Former Baseball Prospectus writer Joe Sheehan posted the following nugget on Facebook the other day:
“The #Twins’ improved run prevention is entirely coming on contact. Last year, the Twins allowed a .320 batting average on balls in play, with 356 doubles and triples conceded. This year, the Twins have allowed a .298 BABIP, and 300 extra-base hits in play. That’s a full season of Byron Buxton, and it’s getting Sano out of the outfield in favor of Kepler, and it’s committing to Polanco at shortstop.
“This is the Twins’ best defensive team — their first good defensive team — since they moved into Target Field in 2010. The commitment to young, athletic players in the field is the biggest reason why they’re going to the playoffs.”
As a fan of a team that doesn’t necessarily put an emphasis on defense, this jumped out at me. Of course, Buxton has been making headlines and highlight reel defensive plays all season. In case you hadn’t heard, Buxton, always considered one of the top prospects in the game while he was making his way up the minors, improved his defense dramatically this season because last year he analyzed Statcast fielding data. Here’s how Travis Sawchik of FanGraphs put it:
Everyone knows Buxton is fast. The eye test tells us that. Statcast confirms that our eyes are not betraying us. But by focusing on his first step, both the quickness and direction of it, Buxton has essentially used Statcast to improve what we have collectively often referred to as “instinct.” While some of that instinct is innate, perhaps some of it is also learned and can be improved through practice. Buxton’s remarkable defensive campaign of 2017 seems to be loud proof of that.
We know that the Mets’ front office is open to Statcast numbers, at least on offense. One of the deciding factors in (finally) picking Lucas Duda over Ike Davis was that Duda had better exit velocity numbers. Is it possible that the same front office could be open to using Statcast defensive numbers to improve the team?
Unfortunately, we can’t answer that question. We can’t even really make any suggestions on who should look to improve upon what because the defensive Statcast numbers are not all publicly available. But we can go back and take the information from Sheehan that opened this piece and apply it to the Mets and see how the team fares.
Let’s go back to 2014, the last year the Mets finished under .500 and count this year as being complete, even though there are still three games left to play. The final 2017 numbers will be different, but not materially so. And let’s add some other numbers to examine, too. We’ll break up our look at defense between the four infield and the three outfield positions initially. Here’s what we have:
|Year||BABIP||BB/9||K/9||HR/9||2B/3B allowed||INF DRS||OF DRS|
DRS was used here because of the whole number aspect of it. If we used UZR the numbers would be different but the conclusions would not significantly change. With that out of the way, let’s look at the numbers in the chart.
The thing that jumps out at you is how shockingly bad the infield defense is this year. And the only thing that’s kept it from being worse is the promotion of Amed Rosario to plug the leak at SS. But the problem has been merely shifted to other positions. In 188.1 innings at 2B, Jose Reyes has a (-4) DRS while Asdrubal Cabrera has a (-6) DRS in 274.1 innings. In 331.1 innings at 3B, Cabrera has a +2 DRS. But Reyes had a (-5), T.J. Rivera had a (-4) and Wilmer Flores had a (-8) at the hot corner.
And unlike other years, even the first basemen have negative DRS numbers here in 2017. Dominic Smith has a (-6) and Flores has a (-3) in a combined 529.1 innings. In 89 innings at the position, Jay Bruce was a league average fielder at the position. And Bruce was a plus defensively in the outfield, with a +10 DRS in 800 innings in right field.
Remarkably, there’s a 6-to-7 win improvement available to the Mets just by utilizing league-average fielders in the four infield positions. Rosario should give the club that at shortstop. The expectation was that Smith would give them that and more at first base but he’s been significantly worse than that in the brief sample we’ve seen. Throughout his minor league career, we’ve heard he was a plus defender at first but the numbers don’t bear that out at all. Is it a small sample fluke or indicative of something else? Being a negative fielder will put more pressure on his bat and it’s far from certain his bat will be able to carry that particular load.
Some may be surprised at how good the outfielders have been because you always hear about how bad the Mets are defensively. But we absolutely need to make a distinction between the infield and outfield units.
Of course, a lot of the credit for the outfield defense goes to Juan Lagares, who has bounced back to his Gold Glove level and has an impressive +13 DRS in 547.2 innings in CF. But in LF, the Mets have received league average fielding this year and over in RF they have a +9 DRS so it’s not like they’ve received infield-level defense from their outfield corners this year.
The Mets believe that power pitchers can negate the influence of poor fielders. A staff filled with hard-throwers should result in a bunch of strikeouts and a bunch of easily-fielded, poor contact. They’ve gotten the strikeouts but the other half of the equation hasn’t quite worked out as planned. It’s impossible to tell how much of that is because of starts going to Chris Flexen and Tommy Milone rather than Noah Syndergaard.
But it’s hard to look at the numbers and conclude that the Mets can just punt infield defense again. The negative factor of the infield defense as rated by DRS is equivalent to the production above average this year of Giancarlo Stanton, who has a 6.4 fWAR. It’s simply mind-boggling. Here’s hoping that with the better quantification of defense available through Statcast numbers that the Mets will change course and look both to identify and address their defensive shortcomings.