Why I Voted for Clayton Kershaw For NL Cy Young

For the second year in a row, I was given the opportunity to cast a ballot for the National League Cy Young Award. For the second year in a row, this was a very difficult task.

Last year, I had to pick between three aces who had historic seasons, finally settling on Jake Arrieta by the tiniest of margins over Clayton Kershaw and Zack Greinke. This year, no pitcher pitched a full season at the level that those three reached a year ago, so this year’s task was more about picking between good-but-flawed seasons rather than trying to decide which great year was the greatest.

And there was no shortage of options. In the end, I strongly considered eight players for the five spots we were asked to rank, and the guys who ended up at six through eight all had very strong cases for spots on the ballot. While the ordinal-rank system will make it look I saw a real difference between the guys ranked #3-#5 and the three notable omissions, the reality is that there was a six car pile-up at the back end of my ballot, and I think you have to split hairs to pick between the three guys who rounded out my ballot and the three guys who just missed.

But, it is an ordinal rank ballot, so let’s go through the ballot spot-by-spot so I can explain my rationale.

With Apologies To: Jon Lester, Johnny Cueto, and Madison Bumgarner

These guys were all terrific, and deserve recognition for the years they put together. Bumgarner and Cueto were second and third in the league in innings pitched respectively, while Lester finished sixth, so they all met the workhorse burden. Lester led the league in RA9-WAR, so if you just want to give a pitcher credit for everything that happens while he’s on the mound, you could make a case for him at the top of your ballot.

Bumgarner can hit too, and since I believe the award is to reward players at a specific position, not for a specific action, I think that a pitcher’s batting line should be a factor in the voting; it is part of the job description for NL pitchers, and impacts their teams chances of winning or losing, just as their own fielding impacts their team’s run prevention when they are on the mound. So while Bumgarner’s pitching-only numbers don’t quite stack up to the rest of this field, his value at the plate put him squarely in this group, and got him strong consideration for a spot on my ballot.

But for each of these guys, there was just enough there for me to push them barely outside the top five.

For Lester, it was the Cubs defense, and their impact on his run prevention, which made it difficult for me to give him full credit for the run prevention that occurred while he was pitching. While we don’t have perfect tools for separating pitching from fielding yet, the metrics we do have suggest that Lester was a little bit less responsible for the outs on balls in play he got than the other Cubs starter who was part of my top eight.

And the impact of those outs on balls in play couldn’t have been larger; Lester ran a .215 BABIP with men in scoring position, and the contact numbers just don’t suggest that Lester had a lot to do with that. So while his run prevention numbers are excellent, I didn’t want to reward him for the performance of his teammates.

For Cueto, the small marks against him were his performance at the plate and the presence of Buster Posey as his catcher. Cueto was one of the worst-hitting pitchers in the game this year, and so the Giants chances of winning his starts were a little bit reduced. Additionally, Posey ranked as the game’s best framing catcher by almost every framing metric out there, so just like we had to adjust Lester’s run prevention for the players behind him, Cueto gets a small downgrade for throwing to a catcher who makes pitchers look better than they are.

Bumgarner also gets a little bit of a downgrade for the Posey factor, but he also took a hit because 10 of the runs he allowed this year were categorized as unearned. The earned/unearned run categorization is pretty arbitrary, and looking at ERA versus RA as a way to try and account for defensive support is misguided, so Bumgarner’s actual results aren’t quite as good as ERA makes them look. And they were already a bit behind the other five guys, so even with his strong offensive results, Bumgarner just missed the cut for me. A great season, unquestionably, but like Lester and Cueto, there were just enough factors there for me to leave him outside the top five.

#5: Jose Fernandez, Miami

To be clear; this was not a vote to honor Fernandez because of his tragic death. After thinking about our responsibilities as voters, I decided that we were asked to solely determine which pitcher performed the best this year, and that it would not honor Fernandez’s legacy to consider his passing as a factor for this award. So I placed Fernandez fifth because of what he did for the Marlins this year, and not as an attempt at a tribute to a person we’ll miss greatly.

On the field, Fernandez was excellent, racking up the league’s best strikeout rate while posting the league’s fifth-lowest home run rate. That’s a pretty nifty combination of offensive suppression. By Baseball Prospectus’ DRA, which takes a stab at untangling all the values that go into run prevention, Fernandez was actually the best pitcher in the NL this year. And he hit pretty well too, running a .242 wOBA in 61 plate appearances, the sixth-best mark in the NL.

But there is one mark against him; that .332 BABIP, which led to worse run prevention than the run estimators would suggest. So, like with the Cubs pitchers, the question is how much of that we think was Fernandez versus his teammate’s performance. Unlike with the Cubs pitchers, though, there are some reasons to think that reflects some poor pitching. For one, Fernandez had a 28% line drive rate, easily the highest in the National League. And while line drive categorization can be subject to interpretation, some Statcast data shows that Fernandez did allow harder average contact than the other contenders for this list, and given that the Marlins had an above-average defense this year, signs start to point to Fernandez’s pitching having more to do with his BABIP than poor defensive support.

Of course, we don’t have all this figured out perfectly, and untangling pitching and defense remains the hardest part of these awards. Perhaps I’m putting too much weight on Fernandez’s BABIP here, and he should be several spots higher on the list, but based on our current understanding of the data, this is where I was most comfortable with him.

#4: Kyle Hendricks, Chicago

I know it looks weird to leave Lester off the ballot while putting Hendricks fourth, given that both succeeded by running low BABIPs in front of the Cubs defense. But at the risk of spending all of our time talking about how to assign credit for batted ball results, let me quote from Tony Blengino’s piece last week.

Kyle Hendricks is our 2016 NL Contact Manager of the Year… Hendricks allowed the least authoritative fly-ball contact in the NL and his average fly-ball velocity allowed was over two full STD lower than average.

Yes, the Cubs defense helped Hendricks this year, but perhaps to a smaller magnitude than it helped Lester. As Tony noted, Hendricks allowed a lot of weak fly balls, the types that are caught by just about anyone. Hendricks also allowed fewer home runs, which are runs that definitively can’t be blamed on the team’s fielders, which is another point in his favor of being the most worthy Cubs candidate this year. Toss in that he didn’t get the benefit of having a personal catcher to help curtail issues with the running game, and I give Hendricks just a few slight nods over Lester as the Cubs best pitcher for 2016.

But, like with Lester, we can’t completely ignore the fact that he pitched in front of one of the best defenses we’ve seen in recent years, and giving him full credit for his run prevention is not something I’m willing to do; his teammates impacted that number in a real way as well. So, combined with the fact that he threw just 190 innings, he fit best on my ballot in the fourth spot. Hendricks was terrific, but in a lower quantity of innings than most of the other excellent run preventers, and the numbers we know were mostly under his control just don’t quite stack up to the three guys in front of him.

#3: Noah Syndergaard, New York

Syndergaard and Fernandez had similar years in almost every respect. Thor was #1 among qualified pitchers in the NL in FIP, at 2.29; Fernandez was second, at 2.30. Syndergaard threw 183 2/3 innings in 30 starts; Fernandez threw 182 1/3 innings in 29 starts. And because extreme batted ball results were present with nearly all of the top pitchers in the discussion, Fernandez had a .332 BABIP; Syndergaard was at .334.

So, really, we’re left with the same question we’ve been wrestling with at each stop along the way; how much blame does the Mets ace get for giving up so many hits? With Fernandez, there were reasons to think that he bore a good bit of the responsibility, but this is where the two differ; with Syndegaard, it looks like the Mets just didn’t give him much help.

It’s not like the Mets made their plan much of a secret. After letting Daniel Murphy leave in free agency, the team doubled down on defensively challenged middle infielders, adding Neil Walker and Asdrubal Cabrera to take over as the least rangey 2B/SS combo in the league. They brought back Yoenis Cespedes and stuck him in center field in order to maximize the number of hitters they could get into the line-up, and when he got hurt, they used Curtis Granderson in center to make room for DH-in-the-making Jay Bruce. David Wright played with a serious back problem, and when he was lost for the year, they signed Jose Reyes and made him a third baseman for the first time in his career.

The 2016 Mets season was one big gamble on defense being overrated. Unsurprisingly, the Mets allowed a .308 BABIP as a team, third-worst in the National League, behind two teams that play at altitude in offensive-friendly ballparks. So, yeah, Syndergaard gave up a bunch of hits, but it probably would have been a surprise if he hadn’t, given what the team decided to put around him.

In terms of contact management, Syndergaard actually rated above-average by Blengino’s metric. He had the same average exit velocity as Lester. There just isn’t as much evidence that Thor was giving up hard-hit balls that helped lead to his high BABIP, and as such, I didn’t hold that number against him as heavily as I did with Fernandez.

And on a rate basis, he even out-hit Bumgarner, putting up a .292 wOBA in 67 plate appearances. The biggest knock against Syndergaard is that he’s terrible at holding runners, and thus he might always underperform his FIP to some degree, but even with that downwards adjustment for his own fielding deficiencies, Syndergaard was still one of the best pitchers in the National League this year, and ranked at the top of the second tier of Cy Young candidates on my ballot.

#2: Max Scherzer, Washington

And now we get to the big two. When it came down to it, this was the big decision for me, as these next two pitchers separated themselves from the pack based on my analysis.

Scherzer, I think, is going to win. As I write this post, all I know is that he’s in the final group with the two Cubs, and of that trio, he has the best case.

Even if we just went with the simple runs allowed model of evaluating pitchers, Scherzer would be neck-and-neck with Hendricks and Lester because of the difference in innings pitched; he threw 26 more innings than Lester and 38 more innings than Hendricks. By WAR that is based on runs allowed, that innings difference pulls Scherzer even with Hendricks at +6.9, and half a win behind Lester’s league-leading +7.4 total.

But since we can go beyond runs allowed, the other numbers give us more confidence that Scherzer deserves a larger portion of the credit for those results. Not only did Scherzer dominate the strike zone at an elite level — his 25% K%-BB% was second-best in the NL, behind only Fernandez — but he also has become one of the league’s most formidable pop-up generators.

This year, batters hit 34 infield flies against Scherzer, the highest total of any NL pitcher. Generating infield pop-ups is the most obvious way to suppress your BABIP, and these are the kinds of outs that we can certainly credit to the pitcher, not the fielder; that’s why they’re counted the same as strikeouts in our standard version of pitcher WAR. And those infield flies help push Scherzer well ahead of both Lester and Hendricks, even though their FIPs are all quite similar.

Scherzer combined a league-leading number of innings pitched with elite control of the strike zone and a batted ball profile that gave his fielders easier outs to work with than most pitchers. That’s a pretty great combination, and certainly makes Scherzer a strong Cy Young contender. And he almost got my vote. Except for the next guy on the list still exists.

#1: Clayton Kershaw, Los Angeles

Yes, I voted to give the Cy Young Award to a pitcher who didn’t even throw enough innings to show up on the leaderboards unless you reduce the innings minimum. Because, when I sat down and looked at everything, I couldn’t get away from the fact that Clayton Kershaw was, once again, the best pitcher on the planet in 2016.

I know he’s not going to win, since he wasn’t even a finalist, meaning I am in the minority in supporting excellence in a shortened season over slightly-less-excellent in a full-season. But in terms of overall value, Kershaw was still right there with the guys who threw an extra 40 to 80 innings.

By our standard pitcher WAR, based on FIP with infield flies added, Kershaw ranked #1 in the NL this year. By runs allowed, he ranked 4th, but was 0.1 WAR from being tied for second with Hendricks and Scherzer. Kershaw had a full-season’s worth of everyone else’s value; he just did it in four months instead of six.

So, it came down to whether I think Kershaw would have had a better season had he thrown a couple of mediocre months on top of what he had already done. Yeah, he threw 79 fewer innings than Scherzer, but he gave up 46 fewer runs. If he had limped through two extra months of pitching with a 5.24 RA9, would we think his 2016 was more impressive? Do we need to see great players perform not great for a while in order to validate their greatness just to get to some arbitrary number of innings minimums?

I don’t. I’m a peak performance guy. When it comes to the Hall of Fame, I’d already put Kershaw and Trout in, even if they didn’t do anything else the rest of their careers, because they’re two of the best we’ve ever seen. And in 2016, Kershaw was at his best again, until his back put him on the shelf for a couple of months.

But what he did before the injury was enough for me to say he was the NL’s best pitcher again this year. Had he stayed healthy, it wouldn’t have even been a race, but he consolidated a full year’s worth of value into the time he spent on the mound. For me, he was the best pitcher in the NL this year, and that’s why I put him at the top of my ballot.