In the fourth episode of the second season of Star Trek, the crew members of the good ship Enterprise find themselves confronted with evil doppelgängers from another dimension. Though only Spock’s malicious counterpart had a beard, with the sinister versions of James T. Kirk, Dr. McCoy, and others remaining clean-shaven, the idea of evil imitators having facial hair has made its way into pop culture. If either Luis Arraez or Kyle Schwarber were beardless, this would work very well as a lede for this article
about the two of them being mirror-universe versions of each other. But they both have facial hair, so that’s right out. Oh well. Worth a shot.
Anyway, Luis Arraez is the Anti-Schwarber. Or Kyle Schwarber is the Anti-Arraez. The anti-parallels (perpendiculars?) between their games are obvious: Schwarber strikes out a ton, but sends the ball to rendezvous with the rovers on Mars with when he does hit it, while Arraez strikes out roughly as often as blue moons appear on a leap year Friday the 13th, but doesn’t have a whole lot of power. But even knowing that, the extent to which the two are inverses of each other is shocking (all stats prior to last night’s game):
Just about the only place they don’t diverge is in Launch Angle Sweet Spot; they’re both about as good as the other at hitting the ball in the optimal launch angle range. But in every other regard, they’re just about polar opposites.
Just by eyeballing the chart above (or a couple of at-bats from each), you get the picture. But we can also quantify their lack of similarity: Baseball Savant offers a Player Similarity tool that provides a numerical comparison based on batting profile (contact stats, walks, strikeouts), with numbers closer to 1 indicating greater similarity. For example, the players most similar to Kyle Schwarber are Nick Kurtz (.78), Owen Caissie (.73), and Munetaka Murakami (.72). The players most similar to Luis Arraez are Ernie Clement (.74) and Javier Sanoja (0.73).
By default, the tool will only show you the players with a similarity score of at least 0.7. For most players, this produces a busy but manageable visual, like so:
(If you’re wondering, each colored circle represents a different batting input, e.g. red represents barrel rate). And for players with highly unusual profiles—the unicorns of the league— you get a very tidy look at the very small number of players who can truly compare.
But if you want to find Arraez on Schwarber’s diagram, or vice-versa, you have to widen the criteria to show players who are much less similar. Much, much less similar. And once you’ve adjusted the settings sufficiently, you see this:
This resembles nothing more than one of those galaxy maps projected onto a planetarium dome; perhaps someday our finest scientists will send a probe to planet Pete Crow-Armstrong. But once you’ve gotten over the awe-inspiring chaos of the diagram, you can find out just how un-similar Arraez and Schwarber are.
Kyle Schwarber isn’t the least Arraez-esque player in the league. That would be Esmerlyn Valdez, a prospect who’s played 45 games for the Pirates this year, with a score of .03. Among league regulars, the least Arraez-like is Elly De La Cruz, with an .04. But Kyle Schwarber isn’t all that much closer to Arraezhood: with an .07, he ranks 413th on the similarity chart. Arraez isn’t the least Schwarberlike player in the league, but he’s close: only Chandler Simpson of the Rays is further away.
You already know the difference in barrel rate, walk rate and strikeout rate from the first table, but here’s a closer look at the contact inputs that go into that lack of similarity:
The difference in those numbers don’t leap out quite as much as the gaps in walk and strikeout rate, but Schwarber is over twice as likely to make solid contact as Arraez, and about half as likely to make weak contact.
Nothing you have read in this article is likely to come as any surprise. But perhaps that itself should come as no surprise. Everywhere you look in literature, film, art, you’ll see the idea of the mirror image, the inverted doppelgänger, the evil twin…
I could’ve started this with an evil Twin joke. Damn.
