Dispelling the Rotation Myth

Season Team G GS CG ShO SV BS IP TBF H R ER HR BB IBB HBP WP BK SO FIP ERA
2 year Yankees 341 309 18 6 0 1 1994 8419 1899 916 838 191 660 24 70 61 10 1869 3.56 3.78
2 year Red Sox 248 247 14 5 0 0 1533 6503 1490 695 656 166 507 9 61 44 3 1365 3.82 3.85

These are the aggregate pitching statistics for the Yankees and Red Sox expected rotation (which includes Joba as #5 on the Yankees, and Buchholz as #5 on the Red Sox). As you can see, over the last two years, these numbers favor the Yankees rather heavily. While the bombers have a slight edge in ERA, they have a much larger edge in FIP (which happens to be a lot more useful in projecting future ERA than….well….ERA.

And while it’d be pretty easy to leave it at that, here are the numbers for the two rotations over looks backs of 2-4 years. Blue is better, Red is worse. Looks pretty onesided.

Boston 2Yr FIP 2Yr ERA 4Yr (3Yr) FIP 4Yr (3Yr) ERA New York 2Yr FIP 2Yr ERA 4Yr (3Yr) FIP 4Yr (3Yr) ERA
Lester 3.39 3.31 3.74 3.67 Sabathia 3.12 3.02 3.12 3.11
Beckett 3.42 3.94 3.72 4.05 Vazquez 3.26 3.75 3.46 4.01
Lackey 4.03 3.79 3.63 3.51 Burnett 3.83 4.06 3.85 3.98
Matsuzaka 4.27 3.66 4.14 4.01 Pettitte 3.92 4.36 3.89 4.26
Buchholz 4.77 5.36 4.51 4.92 Joba 3.90 3.92 3.71 3.62

This data is scrubbed in a few ways. First off, italicized players’ longer track records are only three years instead of four. This can be for two reasons–Matsuzaka, Buchholz and Joba simply only have those three years at the major league level, while Pettitte’s last three years have been in the AL (adding in the year in Houston would introduce avoidable bias into the analysis.) Vazquez, of course, does have a season in the NL included in this analysis–which needs to be taken into account, and CC has 1/3 of a season with the Brewers included. You’ll also note that I’ve slotted Vazquez into the #2 slot in the rotation, while it’s presumed that Burnett will be taking that role. His four year FIP data is head and shoulders above Burnett and Pettitte (and in that sample, his one year in the NL has less weighting than in the two year FIP data). Say what you want about where these guys will be pitching coming into the season, performance will dictate their true slot numbers over the course of the season. The data up above suggests that Vazquez is the second best pitcher on the Yankees, and so I’ll slide him in there. The same is true with Lester as Boston’s ace, though he and Beckett are pretty close.

To start off, all five Yankee starters 2-4 year FIP numbers are under 4.00, many significantly so. The same can not be said for the Red Sox’ numbers. The only important number where the Red Sox have an edge is in Lackey versus Burnett’s four year FIP. Now, discount Vazquez’s advantage over Beckett due to his year in the NL–I’d call that a wash, or at worst a slight edge to Beckett. Lackey is a better pitcher than AJ Burnett (though his ultimate ceiling is lower, given that he can’t match Burnett for pure stuff). Joba’s numbers don’t show his downward trend this past year (but also show Buchholz’s consistently below average performance).

Essentially, it looks like the Yankees are stronger in three of the five rotation slots, weaker in one, and one’s a push (Vazquez – Beckett). Add that to the Yankees’ sizable lineup advantage, and the advantage remains heavily with the Yankees. The Red Sox have a lot of ground to make up.