At its core, baseball’s defensive revolution has been about positioning fielders in places where the ball is most likely to be hit, an idea so simple and sensible that it seems incredible that teams didn’t adopt it earlier. As the Astros’ Sig Mejdal says, “Why weren’t teams positioning their infielders different half a decade ago? I don’t know. The data was all there.”

Part of it the answer is risk aversion, as Mejdal also suggests. But repositioning fielders based on batter tendencies was considered so risky only because the standard alignment was ascendant for so long. Why wasn’t it obvious to everyone—players, coaches, executives, authors—ages ago that more mobile fielders could be a big help? Maybe because embracing the shift requires all of us to do something difficult: to redefine—or at least loosen our deeply ingrained definitions of—what it means to play a position.

On a basic level, we distinguish between most defenders based on their physical locations on the field. If a baseball novice asked you to explain what a shortstop does, you might start by saying that he’s “the fielder who stands between the second baseman and third baseman, to the third-base side of the second-base bag.” That explanation would have worked well a decade ago, but today there are too many exceptions to the classic alignment. Many shortstops move to the right side of second with a pull hitter at the plate, ceding their previous spot to the third baseman. In those cases, the novice could be excused for assuming that a change in positioning corresponds with a change in position.

Baseball’s box score has no patience for philosophical questions about the nature of positionhood. As far as the box score is concerned, the original shortstop’s inherent shortstop-ness travels with him wherever he roams. Aside from shrinking hit totals and batting averages, then, the venerable medium through which many fans follow games offers no indication of a fundamental (don’t say shift don’t say shift) development in the way those games are played. As author and longtime Yankees PR Director Marty Appel observed in a recent Facebook discussion thread,

Shifting so regularly is an enormous change for baseball. It also challenges our future readings of box scores, where a 5-3 putout no longer tells us with certainty approximately where the ball was hit.

Of course, there are other available data points that allow us reconstruct each play more accurately than a box score ever could, including both the horizontal and vertical angles at which the ball came off the bat and the x, y, field coordinates at which it was first touched by a defender. Soon, that information will get even more granular. Still, much of it is proprietary or difficult to process, and angles and coordinates lack the simplicity of the traditional scoring system. So it’s fair to wonder whether the box score itself should change in some way to reflect the new reality, a question asked of me most recently by BP reader Dennis Sterner:

Everybody is talking about the use of shifts this year. And in years past there has been some discussion about The Shift and position eligibility for fantasy purposes. Given the increase in use of shifts this year, here’s my idea for an article that I don’t think anyone has written—why not interview official scorers and anyone else who is responsible for what is in official box scores and ask them why they are ignoring The Shift when they record positions played in box scores.

I’d like to hear an official response to why Derek Jeter is not recorded as having played 3B when he is the farthest to the left of all four Yankees infielders when The Shift is on.

Although the use of unorthodox defensive alignments has exploded in the several few seasons, it’s not a new phenomenon. Sixty years ago last week, the Reds debuted the so-called “Birdie Tebbetts shift,” in which one infielder moves to the outfield to defend against a fly ball hitter who tends to pull his grounders. In that instance, Reds skipper Tebbetts tried to preserve a late lead over St. Louis by replacing starting shortstop Roy McMillan with Nino Escalera, a utility player who appeared at several positions during his short-lived big-league career. However, Escalera didn’t stand where McMillan had; with Stan Musial due up, he trotted out to right-center field. Musial struck out, and Escalera gave way to Rocky Bridges before the next inning. Although he never stood in the infield, Escalera was listed as a shortstop in the box score, and he still qualifies as the last left-handed thrower to play shortstop. More recent examples of the same alignment have been treated the same way.

Major League Baseball’s official rulebook explains why Escalera could be called a shortstop—and addresses Dennis’ question—in a comment on Rule 10.03(a):

When a player does not exchange positions with another fielder but is merely placed in a different spot for a particular batter (for example, if a second baseman goes to the outfield to form a four-man outfield, or if a third baseman moves to a position between the shortstop and second baseman), the official scorer should not list this as a new position.

According to Official Baseball Historian John Thorn, that comment dates to the 1940s, when teams played deep infields or four-man outfields against lumbering Ernie Lombardi, and when Indians manager Lou Boudreau unveiled the “Ted Williams Shift.” In most cases, Rule 10.03(a) makes classifying fielders an easy call. According to Major League Baseball’s Senior Vice President of Club Relations, Phyllis Merhige, who oversees official scoring, there hasn’t been any confusion among scorers about when a player should be listed at a different position. “It’s something that I’m putting on the agenda for next year’s official scorers meeting, to see whether anybody wanted to raise anything,” she says. “But I talked to the Elias Bureau about it, too, and they haven’t heard anything.”

That doesn’t mean they aren’t thinking about it, though, says Elias Director of Research John Labombarda. “We have had discussions in the office about shifts and what to do with fielding statistics, but we have not come to any concrete decision on if we’re going to make any kind of change.”

As Labombarda points out, recording shifts from batter to batter within box scores would require creating entirely new positional designations, which he considers “very impractical.” A shifted fielder, he notes, is “not playing any real position. When Derek Jeter is shaded to the second base side and he fields a ground ball by a left-handed batter, are you going to say it was a four-three instead of a six-three because he was playing in the second base position?

To avoid having to answer that kind of question, scorers simply do things the way they did before teams embraced newfangled fielding. “There is no tweaking that goes on during a game,” Merhige says. “It’s what’s on the lineup card.”

Nor is the process necessarily due for an update. “Where players just shift, I think you’d have to stick to it this way,” she adds. “I think there can be some discussion to make sure people are clear about what to do, but I don’t anticipate any changes.”

However, there are occasional edge cases that cause scoring quandaries. As Daniel Rathman mentioned in an edition of What You Need to Know earlier this month, the Rays employed a five-man infield against the Yankees in the bottom of the 13th inning of a game that began on May 2nd. With runners on second and third with one out for batter Brett Gardner, Rays manager Joe Maddon moved right fielder Wil Myers to first base. Gardner grounded to nominal first baseman Sean Rodriguez, who was playing too far from the bag to take it himself and instead flipped to Myers for an unprecedented 3-9 putout.

It’s unclear what you’d call Rodriguez on that play, if not a first baseman. (Pitcher’s helper?) But it’s equally unclear in what way Myers wasn’t playing first base after Maddon moved him in.

“We have discussed whether Myers should be considered a first baseman for that play, specifically,” Labombarda admits. “We still have not made any decision yet, and anything we do, we have to discuss with Major League Baseball. The first baseman, the catcher, and the pitcher—those three positions need to be defined in the rulebook because they are allowed to stand in certain place and will use a specific glove.” (Labombarda is referring to Rules 1.12 through 1.15.)

Currently, companies like Inside Edge and Baseball Info Solutions chart defensive alignments, but it’s possible to imagine official scorers assuming that task (at least until Major League Baseball Advanced Media’s new field-tracking technology saves human observers the trouble). However, Labombarda doesn’t envision the scorers’ role expanding into that area.

“I don’t foresee it ever being the official scorer’s duty,” he says. “I think the official scorer’s duty will be determining what position the guy is playing. So for example, if we decide to change that Wil Myers kind of play to a first baseman, that will up to the official scorer to keep track of. But in terms of the third baseman going to play shallow right field when Big Papi comes up, I don’t foresee that being the official scorer’s responsibility other than when the ball is hit to that guy for it to be a putout or assist for the third baseman, in quotes, despite that he was playing in shallow right field.”

Box score designs vary by source, just as modern defensive alignments vary by team, but the official statistics are determined by Major League Baseball with input from Elias and the official scorers. Any uncertainty about how to classify fielders should be resolved when those three parties come together at the next official scorers meeting, which typically takes place between January 1st and the beginning of the season. “An official categorization for shift plays,” Merhige says, “would have to go to the scoring rules committee and the playing rules committee*, and probably through the union.”

Box scores are designed to give us essential information about games at a glance, and they’ve proven up to the task for well over a century. In order to cram the core info in, they leave a lot out. Box scores have never told us when fielders were playing in or at double-play depth, and while today’s innovative defensive alignments are more aggressive than that, they’re (more extreme) points along the same spectrum. We can’t expect the box score to encapsulate everything. Fortunately, between HD video, play-by-play and pitch-by-pitch data, and field-tracking technology, we’re preserving a richer record than we ever have before. We just have to look a little deeper.

Defenders are doing things we’ve never seen with the same regularity, if at all. Meanwhile, the box score abides. I don’t know about you, but I take comfort in that.

*The Scoring Rules Committee, which meets sporadically, is a subset of the nine-member Playing Rules Committee, a mixture of former players, team executives, umpires, and league officials.

Thanks to Chris Mosch for transcription assistance.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
I vote for use of the word "rover".
Fielding metrics are, as I understand them, based on the frequency with which a fielder makes plays on balls hit into his "zone". How are these metrics adjusted to accommodate the shift?
It's an issue, and a couple of years ago it seemed to be giving some wonky results in DRS. Admittedly I don't keep up with this as much as I should, but DRS was adjusted to fix it. UZR just chucked shift-plays, not sure if that's changed -- it would seem to be increasingly unsatisfactory as some teams shift nearly ever batter these days.
My question about how to score the shift isn't about the box score, it's about my in-game scorecard. I note the shift often done against Anthony Rizzo thusly: if the 5-3 play involves the 3Bman playing between first and second, I mark it as 5s-3 to indicate to myself that the shift was on. On the 3-9 PO above, I'd've done the same thing: 3-9s.

For those of us still in the age of cardboard and pencils. . .
Thanks for that idea. I haven't been to a game yet where a shift came into play, but I'm sure it will happen.
It shouldn't be a big deal. In both hockey and soccer, players move from their normal relative positions with great frequency. When a winger like Ilya Kovalchuk played the point on power plays, no one ever thought to call him a hybrid LW/D. A position is more a starting point for face offs than a definition.
I'm putting an "s" for shift after the position numbers when a ground ball is hit to a shifted infielder,for example "43s" for a grounder hit to the 2nd baseman playing short right field. If there's any confusion about where the shifted player is actually playing, I put the number of the position shifted to after the "s". Example: "53s4" for a grounder to the third basemen playing to the right of second, but "53s6" for a third baseman shifted into the shortstop hole.

In my experience, confusion doesn't actually occur all that often.