Years ago, during either the second or third time that Mike Trout clearly had a better season of baseball than everyone else on the planet, maybe the fourth, there was a debate if Trout had a slightly better season of baseball than everyone else on the planet. Miguel Cabrera won the Triple Crown, see, which is something that hadn’t been done since the ‘60s, and some people thought ... oh, man, I almost fell right back into the argument.
Point is, it wasn’t a debate between conflicted people who could see both sides. It was a debate between saber-rattling zealots who were sure they were right. Or SABR-rattling zealots. How you evaluated baseball players — your entrenched, unchanging worldview about what makes a baseball player good — was being questioned by the other side. Cabrera did the things that made fans cheer, and he did it far more often. Trout did everything much better, from baserunning to defense, and he ... oh, man, it’s happening again. This will never end.
Both sides thought they were right. Both sides took about .03 seconds before making their decision about who should win. Both sides regret nothing about their decision.
It’s not like that with the NL Cy Young this year. It’s a mess. It’s a mess if you’re into advanced analytics. It’s a mess if you’re into traditional analytics. It’s a mess because it’s clear who the best pitcher is, but he doesn’t even qualify for the ERA title because he missed so much time. It’s a mess because you can make an argument against every candidate easier than you can make an argument for them.
And it’s gonna make for the messiest Cy Young vote in years. Or, year, at least. Last year’s was pretty messy. But in that case, there were three pitchers who all clearly deserved it. This year, the race is about pitchers who have done just enough to mess with their chances.
A breakdown of just how it’s a mess for everyone:
If you’re into traditional metrics ...
Oh, you like wins, do you? [pats head, feeds peanut] Good for you. It’s nice to see someone stick with their convictions in the face of available evidence. That’s never steered anyone wrong before in any field, as far as I can remember.
But the odds are against a 20-game winner. Max Scherzer has a shot, but he needs four more wins. He’s picked up four in his last four starts, so it’s certainly possible for him to do it over his next five. It’s unlikely. Same goes for Jake Arrieta, who needs four wins, but will probably get just five starts. And that’s before you remember that both pitchers are employed by teams who will fidget with the rotation to get their best pitchers properly lined up for the postseason.
And even pitcher-win weirdos know that 19 wins doesn’t mean much more than 18 wins. Once you escape the base-10 bubble, it’s anarchy. Anarchy!
So, you look to ERA and, well, shoot, there aren’t a lot of answers there, either. Because Scherzer and Arrieta might have a chance to lead the league in wins, but they don’t have much of a chance to win the ERA title. Their ERAs are likelier to be closer to 3.00 than 2.00, which makes them imperfect candidates by traditional metrics.
And the guy who might win the ERA title doesn’t have as many wins, possibly because he rarely pitches seven full innings.
This is a mess.
Also, good luck with the old ways. When the seas rise and civilization is set back 400 years, we might need you.
If you’re into advanced metrics ...
Well, you’ll just sort everyone by WAR so you don’t have to think, and ...
- Max Scherzer, 5.5
- Clayton Kershaw, 4.7
- Carlos Martinez, 4.5
- Madison Bumgarner, 4.5
- Tanner Roark, 4.5
There you have it. WAR isn’t a hyper-precise metric, but a gap of nearly a win is probably substantial enough. So, Scherzer takes it. The traditionalists won’t be too offended, what with the wins and the gaudy strikeout totals. Glad to know there’s a consensus on ...
- Noah Syndergaard, 5.7
- Clayton Kershaw, 5.5
- Jose Fernandez, 5.0
- Max Scherzer, 5.0
- Johnny Cueto, 4.4
Three completely different pitchers out of five. A different leader. The same problematic pick in the second spot. The lead guy on the second list doesn’t show up on the first list; the lead guy on the first list is almost off the second list.
The difference is how the metrics are calculated. Baseball-Reference uses the runs that actually scored. FanGraphs extrapolates how the pitchers pitched, defense be damned. And you can see the argument for both sides. Using the runs that actually scored? A novel concept. WAR adjusts for things like innings pitched and park effects, so it’s probably better than clunky ol' ERA. But it has the same limitations as ERA, too.
For example, Syndergaard plays in front of a bad defense. It’s possible to evaluate every Mets player with a dismissive "should be in the lineup, sure, but the defense is going to be a drag." Their center fielders should be corner outfielders; their corner outfielders should be first basemen; their first baseman for most of the year should be a DH; their shortstops should be second basemen; their second baseman should probably be a first baseman; their first baseman for most of the year ... well, you get the point.
Which makes Syndergaard’s league-leading FIP seem relevant.
On the other hand, tallying up what actually happened without the need for an abstract adjustment appeals to most folks. FIP can be a better predictor of future ERA than ERA itself ... but only over a small-to-medium sample. Once you get to two or three seasons, the two stats are very similar in their predictive power, which means that ERA doesn’t have to be pure noise. It also means that FIP doesn’t have to be pure truth. And we’re awarding the pitcher who pitched the best, not who is more likely to pitch better going forward.
Split the difference between the two and go with the guys who appear on both lists? Fine by me, except one of them doesn’t even qualify for the ERA title.
Note that Kyle Hendricks doesn’t appear on either list, and he might win the ERA title with an ERA under 2.00. It’s been over a decade since someone with those qualifications didn’t win the Cy Young. Seemed weird then. And it’ll seem weird now. Of course, that has to do with innings, which ...
If you’re an innings fetishist like me ...
There are far, far too many candidates who might not pitch 200 innings. It’s a different league, different game, where health is more important than that old-timey need to prove that every pitcher is a workhorse.
Still, the only starter to win the Cy Young with fewer than 200 innings pitched was Kershaw in 2014, and he was so much better than his competition that he won the MVP, too. Hendricks hasn’t been quite as dominant, and he’s unlikely to get to 200 innings. If he makes all five starts, as the rotation is set up now, he’ll need seven innings in every outing. If he doesn’t pitch on the final day of the season — which is likely — he’ll need a complete game every time out.
Considering that Hendricks isn’t likely to go over 200 innings because he rarely throws more than six full innings, that seems nearly impossible. And it’s not just Hendricks -- Carlos Martinez and Syndergaard suffer from the same deficit. Kershaw might not reach 150 innings, which would nearly disqualify him, even if he’s clearly been the best pitcher while healthy.
More than the raw innings limit, though, you’re rewarding the workhorses for not giving up runs in the innings they didn’t pitch. Other people had to pitch those innings, and they almost certainly allowed runs at a higher rate than the pitcher contending for the Cy Young. Workload matters when you’re trying to evaluate the best pitcher.
Fine, so go with the guys shouldering the workload, pun intended. That would be the troika at the top of the leaderboard, Scherzer, Madison Bumgarner, and Johnny Cueto. Except Scherzer’s ERA is inferior to both of the Giants pitchers, and the FIP is comparable. Before you make your way to park effects, note that AT&T Park has been a slight hitter’s park this year, with Nationals Park helping pitchers more.
I’m pretty sure I don’t believe that, but the three-year park factors for both ballparks are closer than you think, too. Which brings us back to the unanswered ERA/FIP debate from the last section. Even if you prefer FIP, what’s the tiebreaker for this?
Johnny Cueto, 3.09
Madison Bumgarner, 3.17
Max Scherzer, 3.18
Probably ERA at that point, right? Except WAR might be a better tiebreaker. Except strikeouts might be a better tiebreaker. Except ...
It’s all a mess. The one constant who keeps coming up in all three arguments is Scherzer, so he’s probably the presumptive front-runner. Unless Hendricks gets that ERA under 2.00. Unless Kershaw comes back and rattles off four brilliant starts. Unless Bumgarner finishes with a flourish. Unless Arrieta finishes even better. And where’s Syndergaard in all this? He’s been mesmerizing.
It’s a mess. The twist ending is that the American League is even more of a mess. The Cy Young races in both leagues have broken baseball evaluation.
I miss the old days, when people were wrong about Miguel Cabrera, and we knew it in a very smug, self-satisfied way. This, right here? This is nothing but a happy, fun mess.