Oh shift—a look back at defensive shifts in 2016

Towards the beginning of this year, I took a look at how the Mariners were utilizing the infield shift and how their batters were faring against it. In 2016, a defensive shift was employed well over a quarter of the time the ball was put in play. That represents a significant increase in shifting over last year—more than 10,000 more plate appearances ended with a defensive shift employed, leaguewide. While slow to adopt this strategy in years past, the Mariners bought into the defensive strategy wholesale this year.

First, a few reminders about the data that we’re working with:

• It only includes plate appearances where a ball is put in play. This means strikeouts, walks, and home runs are not accounted for in the data.
• It is broken into four different splits: all shifts, no shifts, traditional shifts (generally three infielders on one side of the diamond), and non-traditional shifts (situational shifts not covered by the definition of the traditional shift). For the purposes of this article, I’ll be using the data covering all shifts.
• As it’s presented to us now, there’s no way to further split the shift data to determine how often a batter grounds into the shift or hits a fly ball over the shift. That means the data is a little bit muddy and doesn’t give us a full picture of how effective defensive shifts have been.

Back in May, after just a month of play, the Mariners showed that they were committed to utilizing this defensive strategy more than ever. In 2015, they shifted just 8.6% of the time a ball was put in play. By the end of this year, they had utilized a shift 41.8% of the time a ball was put in play against them. That was the fifth highest mark in the majors and a massive change in defensive strategy for the team.

How effective were the Mariners while employing all these defensive shifts? To evaluate their performance, I’ll be comparing batting average on balls in play, ground ball rates, the rate of balls hit to the pull side and up the middle, and overall batting performance, represented by wOBA. For the batted ball direction, I’m including balls hit up the middle since grounders hit that way will often be gobbled up by the second baseman or shortstop if a team is employing a traditional shift. Remember, since this shift data is only taking into account balls in play, the wOBA listed below will not include strikeouts, walks, and home runs.

Here are this year’s results alongside last year’s results:

The Mariners were much more effective while employing a defensive shift when compared to their performance last year. And when compared to their performance this year in a traditional defensive alignment, they were a bit more effective. Much of that improvement can be attributed to an improvement in overall defensive efficiency. Robinson Cano bounced back from an injury-plagued year in 2015 to post a stellar defensive season at the keystone position. The pitching staff also played their part by inducing an above average groundball rate and the fifth highest pull rate while the defense behind them was shifted.

Jerry Dipoto and Scott Servais had a vision for the kind of team they would employ on the field. That vision included a massive increase in the number of defensive shifts utilized by the team. It’s hard to isolate the direct effect shifts had on the Mariners’ defensive effectiveness but the data shows us that this shift strategy benefitted the pitching staff. Overall, the Mariners converted 70.8% of the balls put in play against them into outs and converted 71.6% into outs when utilizing a defensive shift. Those marks were eighth best and tenth best, respectively, in the majors.

Offensively, Mariner batters faced a record setting number of defensive shifts. Opposing teams employed the shift 43.2% of the time a Mariner put the ball in play. That mark led all of baseball by a wide margin and represents the largest number of shifts against a single team in baseball history. That may not be surprising for a team that had a lineup full of left-handed pull hitters and employed a platoon at multiple positions. But despite facing a ridiculous number of shifts, the Mariners actually performed fairly well against the shift:

As you can see, Mariner batters have performed fairly well when facing a shift. But much of that success came despite a .273 BABIP on pulled balls in play (27^th in the majors). When hitting to the opposite field, the Mariners were right around league average. Their real success came from hitting the ball up the middle. Their .321 BABIP when hitting towards center was the 8^th best mark in the majors. Even more encouraging is the difference in wOBA.

Of course, this data isn’t very granular. Those BABIPs I just referenced are team marks that include all balls in play hit to a particular section of the field, whether there was a shift employed or not. We can’t know exactly how the team performed on groundballs pulled into the shift for example. Those kind of filters for team split data aren’t readily available. We do have that kind of data granularity for individual players however, and I’ll be covering some of those individual splits later this week.

Edit: Jesse Smith, manager of analytics for the Mariners, tweeted something that I failed to mention in the original version of this article.

Average wRC+ of the 30 players who faced the most shifts in 2016 was 119 (a .350 wOBA). These are players like David Ortiz, Anthony Rizzo, and Kyle Seager. With this context in mind, the Mariners defensive efficiency while employing a shift becomes even more impressive. The data isn’t granular enough to tell us who the Mariner shifted against (it certainly included some of these players), but quality of hitters that see a shift is definitely high.