Baseball Therapy: Why Saber-Savvy Teams Might Want a Shift Ban

Well, congrats go out to Rob Manfred on assuming the post of Commissioner of Major League Baseball. And congratulations go out to Rob Manfred on igniting his first Twitter controversy about three hours into his lifetime term. In an interview that he gave to ESPN’s Karl Ravech, Manfred spoke of his desire to introduce “the clock” into the game, but then continued on to this tasty nugget:

“I think the second set of changes that I would look at is related, and that relates to injecting additional offense into the game. For example, things like eliminating defensive shifts, I would be open to those sorts of ideas.” Ravech picked up on the opening and asked “The forward-thinking, Sabermetric defensive shifts?” Manfred nodded in agreement.

Exactly 4.63 seconds later, #BaseballTwitter officially susploded. Again. And Ned Yost wasn’t even involved.

It’s easy to cast Manfred as an instant villain or as an old fuddy-duddy who is against creativity and innovation. Maybe a little too easy. That’s what Twitter is for, I guess. And yeah, the soundbite made it sound like he blames the nerds for ruining the game. Well, now that everyone’s had a chance to sleep on it, let’s look at this one all the way through.

Plenty of people (correctly) pointed out that while offense is down and infield shifting does have a negative effect on offense, the effect is (all told) relatively minor. If you want to call something out for the decline in offense, the ever-dropping strike zone, and the resulting strikeout epidemic are much more to blame. Teams probably get as much value out of a well-constructed platoon as they do out of shifting, and no one is out to ban platoons. In Manfred’s defense, it probably wouldn’t be wise to pick a fight with the umpire’s union (“Why can’t those guys just call a decent strike zone?”) on his first day, so he blamed the shift.

It’s true that shifting does reduce scoring (a little bit) and it’s easy to see that teams are shifting. The announcers always point it out. It’s like when politicians use “We need to fix our roads and bridges” as shorthand for “We need spend money on infrastructure.” Roads and bridges are certainly a part of infrastructure, although “infrastructure” also includes public transit, electrical power lines, sewers, natural gas and water pipes, landfills, and a bunch of other stuff that most people don’t interact with. Politicians rarely talk about how “We need to fix our landfills.” When you want to talk about infrastructure, you hit something that people see and understand and drive on. When you want to talk about lagging offense, you talk about something that’s easy for people to see, even if it isn’t the biggest issue. If Manfred is guilty of something, it’s being a bit of a politician. Shocking, I know.

I get that Manfred is worried about the drop in offense and that he would throw shifting overboard to bring a few more runs back into the game. He’s not in baseball ops, where the goal is to figure out every little advantage possible to try to win games. His electorate were the 30 MLB owners who are most concerned with making money. Like any business, that comes from putting butts in seats, eyeballs on televisions, and ears on the radio. Attendance, league-wide anyway, is going to be a function of how much disposable income people have for frivolities like taking someone out to a ballgame, and that’s a macroeconomic question that MLB has little control over. But they do have control over how well the game appeals to the marginal fan. This is the person who might spend money going to a baseball game or instead going to see that new movie which is really just a hackish sequel to some blockbuster from last year. Businesses expand their revenue by marketing to those marginal customers. Hint: That person is not reading Baseball Prospectus right now.

The die-hard fans (i.e., you who reads about baseball during Super Bowl week) would watch the game and buy tickets even if the shift were banned. I don’t have these data, but I’m guessing that MLB occasionally does some market research and they’ve found that the marginal fan would rather watch a 10-9 game than a 2-1 game. Maybe that’s pandering to the lowest common denominator—or as seems to be the common coinage, “ruining the sanctity of the game”—but that’s how businesses work, and that’s Rob Manfred’s (new) job. If offense sells, then let there be runs!

But okay, he took a little swipe at those of us who are reading Baseball Prospectus. Defensive shifting had its moment in the sun as the stunning triumph of big data in baseball. All of the hipster favorite baseball teams (many of them with BP alums in their front offices) used shifting the same way those same teams seem to have all traded for catchers who can frame like crazy this offseason. Infield shifting was (and still is) an exercise in the sort of outside-the-box thinking that every motivational speaker ever wants me to do. For too long, teams were bound by some sort of unwritten rule that you needed to have two over here and two over there. If we’re fairly sure of where Smith is going to hit a groundball, why not put an extra defender in that spot? Where in the rulebook does it say that teams must allocate infield real estate equally? (If Manfred gets his way, likely on page 83!)

I’m left to wonder if Manfred’s comments were less rage at Sabermetrics and more a frustrated comment on what big data hath wrought on the game. Of the marquee advancements that Sabermetrics has made over the past decade or so, they nearly all skew toward a better understanding of how to prevent runs. Beginning with the development of the first really well-constructed fielding metrics, teams started to see that defense had real value and began selecting players with that in mind. PITCHf/x probably has as much to offer pitchers as it does batters, but the information that it gives on pitchers is much more fine-grained. Catcher framing is about stealing extra strikes. The emphasis on hitter patience and making the pitcher throw extra pitches has had the counter-consequence of encouraging pitchers to simply throw strike one. Even the greatly trumpeted OBP revolution put the emphasis on players who walked and singled more, rather than the much more popular home run. You can’t pin all the blame on Sabermetrics, but it’s not like it helped.

The infield shift as been held up as a way that “small market” teams (i.e, those not located in the New York metropolitan area) have been able to use their brains to outmaneuver the “big market” teams. Even if it’s only worth about half a win to an average team, shifting persists mostly because it’s “free money.” It’s easy to implement and requires no extra salary commitments, so why not do it? Every little bit helps.

But let me, for a moment, play the contrarian. Or, I’ll let Yahoo’s Jeff Passan do it for me:

This is very telling: I ran Rob Manfred's idea to limit defensive shifts by two sabermetrically inclined GMs — and both said they agree.

— Jeff Passan (@JeffPassan) January 25, 2015

Wait, what?

Of course, we don’t know who those GMs are, but your mind just generated a short-list. Passan elaborated:

Both essentially said same thing: The game is better when the casual fans gets the product they want. Big concern baseball isn't delivering.

— Jeff Passan (@JeffPassan) January 25, 2015

Passan’s right. That’s very telling. But what is it telling us? Why would a GM willingly give up something that helps him to prevent more runs? Perhaps the GMs, under the cover of anonymity, are simply confessing their sins, but is there another motive? What if I told you that banning the infield shift might be the best possible outcome for a small market, Sabermetrically-focused team?

To understand why, let’s hop in the time machine and go back to January 2^nd, 1971. If that doesn’t ring a bell, it’s not surprising. It’s the day on which a ban on radio and television advertisements for cigarettes took place. Many of the people reading this (myself included) have never seen such an ad. The action was the result of a Federal law and driven in part by public health activists, but the surprise is that the tobacco companies breathed a smoke-filled sigh of relief when it went into effect. Advertising is expensive, and when all of your competitors are doing it, it probably means that everyone ends up in a stalemate. So, you are spending money for no appreciable gain, but you can’t stop, because then your competitor would be advertising and you would not be. Now, no one was allowed to advertise! Those of you familiar with Game Theory will recognize this as a modification of the classic Prisoner’s Dilemma.

The thing about the infield shift is that while it is innovative, easy to implement, and saves a few runs over the course of a season, it’s rather easy to spot what the early adopters are doing and copy it. That’s the innovator’s curse. You take the risk and put all the hard work in to come up with something new and people just copy it if it works. When shifts were done by a couple of teams, they were a cute novelty. Now everyone does it and would be fools not to do it. There’s no more relative advantage to shifting any more when everyone else is doing and it makes no sense for an individual team to go back to the old 2-and-2-no-matter-what system. The shift is now just part of the landscape.

Here’s a question that we’ve never discussed. When the shift was a novelty and a way for small-market Saber-savvy teams to grab a couple of free extra runs, we didn’t worry about the side effects. But now that it’s been brought to scale, we need to talk about it. We do have early data that suggest that hitters aren’t actually changing their spray patterns on groundballs. Again, we know that the effects aren’t all that big, and since everyone is shifting now, everyone is saving the same five or six runs. But because there’s no relative advantage, it’s no longer a source of freebies for the small market teams. On top of that, the idea of bunting to beat the shift sounds great, but the math doesn’t actually work out. Offenses can grab some of that lost value back by dropping one down once in a while, but not all of it.

The shift, somewhat by definition, doesn’t affect all players equally. The effects are mostly felt by pull-happy hitters (mostly lefties), while spray hitters are left mostly untouched. For someone like David Ortiz, the amount of value that he loses to shift doesn’t reduce him to nothingness, but it makes sense that at the edges, there will be players whose game would revolve, in part, on pulling base hits who are now diminished. Teams at the lower end of the payroll scale, as we have learned, have to make their bones by finding these guys and turning some of that straw into gold. The shift takes away a source of potentially flawed players who could nonetheless be contributors. It might be a small effect, but it’s more likely to hit those small market teams than the big ones.

Then there’s the optics problem. Whether or not the infield shift is responsible for the well-documented dip in scoring, it is now the poster child. If shifting turns off the casual fan, that has implications for MLB revenue streams, like the national TV deal that brings in something on the order of two and a quarter billion to the sport every year. While small market teams can lose out, comparatively, on local TV deals, they want that common pot (and their share of it) to be as big as possible, because it can bring their relative income (as a percentage) closer to the big market teams. If that number drops a bit, it will hurt a smaller market team’s relative purchasing power.

So what started out as a boon, even if a small one, for a little team that could is now something that everyone can do and it may even be hurting the small-market teams (if only a little bit). But there’s no way to put the genie back in the bottle. Unless Rob Manfred does it for them. It seems strange that some of the very teams that gave birth to the shift might want it deleted from the game, but sometimes life works out funny like that.

It’s true that the actual mechanics of banning the shift would be hard to figure out. MLB could make a rule that says that there must be two infielders to the right and two to the left of second base, although how they would enforce any sort of penalty is hard to fathom (the batter would get first base?) Despite what Rob Manfred might say, we probably are stuck with the shift. The shift might be a bad scapegoat for declining offense, but it’s also a cautionary tale in unintended consequences. Sometimes when you press a button, it changes all of the other buttons.