August 25, 2016
Zach With No K
Hypotheticals are fun. If they weren’t fun, nobody would put any time into thinking about them because, well, they’re hypothetical. Recently, hypothetical scenarios have gotten a lot of press, what with Lebron learning handball, and Tim Tebow figuring out how to waste the time of scouts.
It was that sort of thinking that led the BP Stats team down an interesting path on the afternoon of August 24th. The question at the heart of the matter was equal parts absurd and vexing:
What if Zach Britton couldn’t strike anyone out?
Britton has been excellent this season. His ERA (0.53) is first in the majors, and challenging the all-time record. His RA9 (1.05) is first in the majors. His cFIP (59) is fifth in the majors, and fourth among relievers. His DRA (2.10) fourth among relievers. His RE24 (23.80) is fifth in the majors, and first among relievers. His WPA (4.68) is first in the majors. He might be a Cy Young candidate. He’s been good.
Of course, if Zach Britton couldn’t strike anyone out, he’d be less good. He’d definitely be less good, but how much less good isn’t clear. The reason is simple—nobody has been good enough at other things to not strike anyone out. In fact hardly anyone can be successful while striking out fewer than 10 percent of the batters they face. Only Blaine Boyer (9.4 percent) has struck out less than one out of 10 batters and still been able to pitch more than 40 innings this season.
In order to understand what a no-strikeout Zach Britton might look like, we should start by understanding what exactly real Zach Britton looks like. The following is a table of outcomes from each batter faced by Britton this season:
Britton strikes out a little under a third of the batters he faces, while roughly half of them hit groundballs. This is a result of his absurd 80%+ groundball rate. Britton also allows few home runs. Only 5 percent of plate appearances against Britton end with flyballs, and just one in 10 of those flyballs leaves the park.
These key attributes allow us, with a bit more confidence, to entertain the idea that a no-strikeouts Britton might not be worse than your average beer league softball pitcher.
Roughly a third of Britton’s plate appearances will no longer end in strikeouts, but rather other things. This is where we get into several alternate universes, each of which sees Britton’s strikeouts becoming different outcomes.
The groundballcentric future
Adding those runs to Britton’s ledger means that he goes from a 1.05 RA9 to a 3.51 RA9. Britton goes from being Zach Britton to roughly Yusmeiro Petit. It could be worse. That still makes Britton a better than average pitcher, despite striking nobody out.
The realistic future
The likely reality is that Britton’s missing strikeouts will become all of his other PA outcomes in the same ratios they exist in today. So all of his strikeouts will become walks, groundballs, line drives, and flyballs. This requires a bit more math, but by multiplying the incidence of those outcomes against their linear weights, we can figure out how many more runs Britton would have given up over those 60 plate appearances.
Add it all up and we need to add a measly 2.5 runs to Britton’s account. Don’t forget those missing strikeouts though! Adding in those missing run values means that this version of Britton is actually 16.3 runs worse than the current model.
This raises Britton’s RA9 to 3.86, and pushes him further down the list of effective relievers. He’s not Heath Hembree. Sorry, Zach.
The really realistic future
Unfortunately, we’re really pushing the limits of hypotheticals here, so we’re going to make some rough adjustments for the sake of the argument. We’ll detail them below this updated table of changes:
In this version of the future, Britton has traded some groundballs for other outcomes. Specifically, the following changes have been made:
So if Britton pitches exactly as he does now in every other plate appearance, but his strikeouts are replaced with the outcomes listed above, we’re adding five runs to his account. Plus the 13.8 we established in our baseline for losing all his strikeouts, and we have a Britton that boasts an RA9 of 4.20. We have a Zach Britton that is actually Jhan Marinez.
Taking it a Step Further
This means that those RA9 values listed above are artificially inflated. If you remove those 60 strikeout-outs, and replace them with an unknown number of outs and non-outs, then our divisor changes. Britton will have no longer thrown 53.1 innings this season. He’d have pitched fewer innings, which means each run allowed does more damage to his RA9.
This means that our new Zach Britton isn’t really Heath Hembree or Jhan Marinez. He’s very clearly worse than them. How much worse? It’s hard to say. So the answer to our question from before (What if Zach Britton couldn’t strike anyone out?) is that he’d probably still be a major leaguer. Maybe a fringe guy who comes up from Triple-A whenever someone gets hurt. He’d definitely be different.
Suffice it to say that it’s a good thing for the Orioles and Zach Britton that he can, and does, strike batters out.
Note: all data used in this post comes from 8/23, before Britton came in and gave up a run in a game he had no business pitching in on 8/24. Special thanks to Harry Pavlidis and Russell Carleton for their assistance with the concept and execution.