How much have double plays hurt the Yankees offense?

We all know the script. The Yankees have a runner on first with fewer than two outs, and almost as if it is ordained by the baseball gods, the next batter hits into a groundball double play. At this point, along with bullpen meltdowns, defensive miscues, and triple plays, the GIDP (ground into double play) has become one of the quintessential plays for the 2021 Yankees. In just 122 games, they have hit into an AL-leading 112 of them, a little under one per game. That’s ... a lot of erased baserunners, which certainly has had a very negative effect on the team’s ability to score runs over the course of this season.

But how much value did they lose? Well, there are a few different ways to try to measure it. We can first start by looking at the practical effect the lost baserunners would have had on the team’s OPS.

Let me explain what I mean by that with an example. In the first inning of a hypothetical game, DJ LeMahieu singles to right field, but is immediately erased when Aaron Judge grounds into a double play, and Joey Gallo strikes out; in the second inning, Giancarlo Stanton flies out to right field, Anthony Rizzo lines out to third, and Gary Sánchez grounds out to second base. Despite the fact that the team had a batting average of .333 in the first inning and .000 in the second, they left the bases empty in both innings; functionally, there was no difference.

What does that look like on a bigger scale? Well, as of yesterday afternoon, the Yankees offense on the season had recorded 623 singles, 162 doubles, 9 triples, 149 home runs, and 495 walks, for a slash line of .237/.327/.395, which gives us a perfectly league-average OPS of .722 (100 OPS+). If we remove 112 of those walks, turning them into outs instead, to account for the baserunners removed on the double plays, that OPS drops to .705 (98 OPS+). That is identical to the 58-64 Detroit Tigers* this season, albeit while not doing the same adjustment to their OPS. From a purely statistical standpoint, then, the double plays have done damage to the team’s offense, but not nearly to the extent that it might appear.

*The Tigers have actually slightly outscored the Yankees by runs per game, 4.35 to 4.24, despite their lower OPS. Go figure.

Of course, OPS is only one way to judge value. We also have Win Probability Added (WPA), a game-level stat, and championship Win Probability Added (cWPA), which judges how much of an impact a particular event has on a team’s ability to win the championship. While I will spare you the full spreadsheet of data — it is, after all, 112 lines long — I can tell you that the sum total cWPA of the Yankees’ double plays has amounted to 5.73 percent, a fairly hefty sum considering most plays only provide one or two hundredths of a percentage point.

On an individual game basis, 21 of the 112 double plays (18.75 percent) resulted in a WPA swing of 10 percent or more; four of them were above 20 percent. Additionally, four of them — DJ LeMahieu off Boston’s Matt Barnes on June 25th, Gio Urshela off Oakland’s Lou Trivino on June 18th, Giancarlo Stanton off Houston’s Ryan Pressly on May 6th, and Aaron Judge off Seattle’s Kendall Graveman on July 8th — directly ended their respective games.

No matter how you put it, the double plays have stolen quite a bit of value from the Yankees’ lineup in a year that they needed as much offensive output as they could scrape together. Although the offense has improved since the trade deadline, posting a 109 wRC+ in that time, it’s fair to pin the extreme amount of double plays as one of the many reasons for the hole that the Yankees created for themselves across the season’s first few months.