September 12, 2011
Prospectus Hit and Run
Opponent Quality and the AL Cy Young Race
The past few weeks have seen Justin Verlander solidify his standing as the AL Cy Young favorite as the Tigers have pulled away from the AL Central pack. Not that the 28-year-old righty has been particularly dominant lately. His home-run rate over his last nine turns (1.40 per nine) has been more than double what it was for his previous 22 (0.65), but because he's otherwise done a fine job of keeping opponents off the bases while receiving more than six runs per game of offensive support, he has collected wins in each of his last 10 turns. His 22 victories have matched the majors' highest total since 2003, and even for a voting body that's shown its evolution by awarding consecutive Cys to Zack Greinke (16 wins) and Felix Hernandez (13 wins), that's the kind of thing that will be noticed when ballots are cast.
Should Verlander have such a stranglehold on the award? Look past the won-loss records, and there's considerably less separation between him and two other top AL candidates:
Verlander has the slight edge in innings and strikeouts, but he's in a dead heat for ERA, and trailing Sabathia in FIP. Once you control for sequencing and defensive support via our new version of Fair Run Average, he falls behind Weaver. His advantage in innings still gives him the overall edge in WARP, but it's not as clear-cut as you might think given the volume of awards chatter.
One measure not factored into the computations above is the quality of opposing hitters each pitcher has faced. Here at BP, we do track that sort of thing, of course. Sabathia pitches in the rough-and-tumble AL East, a division where three other teams besides the Yankees own positive run differentials (at least through Saturday). Verlander, by contrast, has the luxury of pitching in the AL Central, where none of his opponents have a positive run differential, and two of them, the Twins and White Sox, rank among the league's most inept offenses. Weaver's division, the AL West, features a pair of low-wattage offenses in Oakland and Seattle.
The spread in Quality of Opposing Batter Faced isn't very big; among the 104 pitchers with at least 140 innings—ERA qualifiers, basically, with an allowance for those a single start away from joining the ranks—the spread is 40 points of OPS from the toughest slate (David Price, .267/.335/.433) to the easiest (C.J. Wilson, .260/.323/.406), and the standard deviation from the median is just seven points of OPS. The cumulative line of the hitters Sabathia has faced is .265/.330/.425, while for Weaver it's .264/.328/.414, and for Verlander it's .262/.326/.412. Those two lines are 15 and 17 points of OPS lower than that of Sabathia, respectively, more than double the standard deviation, enough to take notice. However, if we turn to what those pitchers have actually yielded against their respective slates of hitters, the difference comes out in Verlander's favor, and not by a little. The Tiger holds 60-point advantages over the Yankee in both batting average and on-base percentage allowed, and a 41-point edge in slugging percentage, while against the Angel, he's got advantages of 20, 20 and 12 points in the slash stats:
That last column I'm calling Opponent Quality Plus. The world probably didn't need another pitching metric, but as the Rolling Stones famously sang, "You can't always not get what you don't want," and here I come bearing gifts. Opponent Quality Plus is an OPS+-type figure in which I've used each pitcher's opposing hitter line as the "league average" in the typical OPS+ formula: 100 * (OBP/lgOBP + SLG/lgSLG - 1). I've divided by the Pitcher Park Factor (also in that report) to make a slight additional adjustment. A 100 would mean that a pitcher performed essentially as expected given the quality of opponent and park, while a 90 would be 10 percent better than average (for a pitcher), and 110 percent would be 10 percent worse. In this case, Verlander is 50 percent better than average, considerably ahead of not only Weaver and Sabathia, but also the rest of all other major-league ERA qualifiers. Here's what the top 20 looks like:
I make no pretense that this is the be-all and end-all of pitching stats, but it does help to capture the level of difficulty each pitcher has faced and how well he has conquered those hitters. Note that this applies no correction for the luck and randomness involved in batting average on balls in play or home runs per fly ball, two areas that can have significant effects on those "against" lines. As this is a rate stat, there's also no correction for differences in the number of innings, either. Like I said, imperfect stat.
Through this lens it seems abundantly clear that Verlander has done enough to separate himself from Weaver and Sabathia to merit the Cy Young; while he has faced an easier slate than either, he has disemboweled them more convincingly. This metric also helps to build Kershaw's case in the NL. As great as Halladay and Lee have been, the 23-year-old southpaw has been even more stifling, thanks in large part to his ability to suppress slugging percentage. For what it's worth, he enjoys a margin of 0.7 WARP over Lee, and 1.1 WARP over Halladay, so it's not as though this bathtub metric is producing results entirely out of left field.
Up and down that leaderboard, there are a few interesting comparisons and contrasts to be had. As noted before, Price has faced the majors' toughest slate of hitter, thanks in large part to his status as the nominal ace of a team stuck in the division with the league's two top offensive powerhouses. Nonetheless, among AL East starters, Beckett been more effective relative to his competition, and among Rays starters, Shields has done the better job of rising to his level of competition. Over in the National League, the nearly 20-point advantage in OBP against that Hamels has compiled versus a slightly easier slate than either Halladay or Lee is enough to push him into the majors' top five in this measure. Cain has outpitched teammate Lincecum, as has the Diamondbacks' Kennedy—a point I made on Twitter last week—particularly once ballparks are considered. Kennedy has also outpitched the man whom he's replaced at the head of Arizona's rotation, Haren.
A quick peek at the other end of the leaderboard:
Ladies and gentlemen, we've found a new way to express just how awful Lackey has been. Relative to Boston's other starters, he's faced an easy slate; Beckett and Jon Lester have both faced collections which rank among the majors' 15 most difficult, while Lackey is 52nd among the 104 pitchers who qualify, and yet he's been smoked so badly by that lot that he ranks last in the majors. He's not the only starter on a playoff contender among this sorry lot, either; the Yankees have Burnett, who's not very likely to draw a start, and the Tigers have no less than three in Penny, Porcello and Scherzer—two of whom will help round out their post-season rotation.
Wait, wasn't I just talking the Tigers' rotation up? Indeed I was, noting that their weighted FIP for their prospective playoff rotation was the second-best in the AL behind the Yankees. As noted before, this method makes no allowance for stripping out the effects of high BABIPs; the members of that aforementioned trio are 14 to 26 points above league-average in that department, so they do less well on a purely results-based measure—and here we'll note that, for example, their starters are a mid-pack eighth out of 14 teams in ERA—than a skills-based one.
In any event, Colin Wyers has indicated his eventual intention to factor quality of opposing hitter and pitcher into our new metrics where appropriate, so this little bit of metric moonshine may not make it into our official reports, but I'll leave you with a leaderboard going back to 1950. No surprise who tops the list—we've been talking up the season in question as one of the greatest for awhile now—but whoa: