The Astros Have an Outfield Shift
CLEVELAND — We know the Astros are one of the most forward-thinking, analytically minded organizations in baseball. They’ve led baseball in infield shift usage in recent seasons. They’ve experimented with piggy-back rotations in the minor leagues, they’ve been creative in maximizing draft pools, and have given us a revolutionary bullpen figure, the gift that is Chris Devenski.
They’ve also been as aggressive as any team I’ve observed with regard to outfield alignment.
Outfield alignment doesn’t receive as much attention as infield shifts. There are few, if any, outfield alignment measures publicly available, and we don’t often see outfield alignments in full scope on television broadcasts prior to a batted ball. Average depth is recorded by Statcast, but we’re still working on understanding optimum outfield positioning.
But the Astros are up to something — something which I first noticed last season at PNC Park.
Astros with an extreme outfield shift against Polanco pic.twitter.com/Z92F5xyo4j
— Travis Sawchik (@Travis_Sawchik) August 24, 2016
Since air balls are more evenly distributed than ground balls, there are typically fewer radical defensive alignments in the outfield. Since there are only three fielders tasked with covering a much larger area of ground than in the infield, outfielders are generally kept in equidistant positions, spreading risk. But the above alignment against the left-handed-hitting Gregory Polanco represented an extreme swing to the left. It appeared counterintuitive, too, with the Astros playing Polanco as if he were an extreme right-handed pull hitter. In this case, the left fielder was near the left-field line, the center fielder shading toward left center, and the right fielder nearly in right center.
But the approach appears to be rooted in logic, too. While most ground balls are pulled, air balls are more evenly distributed, with batters often slightly favoring the opposite field.
Consider Polanco’s 2015-16 batted-ball data (eliminating home runs since they are not in play). Polanco pulls most ground balls and sprays air balls with an even distribution:

But if you eliminate line drives — they are the most difficult batted-ball type to catch, and infield shifts are designed to absorb some line drives — then you understand why the Astros are aligning their outfield so aggressively. Note that Polanco most often hits fly balls, air balls with hang time, to left field.

To get a sense of what this looks like from above, let’s consider the starting positions of Houston’s outfielders courtesy of Statcast data, via MLB.com’s Mike Petriello.
The following graphic depicts Astros’ outfield alignment versus left-handed hitters at home. (We can only search the database by park at this point.)

And against right-handed hitters…

Particularly in left field against left-handed hitters, we find some extreme starting positions.
The Astros finished third in outfield Defensive Runs Saved last year (+33), trailing only the Royals (+45) and Red Sox (+41).
What is quite conspicuous about those alignments is that, again, the Astros play most batters as having opposite-field tendencies. While a number of teams shade the other way, this author has observed that the Astros have been more aggressive in that regard.
One reason why the Astros have been so aggressive is they have a willing manager in the open-minded A.J. Hinch, one of the younger managers in the game. Hinch said his mind is racing “at 100 mph.” He’s curious and seemingly willing to try nearly any creative approach that might yield added value. Hinch said he hasn’t been able to be quite as aggressive early this season with outfield alignment due to injuries, but he indicated he wants to be as aggressive — if not more aggressive — moving forward.
“When I have my most athletic outfield out there, I am being equally aggressive,” Hinch said of alignment compared to 2016. “Our best outfield defense is [Josh] Reddick in left, [Jake] Marisnick in center, and [George] Springer in right.… The ballpark will play into that a little bit. We can be a little more aggressive in our ballpark because if it’s over the left fielder’s head it’s either a homer or off the wall. Now when I have literally every player out of position in the outfield — Reddick is not a center fielder, [Nori] Aoki is not a right fielder, [Carlos] Beltran in left, he doesn’t play a lot of left — we are a touch more conservative [with alignment]. We can’t be as free spirited, but I tend to want to test it a little bit. There’s been some benefits to it. I’ve been burned at times.”
How do Astros pitchers react to, say, the right fielder positioned smack-dab middle of the right-center gap with a left-handed hitter like Polanco batting?
“Players don’t like it when it works against them,” Hinch said. “They sure do love it when there’s a bullet hit up the middle and they are shifted that way, or when we play shallow and there’s a looping line drive and we catch it. I’ve learned that players respect all the research we are doing and the method behind it. We spend a lot of time explaining what we are going to do, and why we are going to do it. But let’s be honest: when it doesn’t work, it’s frustrating for everyone… There’s been a lot of double plays we’ve turned, ground balls we’ve caught, balls that we’ve run down in the outfield, and I have to put an asterisk down on my card and go back and remind them ‘Remember when we shifted? That was cool, wasn’t it?’”
In his pre-game media briefing on Thursday, Hinch was asked about another radical outfield scheme that might not be too far away from becoming a regular alignment in baseball: the four-man outfield. It’s been used on rare occasions in the game, to date. But perhaps it is soon to become more common.
FanGraphs alumnus August Fagerstrom looked at candidates for being defended by four-man outfields last season. (Polanco was one of them.)
Earlier this month, ESPN’s Buster Olney speculated that four-man outfields could become a common occurrence soon.
Said Hinch: “We’re taking a look at it.”
Hinch cited Wednesday night’s game in Cleveland as an example of when a four-man outfield would have perhaps made sense to employ.
“We played the infield in with [Edwin] Encarnacion,” Hinch said. “The chance of him hitting a ground ball was so minute it really did not matter how I configured the infield. It was probably going to go to the outfield and it did… There will be a time when guys do that [play four-man outfieldes] for sure, especially the bigger places. I don’t know if Minute Maid [Park] is going to lead the league in [four-man] outfield alignments but Safeco or Anaheim… Center field in Detroit is difficult to cover with one person.”
The ideal hitter to field a four-man outfield against is a hitter with a pull-heavy ground-ball profile, and a hitter who more often lifts the ball than beats it into the infield turf.
Hinch also believes he has two shortstop-caliber defenders on the left side of the infield in Carlos Correa and Alex Bregman, which could better allow for a more aggressive approach to outfield strategy.
“We’ll probably be the team that does it,” Hinch said. “We might as well, just to get our brand out there.”
And the Astros brand is nothing if not tied to innovation and pushing the envelope.
A Cleveland native, FanGraphs writer Travis Sawchik is the author of the New York Times bestselling book, Big Data Baseball. He also contributes to The Athletic Cleveland, and has written for the Pittsburgh Tribune-Review, among other outlets. Follow him on Twitter @Travis_Sawchik.
Regarding Astros opposite field OF shifts for guys that spray fly and liners evenly, I think it is because infielders are playing so far back when shifted that they are taking away a lot of the shallow pulled liners/fly balls (i.e. less pulled liner/FB opportunities for OFs). On the flip side, with less IFs on away-side, OFs have more opportunities to catch shallow flies/liners away. Astros played shallowest in OF (or close to it) if I remember correctly from one of the tweets from Daren Willman last season.
Willman had done some cool OF range maps a year or 2 ago for OFs. It would be cool to see range maps for OFs and IFs overlay-ed on the hitters spray chart.
Does anyone know where one could find the numbers on how often a 4-man outfield or a 5-man infield are used?
Early yes but not too early to think ZIPS was wayyy too optimistic on Correa going forward. I could eat these words but he has looked mortal. I wonder if BABIP fueled minor league numbers should be discounted more? Still a great talent but he hasn’t looked elite in over a year.
Yup nothing good can from a SS who was 21 and put up 122 wRC+ for the year.
You clearly aren’t understanding my point. Not sure where I said nothing good could come from … So while snarky not illuminating in the least. I was commenting about his projections. He hasn’t been reaching them. Last calendar year he is at 117 wrc. Same level as a guy like Steve Pearce. Nothing wrong with that but that is certainly not elite. For the #1 fangraphs trade asset he isn’t even the best at his own position. Feel free to down vote. 🙂
I gave you a +1 southie. Your comment was open-minded in nature and showed a willingness to accept new info. You even said “I could eat these words . . .” to mitigate your perhaps unpopular opinion.
I’d like to hear more from Greaser Bob about why, specifically, he disagrees.
I think you made a fair point. Snark aside, I think Greaser Bob also has a decent point.
I think one issue is that we forget how young Correa is. We’ve also been spoiled by the PED era and some generational talents to expect too much at a young age.
Just since 1980, we’ve had Ripken (twice), Griffey, ARod (twice), Glaus, Pujols, Harper, and Trout (thrice!) put up a 8 WAR or more season before they turned 24. That’s unrealistic expectations for most players.
Other than ARod (who was either the greatest hitting SS in modern baseball or using PEDs very early or both) and Ripken, shortstops especially don’t put up huge numbers until they are a bit older usually.
Since 1980, Correa’s year last year was the 12 best WAR season of any shortstop younger than 24. If he doesn’t improve or takes a step back, that’s going to be a disappointment.
Our valuation of him is also tainted by Lindor becoming a superstar and Seager seemingly passing Correa offensively (for now at least).
I think we’ll see multiple 6+ WAR seasons from Correa and maybe some 7s or 8s, eventually.
Lindor and Seager certainly haven’t helped you’re right. Staggering offensive numbers being out up. I think you’re right about Correa. I could see a 7 win season if his defense holds steady. Maybe a more balanced player than an offensive beast. I don’t see a .300 hitter but a 275/375/475 guy. Perfectly acceptable more than #1 asset
Correa’s YTD slash line from xStats.org: .287/.372/.465.
Steamer’s RoS projection: .269/.351/.449
If anything, he is making better than expected contact thus far.
Lindor is a year older and has a career WAR of 12.2 vs. Correa’s 8.5. Swap their 2017 numbers, and people would be writing about how spoiled we are by Correa and how iffy Lindor is to become a star.
Downvoted 2x for being totally off-topic and reading way too much into 90 PA from a 22-yo.
Correa has a career low HR/FB rate (less than half of what he had in 2016 and a third of his rate in 2015), a career low BABIP (though not much lower than 2015), and a career low line drive rate. He’s still producing good-to-very good exit velocities (max: 111.7; avg: 89.2), so I bet once his batted ball data regresses to the mean, he will start putting up numbers again.
That’s one of the most entertaining manager interviews I’ve read in a while.
I know that Joe Mauer has started seeing some of those extreme outfield shifts this year and with good reason. It’s been disappointing to see him hit a few to the left-center gap only to see the center fielder take a couple steps to make the catch.
I just wish the Astros would stop bunting so much. Seems like they do it a lot for such a forward thinking organization.
I can probably tell you 75% of those were done by the player themselves. They still rarely bunt.
http://www.hardballtimes.com/why-flies-go-one-way-and-grounders-go-the-other/
…so I’m tossing ping pong balls at my fire extinguisher now…
Especially leaving RF open, if he gets it past the 1B/IF, then it’s pretty much a guaranteed triple at least. The potential damage for that seems more than the damage on an IF shift which is usually a single at worst.
It seemed like the Yankees had their RF way over in RCF when they were here in PNC Park. It seemed really weird but didn’t burn them.
I wonder if the graph you show where the Astros are playing lefties the opposite way at home is in some way a response to their ballpark. A ball in the left-center gap at Minute Maid can be deadly.
Aoki is not a right fielder?
Well, he has a very weak arm.
One might ask, what is perhaps one of the only drawbacks of living in New Orleans?
My answer would be that I am blacked out on MLBTV from all Astros games. I truly wish I could watch this team.
‘Remember when we shifted? That was cool, wasn’t it?’ is gonna make a great title to a post like this one day
Is it just me or have I seen the Padres do this sort of shift a lot too. I seems at bats for the last year or so where the Padres basically completely vacated right field.
I don’t think the four-man OF is coming any time soon to Houston, with McCullers, Keuchel and Morton all being extreme groundballers.