On Monday, the Baseball Writers Association of America will announce their six finalists -- three in each league -- for the 2019 Cy Young awards. After that, we'll have more than a week to wonder who actually won them until they're announced on Wednesday, Nov. 13.
But we don't want to wait that long, do we? Of course not. We want to know right now, or at least have a pretty educated guess. In the absence of breaking into the BBWAA offices in order to open up the envelopes -- we assume this works like the Oscars, and let's just agree together that it does -- we'll have to look back upon past history. We'll have to rely on the Cy Young prediction tool.
This is a tool that's been useful in the past. In 2015, it got the top five in both leagues in the right order. In 2016, it correctly predicted the three finalists in both leagues. In 2017, it correctly got the AL finalists in order. You get the idea. It's not something that attempts to predict who should win, based on their on-field performance; it attempts to predict who will win, based on how voters have chosen in the recent past.
(Remember, all votes are due at the end of the regular season, so nothing a pitcher does in October, positive or negative, will be considered.)
The methodology, originally created by MLB.com's Tom Tango, is relatively simple.
Cy Young Points = (IP/2 - ER) + SO/10 + W
There's nothing fancy in there. There's no Statcast metrics. There's no advanced modeling, or attempts to account for good or poor defense. It's an understanding that for as far as we've come with advanced stats, Cy Young voters still place a heavy emphasis on "throwing a lot of innings with a lot of strikeouts and preventing runs."
Or at least they have, anyway. What does the tool show this year -- and how close will the results be?
100 points -- Gerrit Cole, Astros
99 points -- Justin Verlander, Astros
71 points -- Charlie Morton, Rays
70 points -- Shane Bieber, Indians
59 points -- Lance Lynn, Rangers
58 points -- Lucas Giolito, White Sox
Remember, again, that votes were due at the end of the regular season, so voters had no advance notice of what Cole, Verlander, and Morton did in the playoffs.
This is, for all intents and purposes, a dead heat. There's no reason to believe a prediction tool is so precise as to imagine a difference of one percent carries much meaning, so this is a tie -- and that makes sense.
Verlander and Cole were first and third, respectively, in the American League in innings pitched. They were 1-2 (with Cole on top) in ERA, and they were 1-2 (with Cole again on top) in strikeouts. If you prefer something fancier, they were 1-2 (Cole, again, on top) in the best Statcast quality-of-contact metric. It's not hard to see why they're so far ahead of their former teammate Morton, who is essentially tied with Bieber at No. 3.
That's another way of saying that it's perfectly reasonable for either one to win, hence the tie. The advanced metrics lean slightly towards Cole, but in MLB.com's final straw poll of nearly three dozen writers in late September, Verlander collected 24 first-place votes to Cole's 11. Some of those people will have votes in the real ballot, too. Verlander had pretty strong cases to win when he finished in second in 2016 and 2018; we think he'll take this one home this year.
Prediction tool's choice: Cole, but basically a tie
My choice: Cole
My prediction: Verlander
(Note: Zack Greinke collected 73 points, which would have put him in the top three. However, he split his season between the National League [D-backs] and American League [Astros], and there's only been one Cy Young winner to take the award in a year where he played in both leagues -- and Greinke didn't go 16-1 after the trade like Rick Sutcliffe did for the Cubs in 1984. Dwight Gooden should have won the NL award that year, anyway.)
84 points -- Jacob deGrom, Mets
75 points -- Hyun-Jin Ryu, Dodgers
72 points -- Jack Flaherty, Cardinals
71 points -- Stephen Strasburg, Nationals
64 points -- Clayton Kershaw, Dodgers
The NL is a little different, because while the AL has a very clear top two, the NL class is so deep that you could probably make some kind of case for about a dozen different pitchers. (The next four names on this list are Sonny Gray, Mike Soroka, Luis Castillo and Walker Buehler, each of whom have some kind of argument to make.)
This one has been interesting to follow over the course of the season, because Castillo appeared to be the early leader -- in our May 14 straw poll, he and San Diego's Chris Paddack were nearly tied, while deGrom received zero votes -- and then Ryu took over a strong lead for most of the summer as he kept his ERA south of 2.00 until mid-August.
But then Ryu posted a 4.60 ERA over the last two months, as he dealt with injuries. Scherzer made only eight second-half starts (with a 4.81 ERA) as he dealt with injuries of his own. Meanwhile, Flaherty posted a historic second half (0.91 ERA in 15 starts), but an only-okay first half (4.64 ERA in 18 starts) proved too much to overcome.
And in New York, deGrom, the defending Cy Young winner, who didn't receive his first first-place vote in our poll until Aug. 27, put together a pretty stellar season of his own. By the end of the year, deGrom was second only to Ryu in ERA, but he'd done it in nearly 22 more innings, and with 92 more strikeouts. He was tied with Scherzer in that same Statcast quality-of-contact metric, and he did it in 32 more innings.
With the exception of allowing 9 more home runs in a world where everyone allowed more home runs, deGrom put together basically the same season he did when he won in 2018, following a 32.2 percent strikeout rate/5.5 percent walk rate with a ... 31.7 percent strikeout rate/5.5 percent walk rate. Throw in the fact that there's probably some recency bias at the end of the season that boosts deGrom over Ryu, and we agree with the tool on this one.
As for the third finalist, the predictor is basically telling us that Flaherty and Strasburg are basically tied, and don't be surprised if Strasburg -- or Scherzer, for that matter -- end up as finalists.
Prediction tool's choice: deGrom
My choice: deGrom
My prediction: deGrom