February 22, 2011
In an article I wrote for my blog recently, I wondered why we neglect to dock Matt Cain "points" for pitching in AT&T Park, a pitcher-friendly ballpark, while at the same time being very mindful of Coors Field in evaluating the MVP award candidacy of Carlos Gonzalez. During the 2010 season, Gonzalez posted an OPS of 1162 home, but a more average-looking 775 OPS on the road. Similarly, Cain has a career 3.19 ERA at home but 3.76 on the road.
My point for writing this is not to criticize Cain for his home/road splits; I bring it up only to explain the inspiration for this research. Instead, I am interested in finding out how he has evaded his ERA retrodictors so consistently. Since becoming a regular in the Giants' starting rotation back in 2006, his SIERA has ranged from 3.90 to 4.23 but his ERA has ranged from 2.89 to 4.15 (his rookie season).
His home stadium, AT&T Park, which had a home run park factor of 82 for left-handed hitters in 2010, has helped him out quite a bit over the years (SIERA and other run estimators do not factor in park effects) but he has performed better than expected on the road as well. In discussing this issue with commenters on my blog and with the fine folks on Twitter, the consensus seems to be that he is a master at getting the weak fly ball.
From 2008-10, Cain has allowed the fifth-most fly balls (over 44 percent) among Major League starting pitchers. His HR/FB rate ranged from 6.8 to 8.4 percent. Even on the road, Cain has been stingy with the home run, as his career average HR/FB is only 7.4 percent and 6.7 percent at home. With nearly 6,400 total balls put in play, we certainly have a decent enough sample size to conclude that he has an ability to induce weaker contact. But is it necessarily true?
Cain's career BABIP is .270, 30 points lower than the .300 average for pitchers. Contrary to popular belief, the bulk of the difference does not come from fly balls. Cain's ground ball BABIP is 21 points lower, fly balls eight points, and line drives six points.
Can we credit the Giants' infield defense over the years for helping lower Cain's ERA? Park-adjusted defensive efficiency (PADE) tells the story. In 2010, the Giants led the National League, converting over two percent more ground balls than the league average, given AT&T Park. In '09, the Giants led all of the Majors with a PADE approaching 3.0. Aside from '08, the Giants have been at or near the top during Cain's career.
No, Cain does not induce many ground balls, but when he has, they have been converted into outs at a rate much higher than normal (for example: in 2010 17.7 percent of Cain's plate appearances ended in a ground out--Zack Greinke, who had a G/F ratio nearly twice that of Cain, was at 19.3 percent). That means fewer runners are reaching base, fewer are advancing on one of his many fly balls, and more runners are being stranded.
In addition to the groundballs, from 2006-10, Cain threw the fourth-highest percentage of infield fly balls at 13 percent. Almost all of the pitchers at the top of the IFFB rate list have lower-than-normal BABIP. The top-ten range from .265 to .284. The particular pitch with which most pop-ups are generated is Cain's fastball. Since 2008, 134 pop-ups have come from the fastball; 17 by slider; 12 by change-up; and eight by curve ball. The fastball has accounted for over 78 percent of infield flies.
Furthermore, in an e-mail with our own Mike Fast, he pointed out to me that a majority of Cain's outfield fly balls are hit to the opposite field. Opposite field fly balls tend to be weaker and hit towards the deepest parts of the ballpark, which goes a long way towards suppressing home runs as well as hits in general.
I think we now have enough information to conclude that Cain's over-performance of his ERA retrodictors is legitimately assisted by three factors:
So long as he continues to pitch in a spacious home ballpark, has a great infield defense behind him, and keeps his fastball at a consistently elite level, Cain should continue to defy expectations. Saberists will continue to underrate him (as I did prior to this research). Cain may be a lot of things, but he is not a fluke, despite what his ERA retrodictors say. There are exceptions to every rule, and Cain has proven himself to be one of those exceptions.