Many narratives written about the AL Cy Young balloting have tended to present it as a clash between sabermetric, Moneyball-style thinking and a Murray Chase-style obsession with W-L records. The reality, as many people have pointed out (Keith Law and Dave Cameron come to mind), is that the voters prioritized basic pitching statistics like strikeouts, walks, ERA, innings pitched, and opponants' batting average over won-loss record to an unprecedented but unsurprising degree. While Felix Hernandez won the Cy Young because he dominated these traditional "baseball card" stats, many writers have noted that Felix probably would not have won if the vote turned on stats like WAR, FIP, tRA, or SIERA. Matthew, for example, said in the latest podcast that he probably would have voted for Cliff Lee or Justin Verlander over Felix. That made me curious to see just what the sabermetric stats have to say about the Cy Young contenders, so I made a spreadsheet with the key pitching statistics that a sabermetrically-inclined voter would consider.
I posted the chart I made below these definitions of all of the stats, so skip to that if you know this already.
xIP = expected innings pitched (www.statcorner.com)
tRA = "tRA involves assigning run and out values to all events under a pitcher's control and coming up with an expected number of runs allowed and outs generated in a defense and park neutral environment. tRA is on a R/9 scale." Graham and Matthew's introduction to it may be found here. (www.Statcorner.com)
tRA+ = Like ERA+, but for tRA. I can't find the specific article about it, but in general, an ace will tend to have 118 tRA+ or above, a #2 starter will have 106-118 tRA+, a #3 will have 94-106, a #4 will have 82-94, and a #5 starter will have below 82.
lgTRA = the American League's overall tRA (www.statcorner.com)
pRAA = pitcher runs above average (www.statcorner.com)
ERA = Earned Run Average (everywhere)
(f)tERA = FanGraph's version of tRA scaled to ERA (FIP and xFIP are likewise scaled to ERA). tERA = tRA x 0.94. The batted ball input data that tERA uses is slightly different from Statcorner's batted ball data (e.g. one might classify something borderline as a line drive while the other calls it a ground ball) because each gets its batted ball data from a different company. (www.fangraphs.com)
(f)FIP = FanGraph's version of Fielding-Independent Pitching, which is scaled to ERA and measures only stats the defense cannot affect (strikeouts, walks, home runs, HBP). Other sites use slightly different FIP formulas, but FanGraph's is HR*13+(BB+HBP-IBB)*3-K*2)/IP, plus a league-specific factor that scales FIP to match league average ERA for a given season and league. (www.fangraphs.com)
tRA-WAR = Statcorner's wins above replacement, as determined by a tRA-based formula (www.statcorner.com) (By the way, I found an odd thing: if you collapse a season like Cliff Lee's (two teams) or David Price's (he had one relief appearance) then you drastically lower the guy's WAR total. Cliff Lee's went from 4.1+3.9 to 5.8, for example. Is this intentional or a glitch, Matthew or Graham?)
fWAR = FanGraphs' wins above replacement, as determined by a FIP-based formula adjusted for park factors (www.fangraphs.com). In general, players' fWAR tends to be slightly higher than their rWAR because fWAR estimates the total WAR in MLB to be about 1000, whereas rWAR estimates it to be about 870 (for more, see this discussion on The Book Blog). For an introduction to WAR, see Dave Cameron's series on FanGraphs.
rWAR = Baseball Reference's wins above replacement, as determined by an ERA-based formula adjusted for the quality of the defense behind a pitcher. I can't actually find a good introduction to rWAR on www.baseballreference.com or Sean Smith's www.baseballprojection.com, but it's around somewhere.
AvgWAR = the average of a player's tRA-WAR, fWAR, and rWAR
t-lg rank = how close a pitcher was to having the highest tRA-WAR in the AL
r-lg rank = how close a pitcher was to having the highest rWAR in the AL
f-lg rank = how close a pitcher was to having the highest fWAR in the AL
|2010 AL Pitcher||xIP||tRA+||pRAA||lgTRA||tRA||ERA||(f)tERA||(f)FIP||(f)xFIP||tRA-WAR||fWAR||rWAR||AvgWAR||t-lg rank||r-lg rank||f-lg rank|
As you can see, there's no clear-cut winner, and you can make a reasonable case for any of the top seven (Lee, Weaver, Hernandez, Sabathia, Liriano, Verlander, and Lester) depending which WAR statistic you give the most credence to. Cliff Lee, for example dominates tRA-WAR and fWAR but gets hit hard for his high ERA by rWAR. By avgWAR, Cliff Lee, Jered Weaver, and Felix Hernandez are the top tier, while Sabathia, Liriano, Verlander, and Lester are the second tier and Lewis, Price, Buchholz, and Cahill make up the third tier. Felix and Cliff Lee seem like the two most highly rated pitchers (Cliff Lee is 1st, 13th, and 1st, respectively, while Felix is 3rd, 1st, and 3rd). But it's up to you to decide how much weight to give each statistic and who should win the Cy Young. What do you think?