A few thoughts on the shift

My Free Press colleague Jim Rueda had a piece in today's edition about the shift making inroads in amateur baseball as seen in the Mankato area -- colleges, high school, town ball, Northwoods League.

Executive summary: The local coaches are seeing more shifts from opponents, particularly a few opponents, but seem wary of the tactic themselves. They cite lack of specific data on individual hitters and, by implication, doubt that their pitchers have the command to "pitch to the shift."

My sense of this, which applies what the data says about hitters in professional ball (majors and minors) to amateur ball -- which may not be the wisest assumption to make -- is that those are not sound reasons to avoid shifting.

Point one: Almost all ground balls are pulled. It does not matter if it's a left-handed hitter or right-handed. Very few grounders go the other way. Balls hit the other way are almost always hit in the air -- line drives, popups or outfield flies.

And that's what the shift is for, ground balls. Line drives are likely to be hits regardless of defensive alignment (the batting average on line drives is over .750). Popups to the opposite field might test the range of that lone infielder, but they're still generally catchable. Outfield flies, the shift is irrelevant.

Conclusion: More shifts equal fewer ground-ball hits allowed.

Point two: That data comes without pitching to the shift. Logan Morrison (to drag a specific MLB example into this) doesn't try to hit where it's pitched. His approach is about launch angle -- get the ball in the air. Pitchers don't need to try to get him to pull the ball; they need to try to get him to hit it on the ground.

That's a different concept. Don't worry about in and out. If you get ground balls, they're going to pull.

Point three: These rules are universal enough that some MLB organizations employ blanket shifts all the way through their minor league systems. The book Big Data Baseball tells how the Pirates' farm director took short lengths of PVC pipe to their various minor league locales, drove them into the ground and told the infielders: You will play all left handed hitters here, all right handed hitters there, no exceptions allowed or tolerated.

Specific individual data? They don't need that. The general principle is enough to act on.

There is a fourth point, however, that does undermine the shift in amateur ball: Bunts. High school and college teams bunt more than pro teams do, often a lot more. You can shift on Morrison or Chris Davis of Baltimore with impunity; they might bunt once a year if that, and they just aren't very confident in that skill. That's probably not the case with high school and college teams.

Maybe that's enough to keep the shift from becoming as prevalent in amateur ball as in the pros.

Logan Morrison and the four-man outfield

The Houston Astros have deployed a four-man outfield on occasion this young season. The first to get the treatment was Joey Gallo of the Texas Rangers. Logan Morrison on Monday night was the most recent.

Both are left-handed power hitters of, let us say, limited batting averages. Gallo is about as close to a "three true outcomes" (strikeout, walk, home run) hitter as we may ever see. Last year he had 532 plate appearances; 312 of them resulted in one of those three outcomes.

Morrison isn't nearly as extreme (601 PA in 2017, with 268 walks, strikeouts and homers), but the idea is the same. 

In Morrison's case, the four-man outfield was trotted out when he got to two strikes (and no runner on second). The Astros had third baseman Alex Bregman playing back by the warning track in left field, and absolutely nobody on the left side of the infield. The three "real" outfielders were in center to right, and the remaining infielders were all on the first base side of second.

The "obvious" response is the bunt to the left side, but that's unlikely to be attempted with two strikes, especially by somebody as unaccustomed to the bunt as Morrison. 

The flip side of this aggressive defensive positioning is that it may be unnecessary. Morrison went to two strikes in 314 plate appearances; he slashed .148/.258/.311 in those trips. Get to two strikes on him -- the fewer balls the better -- and the real work of the at-bat is pretty much over. 

On bunts and shifts

Brian Dozier's complaint about Chance Sisco's bunt on Sunday drew enough blowback Monday that he tried to defend it by noting that the Orioles had not held Ryan LeMarre on first in the top of the ninth:

“When they didn’t hold our runner on, they conceded to the fact they didn’t want us to steal, so we didn’t steal.”

I'm not going to try to sort out that sentence or hunt for the logic in it. I'm more interested in posing and exploring this question: Why are bunts against the shift so rare that a veteran such as Dozier can believe that doing so violates the game's ethics?

Let's begin by making and testing these assertions:

• Teams shift to get more outs than they would get playing in a conventional alignment.
• Teams shift in the belief that the specific hitter either will not or cannot adjust to take advantage of the open spaces OR
• if he does, that the result is an acceptable tradeoff for keeping the hitter from pulling the ball.

The first seems rather self-evident, although Joe Posnanski also sees a mind-games element to the strategy. Which ties into the second point. Hitting is difficult, and the shift arguably plays with the hitter's mind.

The third might be the most interesting one. Posnanski writes that researchers believe Ted Williams, the most famous inspiration for overshifts, went 13-for-16 bunting against those shifts, "a sweet little .813 average."

OK, let's knock that down a little. Let's say that your typical left-handed hitter facing a shift -- someone like Sisco -- can bunt .600 against the shift. Let's further concede that that will all be singles, with no walks or HBPs, so that .600 batting average on bunts translates to a .600 on-base percentage and a .600 slugging percentage.

I differentiate here between righties and lefties because there is a significant difference in the shifts. The shifts Dozier faces leave the first baseman still relatively close to the bag, because somebody has to take the throw. The shift Sisco bunted against had nobody near third and thus all he had to do was get the bunt past Berrios. It's a little more difficult to bunt against the right-handed shift, but hardly impossible.

Anyway, back to our bunter with the theoretical slash line of .600/.600/.600. How many hitters can reasonably expect an OPS of 1.200 swinging away? Barry Bonds at his steroidal peak. Williams had a few years in that vicinity. The likes of Chance Sisco or Max Kepler can only dream of such heights.

And.600 seems fairly conservative, especially for someone like Kepler or Eddie Rosario, who can actually run. It might be something closer to .800, and even Bonds and Williams didn't put up 1.600 OPSes.

If hitters bunted aggressively against overshifts, the shifts would quickly disappear. If Kepler bunted everytime he came up against a shift that left third base open, teams would quickly stop shifting against him. Why make him a .600 or .750 hitter? It might take only one game for the shifts to stop.

Nobody's tried that yet, however. Maybe it's because hitters doubt their ability to bunt even when there is no defense offered. Maybe they believe it's a surrender to the opposition or an admission of weakness. I have a colleague who derides bunting as "skirt ball."

Just because Williams responded to the shift with defiance doesn't make that the right approach for everybody.

On bunts and no-hitters

Jose Berrios was spectacular on Sunday, and the Twins won 7-0. Then they turned the spotlight away from their young hurler to a ninth-inning bunt single by Baltimore's Chance Sisco, which they termed "bush league" and "not good baseball."

Nonsense. Berrios to that point had pitched out of the stretch to exactly one batter. The O's needed baserunners -- and, crucially, the Twins were in an overshift, with three infielders on the right side against the left-handed Sisco.

If' it's not good form to bunt for a hit when down seven, why is it good form to be in an overshift when up seven? The illogic in the position espoused by Brian Dozier, Eddie Rosario and Berrios himself is glaring.

Sisco reached base and triggered the closest thing Baltimore had to a rally all day. His bunt wasn't "bad baseball." It was good baseball, and the Twins shouldn't be complaining.

You don't want him to bunt? Don't overshift him. He's a slow catcher. It's that easy.

I suspect that part of the Twins angst over the Sisco bunt was the very real possibility that the one hit allowed by Berrios to that point could be changed to an error. It was a fly ball hit by Sisco with two out in the third Eddie Rosario got a glove on at the warning track but couldn't hang on. The scorer ruled it a double.

In the Twins TV booth, Bert Blyleven and Dick Bremer agreed. Me, I thought it was a ball that should have been caught.

Those are subjective opinions. For a more objective call:

The first hit by Chance Sisco in the third inning was a misplay by Eddie Rosario. It had a catch percentage of 90 percent, per Statcast. He needed to cover 83 feet in 5.6 seconds.

— Rhett Bollinger (@RhettBollinger) April 1, 2018

Ball's caught 90 percent of the time? Yeah, there's grounds to appeal that call.

Cisco's bunt single made that point moot in terms of getting a no-hitter, and later in the inning, Berrios gave up a clean single. So he got a three-hit shutout. I'll take that.

Now, let's play a what-if. Say the official scorer had ruled that third-inning fly an error. Would the Twins still have conceded the bunt to Sisco in the ninth? My guess is they would, and I still wouldn't be irritated if Sisco bunted with a no-hitter on the line. I don't know when that "unwritten rule" of not bunting late to break up a no-hitter came into being, but the literature of no-hitters is riddled with attempted ninth-inning bunts.

Sisco was competing, not surrendering. Good for him.

From the Handbook: Shifting

Combining two sections here, complilations of how often teams shifted and how successful those shifts were.

By Baseball Info Systems' accounting, the Twins shifted 906 times in 2017, an increase of 71 shifts from 2016. This includes the at-bats in which the hitter gets to two strikes and THEN Miguel Sano trots over to the first-base side of second base. It might be a one-pitch shift, but it counts because the AB ended with the defense shifted.

This puts the Twins pretty much dead center in the AL in terms of deploying shifts. Seven shifted more often, seven shifted less often.

Brian Dozier is spotlighted in the short introductory essay on hits lost and gained to shifts. Dozier faced 225 shifts in 2017, and BIS says he lost 21 hits to the shift -- and gained 24. Few right-handed hitters face as many shifts as Dozier (Albert Pujols faces more, but that's about it), and it doesn't appear to do much for the opposition.

From the Handbook: Defensive shifts

There are, I suspect, two types of sophisticated defensive metrics: the publicly available ones, and the proprietary ones designed by individual teams' analytics departments.

Baseball Info Systems, which generates the numbers in the annual Bill James Handbook, is prominent in the world of public defensive metrics. Its work has, among other things, laid out for public viewing the rationale behind the current trend toward infield shifts,

In his essay accompanying the shift stats in the 2016 Handbook, John Dewan says that major league teams shifted 13,298 times in 2014 and 17,733 times in 2015. (Those numbers represent balls put in play when the defense is in a shift, so at-bats in which the defense is in a shift early and out of it later aren't counted, and neither shifts in which the batter strikes out, walks, etc.) That's a 33 percent increase last year.

The shifts, by BIS' calculations, prevented 266 runs over all. That's a 36 percent increase over 2014's 196, and suggests that teams are becoming more efficient at shifting.

One other general point: BIS believes that the optimal number of shifts is more than double last year's total, that the upper limit is probably around 40,000 shifts. MLB isn't there yet, but if shifting continues to increase, 40,000 is possible in a few years.

OK, that;s the general stuff. On to the Twins specific shift data:

In 2014 -- with Ron Gardenhire as manager and Paul Molitor assigned to handle shifts (an assignment widely viewed as being an override of Gardenhire's preferences) -- the Twins deployed 478 shifts. Only three teams in the American League used fewer shifts.

In 2015 -- with Molitor now the manager -- the Twins used 724 shifts, sixth most in the AL. That's an increase of 246, That's a 34 percent increase.

The Handbook does not provide a team-by-team breakdown of runs saved by shifts, so I don't have a precise number for the Twins. Across the majors, teams saved 1.5 runs per 100 shifts, so if the Twins shifted at average efficiency they saved roughly 10 runs with the shifts. Ten runs sabermetrically equates to a win, so that's hardly an insignificant matter.

An unworkable fix to a nonexistent problem

Rob Manfred is now officially commissioner.

Sunday was Rob Manfred's first day as commissioner of baseball, and he opened the day with a notion apparently designed to make me miss Bud Selig: Banning defensive shifts.

Manfred didn't get detailed in his interview with ESPN's Karl Ravech on how this was supposed to work,  but the rationale is that runs scored have dropped the past few years as the shifts have become more commonplace, and he believes the fans want more runs.

Which they might. I don't necessarily need to see more runs, but I'm far from being representative of the masses. I would like to see more action, which is not exactly the same thing. I'd like to see fewer strikeouts and fewer walks, more balls in play and more baserunning. That might not result in more runs; it may well result in fewer runs.

I do believe this: The rise in the shifts that trouble Manfred (or, perhaps, that trouble certain people around Manfred)  and the rise in plate appearances that don't result in balls in play that concerns me, have a common cause: Sabermetrics.

Baseball's strategists know more about their game now than they did 30 or even 15 years ago because they have increasingly become open to the insights of the bookish outsiders. Those who reject analytics put themselves at a competitive disadvantage, and they increasingly fail and will continue to.

If the shifts have indeed disrupted the game's equilibrium -- which, as I'll get into shortly, is doubtful -- that equilibrium will recover. Hitters will either learn to "hit 'em where they ain't" or go the way of Jason Kubel, whose inability/unwillingness to adjust helped force him out of the game. ,

About all the shifting: the league batting average on balls in play was .296 from 2000-02 and .298 from 2012-14.
— Matthew Pouliot (@matthewpouliot) January 25, 2015

Shift info because of Manfred's comments. 44 players got shifted regularly in '14. their BABIP only dropped 9 points. http://t.co/cdoGKsrmCs
— Jeff Zimmerman (@jeffwzimmerman) January 25, 2015

The rise in strikeouts is the more likely cause of a drop in offense. The prevailing belief of sabermetricians for decades was that strikeouts are just outs, that there was no real disadvantage to them. Indeed, because strikeouts rise with home runs and walks, a team actually benefits from having players who do all three things. I put it thusly some 20 years ago when complaining about Tom Kelly's obsession with making contact: Strikeouts are the exhaust of the power engine. You want home runs, you're going to get strikeouts as well.

That was a useful insight as long as nobody was acting on it. But last year the strikeout rate exceeded 20 percent. (It was 12.5 percent in 1980). The researchers told teams that they were too worried about hitter strikeouts; teams stopped considering that when evaluating players; strikeouts rose; and now offense is declining.

I suspect the shifts will, in time, work against the strikeouts. The shifts are employed largely against power hitters, pull hitters. If they are forced to go the other way, they'll have to cut down their swings. They'll make more contact. Hit a few balls into the vacated gaps, the defense will stop shifting,

There's no need to order fielders to stand somewhere that they know the hitter isn't going to hit the ball. The game will solve the problem, if there is a problem, itself.

Bunting against the shift

Carlos Santana drops one down for a single Tuesday.

Top of the third inning, two outs, nobody on. Carlos Santana, the Cleveland cleanup hitter, steps up to the plate against the pronounced shift he typically sees when hitting left-handed and promptly pushes a bunt to the vacated third base position.

Easy base hit.

In the Twins TV booth, Bert Blyleven is critical of Santana for bunting. The old pitcher's analysis: With two outs, Santana should be looking for extra bases. A bunt single doesn't put him in scoring position for the next guy (Lonnie Chisenhall in this specific example). The unspoken next step of this line of thought -- unspoken, at least, by Blyleven -- is that the bunt is a selfish play, executed less to help the team than to fatten Santana's meager batting average. (Santana is having the kind of season at the plate only a stat nerd can love: He leads the American League in walks and he has 15 homers, but he entered Tuesday's game batting just .205. He left it hitting .214.)

Later in the game, I am listening to the radio guys, and the Santana bunt comes up. Dan Gladden is almost enthusiastic about the bunt. Santana got on base, he kept the line moving. An unspoken thought in defense of the bunt: Beyond doubt, part of Santana's meager batting average is the heavy use of infield shifts against him. Some bunt singles might discourage teams from shifting.

There's merit in both viewpoints. Blyleven's right in this: Santana's three other hits Tuesday (a home run and a pair of doubles) did more to help Cleveland score runs than the bunt did. On the other hand, had Santana lined a single to the outfield, nobody would criticize him for merely getting a one-base hit. Hitting is difficult.

Personally, I'm fine with the play. It's not the optimal time for Santana to take a bunt hit (leading off the inning would be), but he's got to do something to get the third baseman back on left side of the infield and open up the right side. And he's got hitters behind him who can do some damage. Getting on base -- not making an out -- is the goal of hitting.

Moving beyond this specific case to the abstract: There appears to be a brainless dispute within the game about when it's appropriate to bunt against a shift.

Earlier this year, the Houston Astros and Oakland Athletics got into a plunking war that was rooted in a bunt attempt by the A's Jed Lowrie with a 7-0 lead. The Astros responded by throwing at Lowrie.

This weekend, Colby Lewis, Texas pitcher, griped at Colby Rasmus, Toronto outfielder, for a bunt in essentially the same situation as Santana's bunt. (I can't resist; what's with these cheesy first names?)

My opinion: If it's appropriate for the defense to put on an exaggerated infield shift -- which are designed to take away singles; no infield shift is going to keep a home run ball in the park -- then it's appropriate for the batter to bunt against that shift.

The hitter is under no obligation to cooperate with the defense. Or the pitcher.

I suspect the pitchers unhappy with the opposition bunts are, in truth, unhappy with the shifts being deployed behind them. But they can't publicly vent about the manager giving away the bunt hit, so they direct their frustration at the guy taking what is given them. (Let it be noted that Lewis' ERA this year is 6.37, and Paul Clemens, the Astros pitcher who used Lowrie for target practice, has been up and down with an ERA of 6.08 in the majors. They can give up runs without their manager giving away the bunt.)

I know Tom Kelly thinks the shifts are overdone. But they are prevalent in today's game because they work. When enough hitters do what Santana did Tuesday often enough, the shifts will fade.