Hiroki Kuroda’s Case for the Cy Young Award

When it comes to thinking about the best pitchers in the American League this year, the names that jump out to most might be along these lines: Felix Hernandez (always awesome and having one of his better seasons), Max Scherzer and Anibal Sanchez (two non-Verlander Tigers having big years), Yu Darvish (strikeout machine), Chris Sale (great pitcher on a bad team), and Derek Holland (finally living up to his stuff in a hitter’s park). One pitcher that would probably come up less frequently is Hiroki Kuroda. He probably would not be totally ignored (especially by Yankees fans), but he has flown somewhat under the radar (has never made an All-Star game) for whatever reasons, some having to do the nature of his performance, some not.

External factors aside, an examination of Kuroda’s 2013 performance shows why he might not get as much attention for the year-end awards, but a deeper look also reveals his worthiness.

Since you are reading FanGraphs, you probably already know that at the moment, Hernandez leads American League pitchers in Wins Above Replacement (4.8), closely followed by Scherzer and Holland (4.5 each). Kuroda is currently a distant eighth at 3.2 WAR. While WAR is not the sort of metric where one should quibble over half of win, Kuroda is not just a half-win away — he is closer to two wins away from the leader.

FanGraphs’ implementation of WAR for pitchers uses a version of FIP as the basis for pitcher value. However, if one looks at WAR based on straightforward Runs Allowed (RA9-WAR), Kuroda leads the league at 5.3, just edging out Hernandez (5.1) and Bartolo Colon (5.0). As just said above, the difference between 5.3 and 5.0 really is not decisive, but in terms of RA9-WAR, Kuroda has been just as valuable as Hernandez this year, which certainly would put him into consideration for the Cy Young.

Some will argue that RA should be used for WAR and stuff like determining awards anyway. That is longstanding debate, and I will not address that here. My argument here will be that Kuroda has a case for Cy Young consideration (assuming his performance thus far continues) even if one does accept that FanGraphs WAR for starting pitchers is the right way to go.

Most of those (myself included) who believe that a DIPS-based metric is best for pitching realize metrics like FIP have limitations. Knuckleballers, for example, tend to be exceptions to the persuppositions of defense- and batted ball-independent metrics. Even for non-knuckleballers, as their career grows longer (I have seen this as anywhere from three full seasons of starting to 1000 innings), RA becomes a better guide to their value. While stats like BABIP and strand rate may not relate to true talent in small (even one full year) samples like strikeout and walk rate, this is not to say that they represent no skill. There is less variation in skill among the population for those metrics, but that does not rule out any variation at all. It simply takes a larger (often much larger) sample to get an idea of which players stand out. (This works out differently for various metrics, but I will not get into those details here.)

While this relates to “true talent” as opposed to observed performance, it is not quite the same. When we fans gets into debates about, e.g., “who was the AL’s most valuable pitcher in 2013,” we are not talking about talent, we are talking about observed performance. The problem is that we often have trouble, particularly with pitchers, in figuring out which parts of players true performances should be credited to them.

If Kuroda was a one-year wonder when it came to outpitching his peripherals, I would probably not give him much thought. But as in my Phil Birnbaum-inspired discussion from last year of Clayton Kershaw and R.A. Dickey, we can use past performance to get an idea of how much we might credit stuff like BABIP to a pitcher’s observed skill as opposed to random variation (also known as “luck”).

Kuroda is outperforming his FIP by a long shot this year (2.45 ERA versus 3.33 FIP), hence his high RA9-WAR. It is mostly due to his very low BABIP (.261) and high left on base percentage (82.7), represented in his FDP stats as being worth 1.3 and 0.9 wins, respectively. But, as said above, this is not out of nowhere for Kuroda. For his career, his 3.29 ERA is better than his 3.58 FIP, although not by tons. Starting in 2011, the difference in more pronounced, though: 3.01 ERA versus 3.69 FIP over almost three full seasons of pitching (568 innings). While his career left on base percentage is not exceptional, since 2011 it has been almost 80 percent. Without getting into what may or may not have changed for Kuroda in 2011, that is worth noting. This is not to say that no luck was involved, but the sample is not negligible, either. Moreover, Kuroda has always been a low BABIP pitcher, with a career .279 mark. That is still higher than this season’s .261 mark, but after more than one thousand career innings pitched, I think it is fairly safe to say that Kuroda probably is not some true talent .300 BABIP pitcher who just has been getting really fortunate.

Now, as was pointed out to me, Kuroda does seem to have had exceptionally good fortune with runners in scoring position this year, giving up just a .163 BABIP and .177 wOBA. It is a small enough sample of his performance that it makes it more difficult to say it is just him. On the other hand, Kuroda does have a history of pitching better with in scoring position for his career (.277 wOBA as opposed to .289 with the bases empty, (.253 BABIP with runners in scoring position as opposed to .279 overall career). Since 2011, Kuroda has a very low BABIP with runners on (.221) as opposed to overall (.278).

That does mean I believe Kuroda should full credit for his 2013 BABIP with runners in scoring position. There is probably some skill involved, although I do not know how much. Even if there is some skill, Kuroda is stranding runners this year in a way that suggests he is getting particularly lucky. Still, this should not lead us to dismiss his performance. After all, every metric, even those whose observed realization are close to true talent like strikeout and walk rate, are subject to random variation and luck.

For example, Felix Hernandez is putting up career-best rates in strikeout and walk rates, and I would guess that at least some of that is random variation. Now, they are far more likely to reflect his individual contribution and thus he should get credit for to a greater extent than Kuroda’s performance with runners in scoring position this year. This is true not only because of stuff like the general year-to-year correlation of such metrics but also because of the logic behind DIPS (what a pitcher individually controls). Still, given Kuroda’s career-long low BABIP and his last few years of outpitching his peripherals in general, it would be foolish to simply throw it all out as good luck.

Does this mean that I would vote for Kuroda for Cy Young if I had a ballot and if the vote were today? Well, probably not. Even without attributing it all to luck, I do think that there is probably a fair bit of random variation in Kuroda’s Ra due to FDP factors. Moreover, it is not as if he is blowing away the field when it comes to RA9-WAR — Hernandez, Bartolo Colon, Yu Darvish, and maybe even Max Scherzer are close enough that Kuroda’s lead should not be overemphasized, and Hernandez and Scherzer, in particular, have been far better in DIPS terms. So the answer is no, I probably would not make Kuroda my Cy Young vote. However, I do think that Kuroda should be in the conversation, and would not mock anyone who did vote for him.