John Dewan's Stat of the Week

Stat of the Week: Ban the Shift?


Over the years in Stat of the Week we’ve analyzed the significant increase in shifting in baseball that has occurred since 2012. We have been proponents of this defensive strategy as we have demonstrated in many different ways that, when employed against the right batters in the right situations, shifting is very effective. Of late there has been some additional discussion about the possibility of limiting shifts. According to a recent article written by Jayson Stark:

“At last month’s owners meetings, baseball’s competition committee gave the commissioner ‘strong’ backing to try to ‘put something in place’ to limit shifts, according to sources who spoke directly with members of the committee. So next up, it’s time to run this – and more – past the players’ union.”

Hopefully this “strong backing” doesn’t come to fruition. I will not mince my words. It would be totally ridiculous to try to limit shifting.

Last week, one of the best baseball writers in the game, Joe Sheehan, wrote a thorough essay looking at why banning the shift would be a bad idea. He goes through several arguments before landing on an important part of the discussion. He gave us permission to reprint his comments here.

Let me share some of my thoughts before you get to Joe’s excellent article:

– As effective as shifting is, it only drops the overall batting average in baseball by a few points. And they are pretty much all singles that are saved. If the idea of eliminating the shift is an idea to increase offense, it won’t. Not in any significant way.

Shifting in baseball is a defensive strategy that is similar to defensive strategies employed in other sports. Football is a good example. Think of the 3-4 defense. In 1972, the Miami Dolphins were the first team to win the Super Bowl with the 3–4 defense, going undefeated on the season. Should the NFL have banned the 3-4 defense because it was too successful for the Dolphins? What about the nickel package, the dime package, the cover-2, etc. These defenses where players are positioned in different areas of the field add to the strategy and excitement of the sport. Would the NFL be more exciting if they eliminated some of these strategies?

– What would the commissioner’s new rule look like? It could be that no infielder can position himself different from the old-school “formation” that crosses a line from home plate extending through the second base bag. What if this were done in basketball? What would that look like? There was a time when high school girls basketball did exactly that, like this:

No one can cross the midcourt line. Three guards from one team defend against three forwards from the other at each end. Players are limited to two dribbles. Officials run the ball up the court after a basket. It looked like two separate three-on-three games played on the same court.

Seems ridiculous but these were rules that were in existence as recently as the 1994 state tournament in Iowa.

Doing this in basketball again would be completely silly. The idea of doing something similar in baseball would be just as silly.

By the way, I highly recommend a subscription to Joe Sheehan’s Newsletter!

From The Joe Sheehan Newsletter, December 20, 2018:

Two weeks ago, Jayson Stark (congrats, sir, well deserved) reported that MLB, led by Commissioner Rob Manfred, was seriously considering rule changes to limit how the defense can position its fielders — what we popularly call “the shift.” Jayson’s deeply-reported piece reminded me that I’d promised someone, I think on Twitter, that I would do my own shift piece in the offseason.

My position on this is established, although I am not sure I’ve ever done a full column on it. Simply put, to limit how a team can play, how it can best align its talent to win games, is a generally bad idea. The shift, as much as it has become more common in recent seasons, isn’t a new concept. Teams were shifting on Ted Williams 60 years ago, and did so on comparable hitters in each succeeding generation. We’ve seen four-outfielder concepts as well, back to Willie McCovey in the 1960s and Jim Rice in the 1970s.

Even the standard defensive alignment to which traditionalists cling is an adaptation; the various basemen don’t stand on their bases, but rather, in the general vicinity of them, because early in baseball prehistory they noticed that’s where the balls were being hit.

What’s changed is the collection of data. The shift is the most visible representation of the data revolution within the game. Whereas teams, as recently as the last decade, had to rely on scouting information to determine where players were most likely to hit the ball, we now have perfect information for every player in the majors on their tendencies.

The shift isn’t new; the reliability of the information about when to shift is new. If you’d given this information to Joe McCarthy in the 1930s, he’d likely have shifted on Hal Trosky and Earl Averill. If Casey Stengel had it in the 1950s, Billy Martin would have been playing short right field against Larry Doby. You think Earl Weaver wouldn’t have welcomed knowing exactly where Reggie Jackson and Carl Yastrzemski were hitting their grounders, wouldn’t have used that information to put Mark Belanger exactly where he damn well pleased?

The entire point of having defensive players is to turn balls in play into outs. They’re not decorative, they’re not out there to conform to some airy notion about what a baseball game should look like. Forcing someone to stand somewhere near third base against a player who hits a ball on the ground that way once a month is nonsense. Every manager in baseball history would agree; it’s just the current ones who can act on the idea. Limiting defensive positioning is just the league forcing teams to play baseball badly, and isn’t that what we have pitcher batting for?

The argument against the shift is largely aesthetic. There are people with voices and platforms who just don’t like the way it looks when a hard grounder up the middle or a line drive to short right field becomes an out, rather than a single. To whatever extent there’s a “hard” argument for eliminating the positioning that creates those outs, it’s that doing so would create some more hits, more baserunners, more movement. I’m certainly in favor of that, and in the linked piece, Stark cites Sports Info Solutions data that show eliminating the shift would turn 500 outs into singles, mostly for lefty pull hitters. That’s not nothing — the league batting average would jump three points, and that’s before you consider any knock-on effects, like more pitchers throwing out of the stretch, etc.

On the other hand, Russell Carleton has made a strong case that shifting’s effect on pitcher effectiveness neutralizes its effect on hits on balls in play, so if pitchers pitch better under a new rule set, even a little bit, the effect on overall offense will be muted. Also, the hitters getting those singles back won’t be, for the most part, lively baserunners, so you’re not suddenly bringing back the 1970s.

The fact is, while it’s a highly visible practice whose successes you notice, the shift isn’t that big a factor in the modern offensive environment. It’s taking 500 singles (a handful of doubles are in there, too) a year and turning them into outs, largely from a group of players who aren’t dynamic baserunners. Singles themselves aren’t a key element in run scoring in the modern game, because they require other events, other balls put in play, to have value. One-run strategies and long-sequence offenses have largely been killed by strikeout rates.

There’s a pretty serious causation error driving the thinking about the shift. The idea is that batters are swinging for the fences because the shift has taken away hits on the ground. That’s completely backwards. Batters are swinging for the fences because it’s the most effective way to combat pitchers who throw 97-mph cutters and 90-mph sliders. You’ll hear citations of players like Rod Carew and Tony Gwynn from people who never saw anything faster than 70 mph in high school. Well, pitchers have evolved, mostly by throwing harder but also by getting more movement, into witches, and you can’t just slap the ball against the witches. That kind of batting style is obsolete. Batters have always tried to hit the ball hard and far; we just call those things Exit Velocity and Launch Angle now.

The dominant batting style of the day is an effect, not a cause. If pitchers couldn’t throw above 90, couldn’t get unreal spin on their breaking stuff, you’d see more contact hitting. Hitters have sold out for pull power as a strategy, and defenses have adapted to that by moving their infielders to where the balls land. The shift is the third thing that happened, not the first, and any rule changes that address the shift without addressing the pitcher/batter conflict are intentionally missing the point. Everything. Comes. Back. To. Velocity.

That’s not even my argument, though.

Eliminating the shift is actually going to incentivize the hitters who are being shifted to double down on their behavior. As it stands now, Joey Gallo and his ilk can do what they always do, and risk hitting a one-hop single to right that becomes a 4-3. They can, if they’d rather, drop a bunt down the third-base line or slap a ground ball anywhere to the left of the pitcher’s mound for a single. The incentives for them to change their behavior are clear…and they’re not changing their behavior. Oh, you’ll see the occasional bunt, and any time a guy like Gallo gets jammed and fists a four-hopper into left, the play-by-play guy will excitedly exclaim, “He beat the shift!” For the most part, though, this class of player has made his choice.

If you force a team to leave short right field open and cover the left side of the infield against someone like Gallo, you’re giving him a gift. Now he can employ his preferred style of hitting with much of the penalty for it removed. If you thought these guys were swinging for the fences now, what do you think will happen when you give them 20 extra singles a year for doing so?

That’s not even my argument, though.

In Jayson’s piece, he cites The Bill James Handbook: Of the 30 hitters who saw the most shifts, 29 were left-handed batters. (Edwin Encarnacion is the 30th.) The shift is aimed at hitters with predictable tendencies, which is largely pull hitters with power. The need, within the rules of baseball, to keep first base covered, means you can only do so much against right-handed batters. So the main targets of the shift are left-handed pull hitters with power. Anthony Rendon isn’t seeing many shifts. Andrelton Simmons isn’t. Christian Yelich isn’t.

There have been times in baseball history when rules changes have been necessary to balance offense and defense. We’ve outlawed flat bats and fair-foul hits. We’ve stopped letting pitchers put everything short of Play-Doh on the baseballs. We’ve raised and lowered the mound, manipulated the strike zone. All of these rules have served to balance the scales between the guys trying to score and the guys trying to keep them from scoring.

Outlawing the shift would be the first change in baseball history that is specifically benefiting a subset of hitters or pitchers. Remember, while there’s a gameplay issue right now, there isn’t an overall offense one. There were 4.45 runs scored per team per game last year, pretty much baseball’s historical average. This isn’t 1968, or even 1992. As we’ve seen, too, banning the shift isn’t going to move the needle on offense very much, anyway. It’s 500 singles and a few doubles a year over 185,000 plate appearances.

What banning the shift will do is provide an enormous benefit to the small class of hitters who are losing lots of hits to the shift each year. It’s the Logan Morrison Career Revival Act of 2019. It’s MLB literally picking winners and losers, saying that, aw, poor Travis Shaw, it’s not fair that they know exactly where you’re going to hit the ball, we’ll stop them from putting players there.

How is there any difference between saying you can’t put your fielders where the batter will hit the ball, and you can’t throw the batter pitches you know he can’t hit? Two years from now, will we have a rule that says you can’t throw Cody Bellinger curves down and in? Or that throwing Javier Baez a slider off the plate is an automatic ball?

Banning the shift is doesn’t balance offense and defense. It is a welfare program for a subset of players who won’t adjust on their own. Baseball has never, in its long history, picked out handful of players and made a rule just for them. That’s what banning the shift would be, and it’s a terrible precedent to set.

That’s my argument.

Mandating where a team can play its defenders is a terrible idea. It’s a subsidy for a specific class of players. It won’t put more action into the game. It won’t slow the trend towards dead-pull power hitting, and in fact, it will encourage it by taking away the penalty for doing so. It will provide the illusion of a solution while ignoring the real problem, the evolution of pitchers.

Baseball’s gameplay problem isn’t the fate of balls put in play, it’s the lack of balls put in play. Again, I say to Rob Manfred, if you want to baseball to look more like baseball, stop worrying about where the shortstop is and start giving him more to do.

Contact Information

If you want to react and respond to Joe, or have a question, use the addresses below. Please direct subscription questions through e-mail, thanks.




See what's up.