Here's why starting Alvarez in left is right call

October 29th, 2021

For most of the first four months of the season, Houston’s main center fielder was Myles Straw, who got traded to Cleveland at the Deadline on July 30. After that, the majority of the time went to rookie Jake Meyers, who injured his shoulder in the ALDS and isn’t on the World Series roster. In the first game of the World Series, fellow rookie Chas McCormick was playing in the middle. In Game 2, it was yet another rookie, Jose Siri, who made his Major League debut on Sept. 3, then broke a finger during the final weekend of the season.

In Game 3, as the Series shifts to Atlanta, it's going to be yet another new face in center. It won't be McCormick, and it won't be Siri. It'll be Kyle Tucker, Houston's breakout star of a right fielder. On Thursday, the Astros made it clear that when the Game 3 lineup comes out, DH Yordan Alvarez will be in left field, with left fielder Michael Brantley moving to right, which pushes Tucker to center -- and actual center fielders McCormick and Siri to the bench.

Tucker has just four career starts (five appearances) in center, which tells you how badly Dusty Baker does not want to be without the dangerous bats of Alvarez or Brantley in his lineup.

This is more than just a lineup quirk. This is going to be something we haven't seen in the Fall Classic before, not like this. No one in the post-war era has made a World Series start in center field with as little previous experience there as Tucker’s five games played. Barely anyone has done this at any position, really. (Siri had a comparatively luxurious seven games at center, including his two earlier playoff starts that preceded his Game 2 start.) It’s unorthodox, to say the least.

Now, this isn’t exactly Mickey Stanley switching from center field to shortstop a week before the 1968 World Series. Tucker is a good right fielder, and he was an AL Glove Glove Award finalist as a left fielder just last season. He's familiar with outfield grass. Even so, switching around the outfield spots isn’t quite as seamless as many would like to believe.

“Contrary to most people, they think you can just put anybody any place and they can just play,” Baker told the Houston Chronicle on March 20. “It’s not like that. I played all of the [outfield positions] out there, and if you’re not familiar with them at all, it’s really foreign to your responsibilities, your backup responsibilities, everything.”

So, what might this look like? Did Baker correctly choose offense over defense? How did the options play out. Let's run the numbers.

Wait, no one since World War II, really?

Potentially earlier, too, but that’s as far back as we could reasonably verify the data. The last World Series center fielder to start a game with as little experience at the position in the Majors was … well, Siri, on Wednesday.

But before that, it was Jacoby Ellsbury as a rookie in 2007, when he supplanted struggling veteran Coco Crisp midway through the playoffs and started in Game 1 of the World Series with 18 center field appearances under his belt. Before that, it was Tony Kubek, primarily a shortstop, who started as a rookie in center in Game 5 of the 1957 World Series after 23 games there only because Mickey Mantle had injured his right shoulder.

Even looking at all positions dating back to 1945, there's only a handful of players who earned a World Series start at a position where they’d had five or fewer games of experience.

• 1 game: Carlos Santana, LF, 2016
• 4 games: Elston Howard, 1B, 1957
• 4 games: Jim McAnany, LF, 1959
• 4 games: Ben Zobrist, RF, 2008
• 5 games: Miguel Cabrera, RF, 2003
• 5 games: Brian Hunter, LF, 1991

Put a star next to Santana, perhaps, because his previous game had come as a backup years earlier, making his first World Series start his first start in left field. (Entertainingly, Santana was also playing in place of Crisp. It went OK; Cleveland won 1-0.)

But in center? What the Astros are doing is borderline unprecedented.

How big of an offensive difference are we talking here?

Baker has five outfield choices, really. Three of them are left-handed, in Brantley, Tucker, and Alvarez, and over the past three seasons, they’re three of the 40 best hitters in baseball, making it easy to understand why Houston would want them all playing.

Two are right-handed, the rookies McCormick and Siri. McCormick was slightly above average in 2021 (with a 107 OPS+), while Siri had only 49 plate appearances, albeit after tearing up Triple-A, and entering Friday's World Series Game 3, he has two hits in 12 postseason plate appearances.

Platoon splits being what they are -- Alvarez has small ones, Brantley considerably larger -- and given the relative inexperience of the righty bats, the only thing to do here is to look at hitting projections split by pitcher handedness, graciously provided by Derek Carty via his THE BAT X projection system. These are shown in weighted on-base average (wOBA); the Major League average wOBA this year was .314.

Houston’s outfield projections vs. RHP

Alvarez-- .385
Tucker -- .371
Brantley -- .350
McCormick -- .307
Siri -- .266

Houston’s outfield projections vs. LHP

Alvarez -- .344
Tucker -- .343
McCormick -- .333
Brantley -- .306
Siri -- .292

Two things ought to be clear here. First, you absolutely must start Alvarez against a righty pitcher -- such as Game 3 starter Ian Anderson. Second, McCormick is projected to be a stronger hitter against lefties than Brantley (which played out in reality in 2021), which means there’s little reason to take the defensive hit to force Brantley in against a southpaw in later games.

In Game 3, Anderson will likely be followed by a slew of lefties. (With the injured Charlie Morton replaced by lefty Tucker Davidson, seven of the 12 pitchers on the Atlanta roster are left-handed.) In Games 4 and 5, without Morton, the plan is much less clear, but it’s probably going to be a kitchen sink’s worth of relievers. In the latter two games, maybe you mix and match. In Game 3? You absolutely have to start your three best lefty hitters.

What about the defense?

Of course, there’s a defensive price to pay.

Tucker is an above-average corner outfielder; McCormick has quickly proven he’s an elite defender; one of the best in the game. Brantley is either below average (minus-12 outs above average since 2018, similar to Joc Pederson and J.D. Martinez), or decent (minus-1 OAA in 2020-21), depending on how you want to look at it. Either way, he’s competent, not strong. We just don’t know enough about Siri yet, other than that he’s fast.

Alvarez, somehow, has never been charged with a fielding (non-throwing) error, which is massively misleading considering that in 2019, he looked like this and this. In his brief time in the field that year, the average outfielder would have been expected to turn 94 percent of his chances into outs. Alvarez made it to only 77 percent. It was the largest negative gap of any outfielder in baseball.

Still, we can take a look at 2020-21 numbers to see what we’re looking at.

McCormick: +12 OAA, +6% added value over estimated
Tucker: +9 OAA, +2%
Brantley: -1 OAA, 0%
Alvarez: 0 OAA, -2%
Siri: N/A

There’s a pretty large difference between having Brantley/McCormick/Tucker, which is a quite good defensive outfield, and Alvarez/Tucker/Brantley. (Tucker’s center-field numbers look worse than his corner numbers, but in only four games this year that’s barely relevant, especially when the big error he made there was a lot less about his position and a lot more about the San Francisco evening.)

But keep this in mind, too: There’s also evidence that Alvarez, while certainly not a strong defender, is no longer the cover-your-eyes case he was in 2019. In 41 games in left this year, he was a minus-1 by OAA; in two games in left earlier in the postseason, nothing all that interesting happened. There hasn’t yet been a game ruined by his fielding.

The “worst” missed play of the season for him, per Statcast’s metrics, was this single by Rob Refsnyder in August, an opportunity that would have been caught 85 percent of the time. While Alvarez’s route and reactions don’t look great, it’s not that bad, either.

“Yordan has done a tremendous job of working himself really hard and to the point to where he is a good outfielder,” Astros GM James Click said.

We might not go so far as good, but we can get to decent, enough to have played those 41 games without any major disasters.

This particular potential configuration -- Alvarez in left, Tucker in center and Brantley in right -- has happened in each of Tucker’s four previous starts in center, coming on July 30 in San Francisco, Aug. 3 and 4 in Los Angeles and Sept. 4 in San Diego. That’s obviously not much of a sample to draw from, though it should go without saying those games were also in DH-free National League parks.

For the most part, nothing happened. Most of the balls looked like this ...

... or this ...

… which sort of brings us to the final point here: Lots of outfield plays look like that, in the sense that any Major Leaguer should be able to handle them. Is that true?

How many chances are we really talking about?

This is the trick, really. Despite the long-held baseball truism that the ball will find you if you’re trying to hide someone there, the truth is that it might not happen often at all, because we’re not just talking about chances. We’re talking about competitive chances.

Tons of balls hit to the outfield are extremely easy flies, catchable by any Major Leaguer. Tons of other balls are obvious hits, uncatchable by anyone. It’s that narrow slice -- approximately 20 percent of batted balls -- in the middle that matters. How often do those come along? In the first two games of the Series, there really hasn't been any play like this for Houston's defense, but we're not going to worry too much over two games.

Instead, let's look back on the entire 2021 season, and balls hit to the outfield with a catch probability between 20-80 percent. In examining that range, we are slicing out the very easy plays (81-99 percent catch probability) and the incredibly difficult plays (1-19 percent), and trying to find that middle area where a fielder can make a difference.

How often does that happen? Less than you'd think.

In 2021, Astros outfielders saw 135 such plays, or one every 1.2 games.

In 2021, Braves hitters hit 125 such batted balls, or one every 1.3 games.

All it takes is one poorly timed, poorly fielded play to turn the entire World Series around. But you know you're getting your starting lineup two to three extra plate appearances over a pinch-hitter. You probably won't have the same chance to make an impact on defense, as we saw with Santana, and with a more recent example that's a little more topical: Alvarez, playing the field, in a World Series.

We’ve seen the best-case scenario happen before

In 2019, then-rookie Alvarez was a less-experienced outfielder than he is now, and a similar situation arose in the World Series. When the Astros visited the Nationals for Games 3-5, Alvarez served as a pinch-hitter the first two contests. He played left in Game 5, with Brantley in right, and George Springer in center. (Tucker, not yet fully established, was on the bench.)

Alvarez singlehandedly justified the decision in his first at-bat, blasting a two-run homer off of Joe Ross. He singled in his next time up and scored on Carlos Correa’s home run. Alvarez singled his third time up as well, then left for pinch-runner (and defensive replacement) Jake Marisnick in the top of the seventh with the Astros up, 4-0. Alvarez handled only one ball in the outfield, and it was an easy chance. He’d made a massive contribution on offense though, enough to help make it easy to remove him for defense. It’s how you draw it up.

“It’s a good problem to have,” Click said earlier this week, and he’s not wrong, because imagine if your problems were figuring out which strong player to send to the bench.

Baker had a decision to make, one that no other manager might ever have to make again, depending on the possible return of the DH to the NL. Against Anderson, at least, there's such a massive difference in offensive skill, and a relatively limited defensive risk, that it was always difficult to see Houston not starting Alvarez, Tucker and Brantley. If things go well early, that's not how the outfield will look by the ninth -- and that's exactly the point.