The Yankees’ one-run luck finally caught up with them

The Yankees have very much struggled in one-run games this season. According to Erik Boland, to be precise, the Yankees are 9-18 in one-run games. I’m going to tell you, despite what you’re seeing, that this is partially a factor of luck.

The Yankees have had spectacular luck in close games from 2013 to last season. The first piece of evidence is the obvious—that the Yankees have been incredibly bad in one-run games. That wasn’t necessarily the case in the past. Here are their winning percentages in one-run games from 2013 to 2016: .542, .511, .497, and .542.

OK, so they were pretty good in one-run games. But what about the actual talent level of their bullpens, and what about win probability? Here’s how weird this is: the Yankees bullpen in 2017 have the sixth-best ERA- but the fifth-worst Win Probability Added. If we were to compare this to 2013-2016, then we see how much these two stats usually track each other: in that time, they have the ninth-best ERA- and the third-best WPA.

What’s causing the discrepancy in 2017? One could say it’s the home run boost, but it’s not that. The Yankees’ HR/FB% ratio is lower now than it was in the previous four seasons, and lower than the current league average. What it could be, if I were to guess, is command and sequencing.

For one, the walk rate for the bullpen is abysmal. The Yankees bullpen has the fourth-highest walk rate in the league despite having the fourth-most collective fWAR. This kind of makes sense, and we saw it in the Red Sox walk-off walk on Friday. If you’re walking more batters—and both Dellin Betances and Tyler Clippard are big culprits—then things can fall apart quickly. This brings me to my next point.

The other issue, as I’ve brought up in the past, is that Joe Girardi just doesn’t know how to manage this particular bullpen. In the past, having a back-end of Dellin Betances, Andrew Miller, and Aroldis Chapman made things easy, as did having Mariano Rivera a few years ago. But with Betances having poor command and Chapman struggling/hurt, Girardi’s decisions haven’t really been clear. For example, look at how a pitcher’s leverage when they enter the game stacks up against their ERA- this season:

In fact, there’s almost no correlation at all. This means that the pitcher’s actual performance this season has had only a weak positive correlation to the importance of the situation at which they entered the game. That’s really, really bad. While in 2013-2016, it was a much stronger relationship:

This could mean Girardi just isn’t as clear on who is capable of what, but I’m not sure. Either way, he needs to figure this out pretty quickly, because the clock is ticking.

So, the issue is two-fold. On one hand, the team is likely to regress in a positive direction. Having Betances and Chapman is still better than what most teams have, and I can’t imagine they’re bad indefinitely. The overall performance of the bullpen is also currently better than when the bullpen routinely ranked in the top of the league. On the other hand, pitchers are not entering when they really should be, putting guys like Clippard in really important spots, only to blow it.

That needs to change. If they start sitting on Chad Green, Adam Warren, Betances, and Chapman, and maybe they make an acquisition at the deadline, they should be in a better place moving forward. For a team that is fading as fast as the Yankees, they’re going to need this change to happen nearly overnight.