In NLCS Game 7, the Dodgers won a pennant partly by preventing or hitting home runs. Mookie Betts robbed Freddie Freeman of what would have been a lead-extending dinger in the fifth, and Enrique Hernández and Cody Bellinger launched solo shots in the sixth and seventh. Thus, the Dodgers outhomered the Braves and won 4-3, running the record of teams that have outhomered their opponents during these homer-centric playoffs to 31-4.
But the blasts by Hernández and Bellinger accounted for only two of the Dodgers’ four runs. By win probability, the second-biggest hit in the game was the one that produced the other two: the two-out, two-run single Will Smith smacked in the third.
A few years ago, Smith’s grounder might have been an easy out for the second baseman, and the Braves would have been out of the inning. In 2020, though, there was nobody on the right side of second base to stop the ball. The Braves were shifting against Smith, a right-handed hitter, and Braves second baseman Ozzie Albies was stationed on the left side of second. That alignment may have cost Atlanta a pennant.
A similar decision nearly cost the Rays a pennant, too. In the fifth inning of ALCS Game 6, the Astros’ George Springer—also a right-handed hitter—bounced a grounder past a drawn-in, shifted infield with runners on second and third.
By win probability, Springer’s single was the biggest play of the game: It erased the Rays’ early lead and gave the Astros an edge that they never relinquished. Unlike the Braves, the Rays survived their seventh game, but the shift-beating single helped push them to the brink.
Infield shifts against left-handed hitters have become a common occurrence in the past several seasons. In fact, teams “shifted” on 50.1 percent of pitches thrown to left-handed hitters during the 2020 regular season—in other words, what we think of as “the shift” is now the standard defensive alignment against lefties. The infield shift against right-handed hitters is a more recent innovation. But the shift rate against righties is rising rapidly, too. Since 2017, the shift rate against lefties has more than doubled, but the shift rate against righties has more than quadrupled (to 22.9 percent of pitches in 2020). Shifts against righties are now more common than shifts against lefties were in 2017.
The rationale for the shift is pretty intuitive: Both left-handed and right-handed hitters tend to pull about three-quarters of their ground balls, so it seems smart to position more fielders where the ball is likely to go. Shifts against lefties have been proved to prevent hits on the whole, so it’s not surprising that teams have more than doubled down on that strategy in recent seasons. Nor is it surprising that teams would seek to extend the strategy to right-handed hitters, having experienced such success against hitters who swing from the other side. But here’s something that is surprising, especially considering how often right-handed hitters have been shifted this postseason: In the aggregate, shifts against righties evidently don’t work. And just as those often ill-advised shifts shaped both Championship Series, they may alter the outcome of the Fall Classic.
Anecdotal evidence of the shift’s futility against righties is easy to spot. Although some hitters in the postseason have singled into the teeth of the shift—as Springer did in the ALDS and Mike Zunino did during the ALCS—the shift’s shortcomings are most apparent when righties hit ’em where they ain’t. In the first three rounds of the playoffs, right-handed hitters singled 14 times on grounders to the right side of second when the shift was on. Not all of those hits would have been prevented by a second baseman playing straight up; two were chopped down the line past the first baseman, and one was whistled up the middle. However, the rest probably or possibly could have been played by a second baseman stationed on the right side of second. Here’s a montage of those plausibly preventable singles (minus Smith’s and Springer’s), from most to least recent.
As you’ll see if you watch the whole video, the Dodgers have benefited from and been burned by shifts against righties multiple times, and the Rays have been stung multiple times too. Those two teams ranked first (47.3 percent) and seventh (32.3 percent), respectively, in shift rate against righties during the regular season and have shifted only slightly less often against them in the postseason. (Ironically, the Braves were one of the least-prolific shifters against righties during the regular season, although they roughly doubled their rate in the playoffs.) Thanks in part to the Dodgers and the Rays, the percentage of pitches to righties on which the shift was in effect has been higher in the postseason (24.4 percent through the Championship Series) than it was during the regular season. And it’s likely to climb even more in the World Series, now that the shift-inclined Dodgers and Rays are the only two teams still standing.
That montage I made might be suggestive, but it’s far from conclusive. I could cherry-pick examples of shifts backfiring against left-handed hitters, too, but that wouldn’t change the reality that the robberies outnumber the extra singles that slip through. I could also show you shifts against righties that paid off. In ALDS Game 4, for instance, Springer crushed a ball up the middle that a shifted Tommy La Stella was perfectly positioned to field.
And in ALCS Game 2, Springer smashed a bases-loaded bouncer that might have been a hit if the Rays’ Brandon Lowe hadn’t relocated to the left side of second.
The global effect of the shift is clear in league-level batting averages on ground balls. In 2017, when right-handed hitters were shifted less than 6 percent of the time, righties batted .263 on grounders to the left side of second and .251 on grounders to the right side of second. In 2020, with shifts much more common, righties batted .239 on pulled grounders and .311 on opposite-side grounders. Righties’ results have gotten somewhat worse when they pull grounders (often into the shift) and way better when they hit grounders the other way (often exploiting the shift). Pulled grounders are the much more common kind, so in theory, the math works out in the defense’s favor.
Consider this, though. Entering the World Series, left-handed hitters this postseason had managed a .319 wOBA against standard alignments and only a .289 wOBA against the shift: They were 30 points worse when the shift was on. Right-handed hitters, however, had produced a .343 wOBA with the shift on, and a .305 wOBA against standard alignments: They were 38 points better when the shift was on. (During the regular season, the gap was 21 points in the same direction.) That seems like strong evidence that the shift is a good idea against lefties but a bad idea against righties. Other analysts, including Tom Tango in 2017 and Russell Eassom in 2018 and 2019, have noted this apparent righty/lefty discrepancy before.
Even that doesn’t settle the matter, though, because those figures aren’t comparing the performance of the same hitters when they were and weren’t shifted. It’s possible that the right-handed hitters who did get shifted simply weren’t as good as those who didn’t get shifted. It’s also possible that one of the groups faced a better or worse collection of pitchers. Either or both of those factors could skew the stats. To get a definitive answer, one has to control for the identities of the batters and pitchers to ensure that the talent is consistent across situations and that any difference in results is attributable to the shift.
In September, Tango took a more rigorous look. The noted sabermetrician, who now serves as MLB’s senior data architect, examined the results of bases-empty matchups since 2015 between the same batter-pitcher duos, with and without three fielders on one side of second. He found that left-handed hitters suffered a 24-point drop in wOBA against the same pitchers when the shift was on. Right-handed hitters fared 38 points better with the shift on. When Tango repeated his study but treated batter-pitcher duos as different if the pitcher had changed teams, the takeaway was even clearer: a 25-point drop for lefties against the shift, and a 43-point improvement for righties. Tango also found a similar boost to right-handed hitters’ stats when he looked into 2020 alone.
A few days before Tango published his September study, Russell Carleton came to similar conclusions in an article at Baseball Prospectus. Carleton had previously cast doubt on the shift’s overall effectiveness because of what he called the “walk penalty”: When the shift is on, pitchers walk batters at a higher rate, seemingly because they’re uncomfortable or because they’re avoiding the area of the strike zone where pitches are most likely to turn into opposite-field hits. Those extra free passes cancel out some of the shift’s benefits on batted balls. In September, Carleton clarified that while the walk penalty is a concern against all hitters, the shift is still a net positive for the fielding team when a lefty is at the plate. Against righties, though, the shift is disastrous for defenders.
Both Carleton and Tango found that while lefties are much more likely than usual to strike out when the shift is on—possibly because they’re trying to lift the ball over the infield or otherwise adapt their approach, which is difficult—righties are much more likely than usual to put the ball in play. Righties seem eager to hit against the shift, and according to Carleton, they’re rewarded for that extra contact by raising their baseline rates for hits of every type. Thus, Carleton concluded that shifts against righties “are toxic and cancel out much of the benefit [teams] would have gotten by just sticking to shifting the lefties.”
Sports Info Solutions, a company that provides data on defense and shifts to teams and media clients, doesn’t dispute most of Carleton and Tango’s findings. “We’re generally on the same wavelength as those guys,” says SIS analyst Alex Vigderman. SIS advises that right-handed hitters are shift candidates if they pull at least 80 percent of their grounders and short liners to the left side of second. According to Vigderman, 26 of the 100 righties with at least 100 applicable balls in play during the 2020 regular season met that threshold. However, only seven of those 100 hitters weren’t shifted on any balls in play, which suggests that teams are routinely shifting against the wrong righties.
Vigderman’s analysis determined that in 2020, “only 43 percent of the full shifts on balls in play against righties were against the appropriate players.” He adds that partial shifts, or “strategic” shifts, as they’re labeled by Baseball Savant—alignments that are nonstandard in some way but feature only two infielders on the left side of second—are also largely ineffective against righties. “The short version is that we still contend that shifting right-handed batters isn’t unreasonable, but teams should do it on way fewer batters,” Vigderman says. “So while it might be a useful tactic, at its current usage it’s counterproductive.”
Carleton and Tango are even less sanguine about the shift being useful against righties under any circumstances. According to Carleton’s model, 84.1 percent of actual shifts against left-handed hitters from 2015 to 2019 were analytically correct, but only 1.6 percent of shifts against righties were. And Tango established that even the righties who are shifted most often are unsuited to the tactic. “Across the board, regardless of how often a RHH was shifted, they had a huge gain with the shift,” Tango wrote, asking in conclusion, “Why are clubs shifting RHH at all?”
This week, Tango investigated whether any configuration of three infielders on the left side of second can be beneficial against righties and came up almost empty. If teams are determined to do it, he found, their best bet is to have the third baseman hug the line and position the shortstop about 15 degrees to his left, with the second baseman another 15 to 20 degrees toward the middle. “But if you really have to do something,” he added, “don’t put three infielders to the left side against a RHH so often. Or even maybe ever.”
Shifting against righties would work just fine for defenders if not for one unfortunate fact about baseball: On offense, players run the bases counterclockwise. The rules of the game and the geography of the diamond give rise to a host of logistical challenges that make righties resistant to shifts.
The most glaring problem with shifting against righties is the huge hole it creates on the right side. When the shift is on against a lefty, the lone fielder on the left side of second can stand somewhere in the vicinity of the standard shortstop position, not too far from the second-base bag. There’s still plenty of unguarded territory, but at least the solitary fielder can take up position in the highest-traffic lane for opposite-field grounders. When the shift is on against a righty, though, the lone fielder on the right side of second is the first baseman, who has to stay close enough to the baseline to get back to the bag for force-outs (and, sometimes, to hold the runner). That leaves an even wider swatch of prime real estate for hitters vacant. What’s more, the throw to first is more difficult for fielders on the left side of second, and weak-armed second basemen may have trouble firing to first while moving to their right in distant territory.
There’s reason to think that shifting on righties is an even worse proposition in the postseason, when pitches tend to be faster. The percentages of pitches that are fastballs has been slightly higher this postseason than it was during the regular season, and those playoff fastballs have flown more than 1.5 mph harder. Four-seamers this postseason have averaged 95 mph—a new October record—and four-seamers thrown by Rays and Dodgers hurlers prior to the World Series averaged 96.6 and 95.1, respectively. Against that type of heat, batters are more likely to be late and less likely to pull pitches. During the regular season, righties hit 33.3 percent of their grounders to the right side of second when facing four-seam fastballs 95 mph or faster, compared to only 23.3 percent when facing four-seam fastballs 94 mph or slower.
The Rays’ and Dodgers’ lineups lean left-handed: They placed second and 11th, respectively, in percentage of pitches seen by lefties during the regular season, and entering the World Series they ranked sixth and seventh in the playoffs among the 16 postseason teams. Based on the two pennant winners’ fielding tendencies, though, this series is ripe for shifts on any righties on the rosters. According to SIS, Manuel Margot, Willy Adames, and Hernández are good right-handed candidates for the shift, and Smith is borderline. (The Rays shifted against Hernández in World Series Game 1, but he sent a run-scoring single through the left side anyway.) But if Carleton and Tango are right, even those righties may be bad bets, especially in October’s high-velo environment. Historically, sabermetricians have championed revolutionary shifting over orthodox defense. Now they find themselves in the unfamiliar position of pumping the brakes and recommending a return to tradition.
All of which leaves us with an unsolved sabermetric mystery. The Dodgers and Rays—among the majors’ most data-driven, forward-thinking, early-adopting teams—have presumably seen the studies cited above, and they have access to the same Statcast data Tango and Vigderman do. So why are they—and virtually every other team except the Padres, who shifted against righties least often in 2020—still pursuing a strategy that the stats say is usually self-sabotage?
It’s possible that Tango, Carleton, Vigderman, and Eassom are all missing something. It’s also possible that some teams are implementing shifts against righties in a more optimal fashion than others. Maybe some teams have been blinded by BABIP: Yes, shifts help fielders turn grounders into outs, but it’s easy to overlook their offense-friendly effects on walks, strikeouts, and extra-base hits. And perhaps the front offices that have pushed longest and hardest to overcome coaches’ and players’ past resistance to shifting against lefties are the most susceptible to believing that all forms of shifting work well.
The Dodgers’ and Rays’ wOBAs allowed against shifted righties don’t stand out on the regular-season or postseason leaderboards, but the behavior of the two trendsetting teams gives their less shift-happy opponents pause. “The amount the Dodgers shift is very tough to justify, but also I think they do everything about as well as can be done,” says a front-office executive for a rival team. “So the fact that the Dodgers are this aggressive in shifting leads me to believe they’ve got info we don’t.” Maybe the Dodgers are drawing on proprietary data, or maybe they’ve concluded that pitchers and fielders perform better when they shift on most pitches rather than adjusting their approach and positioning from batter to batter.
Whatever the reason for the Rays’ and Dodgers’ adherence to the rare new-age strategy that seems unsupported by the stats, the World Series, like the ALCS and NLCS, may swing toward one team or the other based on where a grounder rolls or where a second baseman stands. For a few more games, each team will have to hope that the shift against righties giveth more than it taketh away.
