This chart is really incredible. It's the number of times a club used the Shift Defense in the last few years.



Something absolutely new is happening is baseball strategy. That is so cool.

If I may speculate wildly -

I wonder if the shift is going to lead to an increase in the value of bat control. If you can make a significant dent in the value of a power hitter by changing how the defense plays him, hitters who can hit 'em where they ain't are going to become more valuable. Swinging as hard as you can might become a little less of a dominant strategy if clubs are loading up the defense where you're most likely to hit the ball.

Likewise, I think one of the reasons the shift defense is a good idea in the contemporary game is that it isn't just your big slugging lugs who load up and swing as hard as they can every time - on the Red Sox, Jarrod Saltalamacchia and Mike Aviles are just as vulnerable to the shift as anyone else. In earlier times, your weaker hitting up-the-middle guys were bat control hitters whom you couldn't really shift all that effectively.

I could be entirely wrong in that speculation, but I do think that we're likely to see some sort of game theoretical response from hitters to this defensive change.

The two teams with no shifts on the season at all are the two teams with newbie managers with zero previous managing or coaching experience.

A couple other interesting things I noted:

The Rays' 216 was eough for first last year, but would only be projected to be 7th this year.

There also seems to be a big AL/NL split. Nine of the top 10 are from the AL (and the Brewers used to be!) and 4 out of the bottom 5 are from the NL.

I could be entirely wrong in that speculation, but I do think that we're likely to see some sort of game theoretical response from hitters to this defensive change.

Well if they don't respond they're morons. Its not like a major league hitter can't choose to bunt, and if you're a lefty and don't do that at least occasionally when a shift is on, you're cheating yourself.

There also seems to be a big AL/NL split. Nine of the top 10 are from the AL (and the Brewers used to be!) and 4 out of the bottom 5 are from the NL.

Yeah. And all five AL East teams are in the top ten.

There also seems to be a big AL/NL split. Nine of the top 10 are from the AL (and the Brewers used to be!) and 4 out of the bottom 5 are from the NL.

I think that is to be expected. You aren't going to shift except on the big sluggers and the AL teams, especially with the DH, are going to be able to carry the big plodding slugger a bit more. Just thinking through some DHs who get shifted on;

David Ortiz
Luke Scott
Travis Hafner
Adam Dunn

That's a group of guys who would be giving up an awful lot defensively if they were in the NL.

You aren't going to shift except on the big sluggers

This is the thing about the new shift, though. It's not just being used on Papi and his ilk. The Rays are shifting on Saltalamacchia. The shift should work against anyone who loads up and swings hard every time, regardless of how effective they are in doing so.

It would have been nice if Dewan had offered a specific definition for what counts as a shift, e.g., the middle infielder has to be positioned on the other side of second base.

Al Leiter: "Players were given those positions over a hundred years ago for a reason!"

Carry on.

It's crazy that the Rays have already shifted more in 1/4 of a season than any other team did all of last year.

I don't watch enough Angels baseball to tell, but are teams defending against Pujols with anything out of the ordinary or that could be interpreted as "shift-like"?

It's not just being used on Papi and his ilk

Exactly. The Rangers used a shift the other night on ... dang if I can even remember who it was on, Eric Hosmer or Brayan Peña or somebody who, not long ago, you wouldn't have felt was worth moving a step one way or the other for.

It was interesting to see the shift that the Red Sox played against Carlos Peña last night. Since he's been frequently bunting against the "standard" overshift, the Red Sox were playing a variation where their third baseman was standing about where the SS stands with a standard "infield in" positioning, while the SS was just to the left of the bag, and the second baseman was about midway between 1st and 2nd. This increased the size of the hole between SS and 2B, which Peña promptly took advantage of by hooking an outside fastball right through that hole in his first AB.

Exactly. The Rangers used a shift the other night on ... dang if I can even remember who it was on, Eric Hosmer or Brayan Peña or somebody who, not long ago, you wouldn't have felt was worth moving a step one way or the other for.

Lots of teams have been shifting on Eric Hosmer, and it's been killing his numbers. He's struggling in other ways too, but he's losing hits to the shift which is compounding his funk.

Maddon threw a new wrinkle into defensive positioning last night in the 9th inning. The Rays started the inning with Jeff Keppinger (poor defender) at 2B and Sean Rodriguez (great defender) at 3B; both are capable of playing either position. They stayed in that alignment through two righties and a switch hitter but after Ross got a hit Maddon had Rodriguez and Keppinger switch positions for the lefty Nava so the stronger defender would be on the right side of the infield.

It used to be mainly the occasional shift for a power lefty. More rarely it was for a power righty. Now, some teams are using shifts regularly.

I honestly don't know how hitters should adapt. They made the majors using a specific style, and I'm really not sure how it would work -- whether they'd be better off or worse -- if they tried to hit against the shift. Hell, do hitters even "protect the plate" with two strikes anymore? They seem to just be loading up and swinging away regardless.

IF there are hitters who are better at hitting against the shift -- I'm skeptical, as this isn't beer league softball -- it would be another skill that would have value.

Maddon threw a new wrinkle into defensive positioning last night in the 9th inning. The Rays started the inning with Jeff Keppinger (poor defender) at 2B and Sean Rodriguez (great defender) at 3B; both are capable of playing either position. They stayed in that alignment through two righties and a switch hitter but after Ross got a hit Maddon had Rodriguez and Keppinger switch positions for the lefty Nava so the stronger defender would be on the right side of the infield.

I love this. It's the kind of stuff you do in DMB but we'd been told real life players would be too insulted to make it worth trying.

Ray, just like with any change in playing conditions, the guys who can't adjust will lose playing time to guys who can.

The move I like with the shift that some teams do is to have the third baseman move into "the shotgun" portion of the infield while the second baseman and shortstop remain in fairly familiar locations. It seems very logical to me, particularly in a DP situation where I can't imagine why you would want the third baseman in the pivot position. Plus, with more ground to cover individually the shortstop is likely to have better range than a third baseman playing solo on the left side of the infield.

I'm glad to see a thread on this...

How does this impact our ability to use defensive stats (cue the how good is Zobrist debate)?

How does this impact our ability to use defensive stats (cue the how good is Zobrist debate)?

Good point. At the rate that Maddon and some others are using the shift, I have to think that it's going to be really difficult (even more than it is now) to get truly comparable measurements for some of these guys.

I hope this keeps its original name (the "Williams Shift"), because its very cool to have something as fundamental as this named after the greatest hitter who ever lived, non-Babe division.

More than anything, this seems like a Joe Maddon/Rays thing, doesn't it? I mean, if you take the Rays out of the picture, the change is much more gradual and slow.

I love this. It's the kind of stuff you do in DMB but we'd been told real life players would be too insulted to make it worth trying.

Maddon really puts a lot of the traditionalist statements on their heels. You wonder if it would work in a larger market with all the second guessing. The Rays are able to function like a baseball laboratory almost.

the brewers have been doing the shift thing on what i thought was an increasing basis for what i think is five years or so to compensate for having horrible infielders. melvin wanted to keep the guys he had because they could hit but knew the team had to help the pitchers somehow

one aspect of the shifts for all teams that i find interesting is that even on grass the second baseman will stand what looks to be 30 feet or so beyond the lip of the infield half the time. it's like something out of softball.

I wonder if the shift is going to lead to an increase in the value of bat control. If you can make a significant dent in the value of a power hitter by changing how the defense plays him, hitters who can hit 'em where they ain't are going to become more valuable. Swinging as hard as you can might become a little less of a dominant strategy if clubs are loading up the defense where you're most likely to hit the ball.

Well, sort of. It's not like teams either do "shift or no shift". They shift for individual batters. The bat control players will face the same defense they've always faced. Their intrinsic value will stay the same. But their relative value could certainly increase if the shift decreases the value of the TTO slugger. That may be what you're saying, but I'm not sure.

Now, what will happen to the TTO slugger? Well, they may try even harder to hit one out of the park. No point in swinging at a borderline pitch because if you don't absolutely kill it you will be out. So they take even more pitches and swing even harder. So the question is, will teams start to de-emphasize those hitters? You know, not promoting them, changing their style before they get to the majors, etc.

Questions:

1. Does the shift usually include moving OF over or just the IF?
2. Do pitchers typically also pitch differently, with more inside stuff or more breaking stuff? I realize this may be hard to quantify.

If I may speculate wildly -

I wonder if the shift is going to lead to an increase in the value of bat control.

I'd sure hope so, and if it doesn't, it sure would say a lot about those batters' lack of intelligence, or at least to their inability to adjust to changing strategies. The Fielding Bible came out several years ago with charts of many prominent batters' pull-hitting tendencies, and of course BB-Reference takes it even further with detailed breakdowns for every player. I've never quite figured why it's taken so long for most managers to get around to absorbing this sort of basic information and adjusting their fielders accordingly, but all props to Joe Madden for not looking a gift horse in the mouth.

Well if they don't respond they're morons. Its not like a major league hitter can't choose to bunt, and if you're a lefty and don't do that at least occasionally when a shift is on, you're cheating yourself.

When the Blue Jays had the shift on for Carlos Pena back in April at the Rogers Centre, he bunted twice(!) for base hits in one game.

Now, what will happen to the TTO slugger? Well, they may try even harder to hit one out of the park. No point in swinging at a borderline pitch because if you don't absolutely kill it you will be out. So they take even more pitches and swing even harder. So the question is, will teams start to de-emphasize those hitters? You know, not promoting them, changing their style before they get to the majors, etc.

Yeah, one way to beat the shift is to BB and HR more -- take the defense completely out of play. But batters will have the same problems doing that as they will trying to hit against the shift.

Let's put it this way: Does anyone seriously doubt that the ability to hit to all fields is an added value? Does anyone doubt that unpredictability is an asset?

Yeah, one way to beat the shift is to BB and HR more -- take the defense completely out of play. But batters will have the same problems doing that as they will trying to hit against the shift.

Plus bunting. David Ortiz did it the other day.

Let's put it this way: Does anyone seriously doubt that the ability to hit to all fields is an added value? Does anyone doubt that unpredictability is an asset?

If the hitter is already hitting to all fields, then there won't be much of a shift and nothing will really change.

If the hitter is not already hitting to all fields, it would be an added value to start hitting to all fields if the hitter has the ability to do that. But after some time, the defense will just shift back, so after a time it may not have much added value.

(And the spray charts are going to lag behind the new shifts; it will be a while before we see how hitters are reacting to the shifts, if they're reacting. I mean, bunting is easy to see right away - but an altered hitting approach is not.)

Last year I saw a spray chart on a hitter -- I forget who, maybe Ortiz -- and he almost never grounded to the left side. So why have more than one fielder on the left side of the infield? And is it really going to help Ortiz to try to hit grounders to 3B?

The other day it was bases empty and no outs when Ortiz bunted to third base. He got on safely. If we can accept that Ortiz walking is a good thing for Ortiz and the Red Sox then him bunting in that situation is a good thing as well.

Let's put it this way: Does anyone seriously doubt that the ability to hit to all fields is an added value? Does anyone doubt that unpredictability is an asset?

If the hitter is already hitting to all fields, then there won't be much of a shift and nothing will really change.

I thought that would have been understood when I wrote "added value".

If the hitter is not already hitting to all fields, it would be an added value to start hitting to all fields if the hitter has the ability to do that. But after some time, the defense will just shift back, so after a time it may not have much added value.

Well, it would prevent the sort of situation you have now with a hitter like Teixeira, where a hugely disproportionate number of his hardest hit balls are hit straight into the teeth of a bunched-up defense. It gets back to the whole predictability vs lack of predictability factor that I mentioned above, where unpredictability by itself gives a player** a certain added value and predictability lessens it.

**If pitchers benefit from mixing up their pitches, I have a hard time seeing why batters wouldn't benefit by not giving opponents reasons to profit by bunching up their defenders.

An Ortiz bunt: the fourth true outcome.

Seriously, Wikipedia's Glossary of Baseball defines the Three True Outcomes as, "The three ways a plate appearance can end without fielders coming into play..." On Ortiz' bunt the other day, there wasn't a fielder within about a half a mile of the play.

What's the average batters faced for a team in a season? So far the Rays have faced 1437 batters this season, so they're using the shift nearly 12% of the time. That's ridiculously high considering that the next closest team is probably only around 4-5%.

Ted Williams-Leigh Montville interview

TW: It certainly was the slider. I could hit the slider just as good as I could hit any ball, but I couldn't get it in the air as good. And you know they throw the shift there, and there's nine guys playing on this side and there's nobody over there, and I'm close to the plate and the only way I can get the ball over there is to uppercut it and hit it with the width of the bat, not the length of the bat. So I started looking for sliders and here they came. Boy I was laying for it. I was ready for it and I was crashing it. Now they're a little in doubt, now what are we gonna do? He hits high balls, fastballs, inside, outside, slider is the least dangerous and then they start throwing me a lot of sliders and I was starting to hit them out because I knew that was all I was gonna get. I knew I was gonna get it. If they're getting all of the good hitters out in the league, you knew damn well they were gonna throw it to Williams. So all those things fell in place.

LM: So the slider made you hit to the left?

TW: No, I didn't say that. You're putting words, you're trying to put words…

LM: I'm sorry

TW: No, I said for me to hit the slider, I had to look for it. I didn't care where a pitch was, if I was looking for it. But the slider was a pitch you kinda had to look for 'cause it was quick at the end. If I couldn't get under it, that's a little lag there, I had to hit it occasionally get one in the air, but it was a ground ball pitch. I didn't want any of those. But when they said go to left field, get a little further away from the plate, then that registered.

LM: And that opened it up for the .388

TW: Well, didn't open it up. Why did it open it up?

LM: Because you were getting hits… because you were hitting to the left field and they couldn't do the shift.

TW: That's right. They couldn't all go one way, they had to open up. Now I had more holes than I ever had in my life. Just hit a ball, a base hit.

LM: Did you say to yourself, why didn't I do this before?

TW: No I didn't, no I didn't. They hadn't changed the shift yet. That year I started hitting to left field a bit more and now they're opening up. I went to Chicago and I think I got 3 or 4 hits at least in the first game and then 3 or 4 hits the next day, and boy that shift was going out the window. Because I'd hit them all to left-center, through the box, shortstop, nobody there. So now they start opening up. Now the weather's getting hotter, June, July, and August. Now, I'm really starting to hit my own way and I had all kinds of room. That's what happened. Mantle had that tremendous year, and I happened to have a year that fell right in place

I wonder if they shift much in the minor leagues? I suspect they don't nearly as much, and probably shouldn't, but if the shift "difference" is growing rapidly between the two, then that means one more way where a successful hitter in the minors can become much less successful in the show - and there are already plenty of those.

If you believe having the ability to hit all fields and change your approach is important (yes) and better identified through scouting than stats (likely I think) then a stats type change (I suspect this change is driven, or at least enabled, but computer analysis of hit patterns byplayer) in the big league could make scouting more important for the minors.

Sure there is a whole pile of conjecture above, but I like how it shows that there is no one static right balance of scouts and stats, but rather it is a dynamic equilibrium. All of baseball is a dynamic system and, much like hitting/pitching is dynamic.

#27

For the Rays the pitchers are a big part of the shifting. The positioning will be different depending on the pitcher on the mound and they're taught to pitch to maximize the effectiveness of the shift. They also can call off or modify the shift if they want to pitch a different way.

The Rays' defense hasn't been particularly good this year. Their team DER is .694, just below the league average. Is there any reason to think all this shifting is doing any good?

Maddon really puts a lot of the traditionalist statements on their heels. You wonder if it would work in a larger market with all the second guessing. The Rays are able to function like a baseball laboratory almost.

Well, it's obviously worked for the Rays, maybe not as much this year per other commenters, which is why the O's are second (copycat league).

Clearly Maddon provided cover for the rest of the AL East by shifting last year and making the playoffs. Now shifting is barely more controversial than saving your closer until your team takes the lead. I don't listen to the press conferences, but I doubt that Farrell and Showalter field many accusatory questions about employing the shift - it's quickly becoming the new orthodoxy.

On the MASN broadcasts Palmer constantly points out that the value of the shift depends on the pitcher hitting his spots and inducing contact to a particular area - something he believes the Rays staff can do. That same something that extreme DIPs say is a non-existent talent, and (co-incidentally?) a talent that Palmer himself may have posessed.

So the hypothesis would be that (Ability to pitch to the defense + defense tailored to the pitching = lower batting average on balls in play):



Looks inconclusive to me. We really need another statistical proxy for "ability to hit spots."

So, let's suppose the trend picks up even more speed, and all teams are shifting frequently. And the reaction is for teams to shy away from extreme pull hitters and start prioritizing bat control. What does the equilibrium look like?

Tom (#9) -It would have been nice if Dewan had offered a specific definition for what counts as a shift, e.g., the middle infielder has to be positioned on the other side of second base

Greg (#27) - "1. Does the shift usually include moving OF over or just the IF?
2. Do pitchers typically also pitch differently, with more inside stuff or more breaking stuff? I realize this may be hard to quantify."

The term "the Williams shift" refers to the practice of moving the SS to right at second base (either side, but right at the bag), and moving the 2B over into the exact middle of the hole between the 1B and the new SS position. The third baseman tries to guess where best to play to prevent you from hitting to the opposite side. The term for moving the outfield way over is to just say "the outfield is pulled way over towards right." I've never heard the term "shift" applied to the outfield, although I suppose someone must have said it sometime.

The origin of this, which is very odd, is actually the NORMAL playing positions at the beginnings of major league ball, in 1971, if not even earlier. At that time, the first, second, and third baseman were expected (unless they were King Kelly) to just stand right there on the bag. This presents the shortstop with the exact situation the second baseman gets in the Williams shift - play right in between the bags. But this is a shift against RIGHTY pull hitters. The rise of the lefty batter may have been the factor that caused someone to finally pull their second baseman off the bag. Either that, or the righties figured out how to hit to the opposite field. Or maybe the shortstop moved from the left field side of the diamond to the right side, when he was up against lefties. I've never seen any serious evidence of which it was that happened, or when.

To give you an idea of how easy it is for managers to think of the shift, I myself have been subjected to it in two college inTRAmural leagues (I was never a varsity player in any sport, being FAR too slow, due to a missing ligament in each foot - or so says the military). But I do have real good bat control, and spend my PAs looking for something soft to pull a soft line drive into RF. Twice, although I wasn't leading the league in anything, I was hitting over .400, and sure enough, got the shift. It didn't work, because I can slap a pitch right down the 3B line, and can also slap one into the vacated SS hole, if the 3b comes over to the line. I don't have to bunt, but then, I wasn't playing against major league pitchers, either. Anyway, after a few PAs of that, teams would quit shifting on me. At the major league level, where I am outmatched athletically, you would think the hitters could compensate at least as well as I can. You would also assume that most pitchers have enough pitches that at least one of them ends up inside to a lefty batter, making hitting to the empty side hard. Williams apparently thought that he could hit just one pitch - the slider - to the opposite field (see #29).

BTW, the earliest days of MLB have a feature that may well be an attempt to deal with this inherent shift vs. righties. I have a set of all the guides from the 19th century, including the years when they used a system where the hitter could call for a high, low, or complete strike zone. They didn't print this information for everyone, but they did do a whole page on the batting average champions for each year, in both the NL and the AA. And in every single one of those pages, they note that the hitter called for a HIGH strike zone. That, I would assume, is an attempt to hit the ball in the air, where the infield can't get to it, and the shift becomes meaningless. And it's a characteristic of the best hitters. Hitting the ball into the air seems to have been the standard defense against the early shift, but, then, the batter gets to choose to swing only at high pitches. No pitching to contact for these guys. Those 2-seamers are balls, not low strikes.

- Brock Hanke

I hope this keeps its original name (the "Williams Shift"), because its very cool to have something as fundamental as this named after the greatest hitter who ever lived, non-Babe division.

I think you have a higher opinion of Cy Williams than I do.

The origin of this, which is very odd, is actually the NORMAL playing positions at the beginnings of major league ball, in 1971,

MLB and I are the same age. Cool.

David - Hah! Sometimes you can't even feel bad bout a typo, because it leads to something like this. - Brock

Dewan defines a shift as three fielders to either side of second base. So the cubs and their shifting is not being considered as a shift.

The cubs this year have induced more groundballs than last year but they are worse at converting them into outs than last year.

A couple nights ago Molina got the game winning hit because the second baseman was playing the shift.

The "Williams Shift" would move the outfielders over as well. That shift wasn't just about moving the infielders over. The "Williams Shift" was not a static tactic but an evolving one and one that many managers arranged differently. The first shifts against Ted were really really extreme. Lou would leave the entire left side of the field empty of any fielders. The Cardinals in the series I believe employed several versions of the shift against Ted. In one version the outfielders didn't really shift and in another they did.

Against Cy they basically left the entire left side of the field open as well.

Here is a link to a thread that explores the "Williams Shift" in detail.

If I'm not mistaken, the "slider" that Ted Williams is talking about is actually what we now call the cutter.

The shift or just that hitters are hitting more ground balls?

AL BABIP GB/FB

2012 287 .83
2011 294 .80
2010 295 .79
2009 300 .76

The easiest "solutions" to the shift are (a) get men on base, especially 2nd base so the 3B has to cover 3rd; (b) hit fly balls.

For his career, Ortiz had a GB/FB ratio of .57 for his career. He makes contact about 68% of the time. Only about 25% of his PAs result in a ground ball. (Note he has had a slightly higher GB/FB ratio the last few years.) His career H/PA is 244. 52 points of that is HR. About 12.7% of his PAs result in a LD (which might include some HR, I'm not sure) which accounts for about another 92 points (if HR aren't included). So about half the time he hits a GB or a non-LD FB and he has a BABIP of 200 on those. How many hits could the shift really be taking away?

Teixeira's just a mess all over the place. His XBH% is way down which is largely a severe drop in doubles -- hard to see how the shift explains that unless it's just the 1B playing dead on the line. Meanwhile his pop-up rate was well up in 2010-11. This year's he's been a totally different hitter -- career high GB rate, low pop-up rate but very low HR/FB rate, career high IP%, career low LD%, career low K and BB rates and an even lower XBH%. He's completely off his game right now, shift or no shift.

EDIT: Take back the earlier edit, I think they're OK.

Maddon threw a new wrinkle into defensive positioning last night in the 9th inning. The Rays started the inning with Jeff Keppinger (poor defender) at 2B and Sean Rodriguez (great defender) at 3B; both are capable of playing either position. They stayed in that alignment through two righties and a switch hitter but after Ross got a hit Maddon had Rodriguez and Keppinger switch positions for the lefty Nava so the stronger defender would be on the right side of the infield.

And IIRC, the batter hit the ball directly at Rodriguez.

When the Blue Jays had the shift on for Carlos Pena back in April at the Rogers Centre, he bunted twice(!) for base hits in one game.

If I recall correctly Carlos Pena led the NL in bunt hits last year.

i'm glad to see the actual numbers because it certainly SEEMS as if BABIP is down, ground balls are up, BA is down.

offense is down all over the place. i can't ever remember so many pitchers having ERA in the 1s and 2s this late into the season. They gonna hafta lower the mound

I have a copy of the 1970 San Francisco Giants Yearbook that has a photo of a shift the Reds used against Willie McCovey in 1969. Three infielders are playing on the right side of the infield, with the left side completely open. And the Reds are using four outfielders at normal depth - one in left-center, and three positioned in center field and right field. Never seen anything like it...

They gonna hafta lower the mound

Offense is where it was for most of the '80s. Nobody has to lower anything. And there's still an awful lot of crappy pitching going on everywhere you look.

I'm really happy with the level of offense, personally. So far the average team is on pace to score about 675 runs with a .715 average OPS. Compare to 2000, the height of the sillyball era, when the averages were 832 runs and a .782 OPS.