Why Clayton Kershaw Should Win the National League Cy Young Award
The 2010 awards season marked a milestone of sorts for statistically-inclined baseball fans.
Felix Hernandez, who was 13-12, was named the Cy Young winner, as the Baseball Writers of America gave more weight to his MLB-best earned run average than win/loss record.
One day, perhaps, we will see more accurate and useful defense-independent pitching stats (DIPS) supersede ERA in the eyes of voters.
The Baseball Writer's Association (almost certainly) is not there yet, but if it were, Clayton Kershaw would be the choice for National League Cy Young this season.
Still, the advanced stats suggest Kershaw has a stronger case.
The Case for Clayton
The 27-year-old southpaw posted a 2.13 ERA in 232.2 innings this season, striking out 301 and walking 42 batters in the process.
His ERA and 0.88 WHIP were both third best in the Majors, behind Greinke (1.66 ERA, 0.84 WHIP) and Arrieta (1.77, 0.86).
The combination of this dominance on a per-inning basis (or per-batter basis in the case of cFIP and DRA) with the most innings pitched the Majors put Kershaw on top in both fWAR (8.6), and Baseball Prospectus’ DRA-based WARP (7.94).
Greinke (9.3 rWAR) and Arrieta (8.6) both finished ahead of Kershaw (7.5) at Baseball-Reference, whose wins above replacement formula is based on runs allowed.
So despite his dominance in the defense-independent metrics, why did Kershaw allow more runs than his fellow excellent peers?
As you probably guessed, our friends BABIP and home run per fly ball rate were up to their usual tricks.
Kershaw allowed a .281 BABIP, while only Marco Estrada was more fortunate than Greinke (.229) and Arrieta (.246) in the balls-in-play department.
While it might be tempting to just assume Kershaw was not as good at inducing weak contact, the tools at our disposal do not support this.
Of the three pitchers in Cy Young contention, Greinke was actually hit the hardest, allowing the highest exit velocity of the trio, according to Baseball-Savant.
The average ball left the bat against Greinke at 87.78 miles per hour, compared to 84.91 for Kershaw and 84.89 for Arrieta (the league average is roughly 88).
In terms of Baseball Info Solutions’ hard-hit ball percentage (which can be found at FanGraphs), Greinke was hit hard the most frequently. Arrieta was best in the group in hard-hit ball rate (22.2%), followed by Kershaw (25.1%) and Greinke (26.6%; the NL average was 29.1%).
While Arrieta is technically superior to Kershaw here, they are roughly 1.5 standard deviations apart in terms of BABIP, while less than 1.0 standard deviation separates their hard-hit ball percentages.
Arrieta also had the benefit of a comparatively better defense behind him. Only four pitchers were hurt by their teammates’ fielding more than Kershaw (Greinke, Max Scherzer, and Brett Anderson), as his fielders cost him about 7.5 runs, according to Baseball Prospectus.
In the case of Arrieta, his fielders did not seem to help or hurt him, as they cost him only 0.24 runs.
Then there is the case of home runs per fly ball, which we would normally expect to not stray too far away from 10.0% in either direction.
Kershaw finished at 10.1%, while Greinke finished at 7.3% and Arrieta posted a rate of 7.8%. Again, this does not seem like a quality of contact matter; Kershaw actually had a marginally higher infield fly per fly ball rate (9.5%), while the other two pitchers were tied at 9.3%.
These factors explain why Kershaw’s FIP, xFIP, SIERA and cFIP are all lower than Greinke’s and Arrieta’s.
Perhaps, though, you’re of the opinion that these defense-independent numbers have more value as predictive stats, rather than descriptive ones. While BABIP and home run rates are unstable and prone to randomness, it is not unreasonable to hold the pitcher accountable for them for awards that judge past performance.
These are reasonable positions to take, but ERA should still not be your metric of choice here. There are simply too many flaws in ERA, most broadly the fact that it does not factor in context at all.
Even if we disregard factors like BABIP we would normally ascribe to “luck,” we still have to acknowledge some pitchers play with worse fielders, play in unfavorable run environments, have the platoon advantage less often, and face stronger competition.
If an awards voter wants to blame a pitcher for bloop hits, that’s his prerogative, but it would be completely unfair to punish him because his catcher can’t frame pitches or he constantly faces elite hitters.
This is where DRA comes in, and the folks at Baseball Prospectus deserve credit because it might be the best “backwards looking” pitching stat available.
As mentioned, Kershaw leads the way in DRA, park-adjusted DRA and DRA-based wins above replacement.
These numbers indicate there was more to Kershaw's (comparatively) higher ERA than ball in play luck, as there were other things he had even less control over.
As mentioned, Kershaw has been hurt by subpar defense, and relative to Kershaw, Arrieta and Greinke have had the benefit of better pitcher framing. In terms of the factors listed at BP, Arrieta and Greinke have a significant advantage in terms of runs added from factors out of their control.
This probably puts Kershaw over the top, as in addition to his advantage in terms of the forward looking defense independent statistics, he also has the edge in DRA.