The AL Cy Young Can’t Go Wrong, or Right

Though we won’t know the winner of the 2014 American League Cy Young Award until later Wednesday, we’ve already been given some clues. The BBWAA has told us the three finalists are Felix Hernandez, Corey Kluber, and Chris Sale. Based on the association’s own precedent, Sale isn’t going to win because he didn’t throw enough innings, so this is coming down to Felix vs. Kluber, as we’ve been assuming for months. The feeling is that Felix is going to win, and ESPN agrees with that pretty strongly, but Kluber’s case only got stronger as the season wore on, so it’s hard to imagine a bad choice. Which, from another perspective, means it’s hard to imagine a good choice.

There were two Cy Young winners in 2013. There were two Cy Young winners in 2012! There were two Cy Young winners in 2011, and in 2010, and in 2009, and in 2008, and in…you get it. There have been two Cy Young winners every year since 1970. In 1969, there were three, as AL voting was split between Mike Cuellar and Denny McLain. That’s our one existing case of there being co-Cy Youngs, meaning I think it’s safe to presume Wednesday will reveal a single winner. That’s too bad when you’ve got a pair of guys who are equally worthy.

We’ve gone over this in a lot of detail before. Let’s make things simple, before we make things complicated. So we’ve agreed to eliminate Chris Sale, because of his DL stint. Here’s what we have, over basically identical innings:

  1. Felix Hernandez, .244 wOBA allowed
  2. Corey Kluber, .277 wOBA allowed

That’s a pretty sizable gap. Taken as is, it would work out to a difference of about 26 runs. So, from there, you’d think this isn’t a particularly close race. But, of course, there are adjustments to make. The question is: do those adjustments make up a 33-point gap in wOBA against? From the get go, Felix leads; the burden of proof is on Kluber, you could say.

One thing you can note: Felix’s wOBA allowed was 19 points better than Kluber’s with the bases empty. With the bases not empty, it was 58 points better. So Felix, in that sense, was more effective in more threatening situations. Yet Kluber finished with a higher left-on-base rate. So this doesn’t really get us anywhere. We might as well move on to the bigger adjustments.

Namely, ballpark and defense. You notice that Felix allowed a .258 BABIP. Kluber, .316. Yet, Felix pitched half the time in an extreme pitcher-friendly environment, and Kluber didn’t. Also, Felix pitched mostly in front of a good defense, and Kluber pitched mostly in front of a bad defense. These things need to be adjusted for, to the best of our abilities, so we can try to isolate actual quality of batted balls.

Thankfully, a month and a half ago, Tony Blengino tackled the assignment. Blengino broke down all the batted balls in incredible detail, using data we don’t have access to, and you should read that if you didn’t, or if you forgot about it. After making all necessary adjustments, Blengino calculated that Felix had a “true” ERA of 2.29, while Kluber had a “true” ERA of 2.45. That works out to a difference of just over four runs, and, as Blengino concluded:

Yes, Safeco deflates offense, but it affects Felix much less than most because of his prolific grounder-inducing ability. Kluber’s subpar defense did hurt him, but his ability to work out of jams prevented it from negatively impacting his ERA. These are two clearly Cy-worthy seasons – but Felix Hernandez was a little bit better. He would get my vote.

So, still, we’re working from the position of Felix being just a tiny bit better. The gap is never going to be bigger than tiny. But there’s still more that we can look at, as long as we’re going to obsess over this. If there has to be an answer, we might as well know as much as possible, right? The first thing that might come to your mind is framing. Felix and Kluber didn’t pitch to the same catchers. Felix pitched mostly to Mike Zunino; Kluber pitched mostly to Yan Gomes. Baseball Prospectus is going to help us out a bit here.

Over there, you have framing data broken down by battery. According to their numbers, Zunino was worth almost eight runs to Felix. Gomes, meanwhile, was worth more like two runs to Kluber. Adding all the catchers together, Felix got 8.3 runs of framing help, while Kluber got 0.9. Taken at 100%, that swings things slightly in Kluber’s favor. If you figure this exaggerates the impact, and that pitchers deserve some of the credit, maybe you take things at 50%. Then Felix and Kluber are about exactly tied. Exactly tied, after all this.

There’s one more thing that we can do. I guess there are countless more things that we can do, but there’s one more potentially big thing: we can try to consider quality of opposition. After all, shutting out the Angels isn’t the same as shutting out the Astros. Now, this isn’t actually a simple thing to calculate. How do you capture an opponent’s true-talent level in the moment? We can try a couple different things. I’ll also note right here that Kluber pitched 46% of the time with the platoon advantage. Felix pitched 42% of the time with the platoon advantage.

Let’s take all of Kluber and Felix’s individual plate appearances and match them with hitter wRC+. We can use actual 2014 wRC+, and we can use projected 2015 wRC+, in case that provides a better idea of true talent. Then it’s easy enough to weight the numbers appropriately.

Based on actual 2014 wRC+

Kluber: 100 wRC+, average opponent
Felix: 100

So that accomplished absolutely nothing. They faced league-average hitters. But here we can find some separation:

Based on projected 2015 wRC+

Kluber: 102 wRC+, average opponent
Felix: 104

If you assume the projections better capture true talent than in-season performance, Felix faced a slightly more difficult slate. The best way to do this would be to take the hitter projections the day of, as opposed to after the season, but the projections don’t change that much. I think the safe conclusion is this: Kluber and Felix faced mostly similar schedules, but Felix’s opponents, on average, were a little bit more challenging. By maybe 1-2%. This whole race is separated by about 1-2%.

So you see the folly of trying to pick one. Felix and Kluber are basically exactly even. It’s Kluber, if you think Felix got a lot of help from his catchers. It’s Felix, if you think the catcher effect is somewhat exaggerated. Our best understanding of the difference would put it somewhere in the vicinity of, say, 3-5 runs, and when you think about the error bars in our estimates, that’s nothing. That’s noise. We can’t actually tell the difference between two pitchers separated by that few runs. It becomes a flip of the coin.

I assume Felix is going to win, because he begins in the top slot. He beat Kluber in ERA, he beat Kluber in hitting numbers against, and he beat Kluber in first establishing award momentum. Think of it like an instant-replay review: the ruling on the field is that Felix wins, and that needs to be overturned. Absent conclusive evidence, the call stands, and I don’t think it can be conclusively proved that Kluber was better. If the research doesn’t lean either way, you go back to what you had at the beginning.

But the numbers don’t lean either way. The best result of the voting would be that Felix Hernandez and Corey Kluber are co-Cy Young winners. Assuming that doesn’t end up the case, the voters are going to get this both right and wrong. In their defense, they were put in an impossible position.

Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

Newest Most Voted
Inline Feedbacks
View all comments
Pale Hose
8 years ago

“Yet, Felix pitched half the time in an extreme pitcher-friendly environment, and Kluber didn’t.”

Are we so sure this is true? By FG park factors both stadiums are rated at 97.

8 years ago
Reply to  Jeff Sullivan

I think SAFECO is more pitcher friendly against RHB’ers. RHP’ers face more LHB’ers where SAFECO does not help as much

Dave Cameronmember
8 years ago
Reply to  Pale Hose

Whenever a park radically realigns its dimensions, as the Mariners did to Safeco in 2013, the earlier park factor data becomes somewhat less useful, and so our calculations (based on five years of data) add in a larger share of regression to the mean, since we don’t know if the park will still play like it did before the redesign.

In cases where the realignment doesn’t end up changing things that much, then the result is that our park factors probably overcorrect for the changes. I think there’s a pretty good case to be made that Safeco is still extremely pitcher friendly, and our park factors just haven’t picked up on that after only two years of the “new Safeco” design.

I think, in general, all Mariners pitchers are a little overrated by FG data, and all Mariners hitters are a little underrated, because the park isn’t as neutral as our current PFs suggest.