite field, although it’s worth wondering how much of that has been a side effect of his injuries. Adam Lind, surprisingly, doesn’t pull the ball on the ground even half the time.
Lastly, the Marlins. And the Marlins might exemplify why an overall team metric might not be the best idea. Mostly it works, but the Marlins’ offense seems a bit bipolar, with some good shift candidates and some awful shift candidates. Teams will want to shift against Jarrod Saltalamacchia. They’ll be at least tempted to shift against Marcell Ozuna and Giancarlo Stanton. But, Christian Yelich is like Nori Aoki with more pop in his bat. Martin Prado barely even pulls 40 percent of his grounders. Dee Gordon’s even lower than Prado is. The Marlins have three of the least-shiftable players in baseball in their projected starting lineup. Not bad, considering the lineup has eight players and a pitcher. The Marlins will still lose some hits when an opportunistic infielder robs Saltalamacchia or Ozuna, but with the three just named, there’s not a lot you can prepare for. The hitter reacts to the pitcher, and the fielders react to the hitter.
No one builds an offense specifically to work against the shift. It’s just not that important, relative to drawing walks and hitting the ball out of the yard. But some offenses are less shiftable than others, and we can see that here, with the Rays and Yankees in particular lined up to lose a bunch of singles. That’s just a part of the new baseball reality. And at the other end, while the less-shiftable teams presumably weren’t put together with that directly in mind, no one’s going to complain. Every year, it’s getting harder and harder to hit. A guy like Nori Aoki is at least partially immune.