Right now, the Phillies suck. You know it, and I know it, and since we don't like to talk about terrible things, let's discuss something more positive. Back in 2011, the Phillies notably didn't suck — they won 102 games, the most in the major leagues. For one final, glorious year, the Philadelphia dynasty lived on.
But that Philly squad differed from the ones that came before it. In 2011, the Phillies didn't get that much production from their position players, who combined for 20.6 fWAR. While that was certainly a respectable total — it ranked 12th in baseball — it was the team's lowest since 2000. Rather, it was the Philadelphia pitchers that elevated the team to greatness, particularly the starters. In the preceding offseason, Cliff Lee joined Roy Halladay, Cole Hamels, and Roy Oswalt* to form an incredibly fearsome foursome.
*And Vance Worley/Kyle Kendrick/Joe Blanton, but those guys don't matter.
Oftentimes, offseason headlines can come back to haunt teams. When everyone from Dave Cameron to Jayson Stark to Dayn Perry thinks that you might have the best rotation of all time, you might find it hard to live up to those expectations. The Phillies had no such difficulty in 2011: As Josh Goldman wrote at FanGraphs that November, they finished with the best FIP- since 1900, as well as the 11th-best ERA-. By both predictive and descriptive metrics, the club's starting corps truly earned its keep.
ERA- and FIP- don't work as well as they should, though. That's why we have new metrics for pitcher evaluation — DRA and cFIP. While they may be more complicated than their predecessors, they can better reflect how well a pitcher pitched, and that's what matters for us. So how do these newer, shinier statistics feel about the 2011 Phillies rotation? Can it withstand the test of time?
Let's start with DRA, which attempts to model runs allowed. Phillies starters posted a 74 ERA- in 2011, but their DRA- lagged behind, at just 83. That means that, while they had an ERA 26 percent better than average, their DRA was only 17 percent better than average. With that said, this squad still stands apart historically — they're one of only 26 teams with a DRA- of 85 or lower:
DRA dates back to 1953, so this sample contains 1,574 clubs. Finishing here puts Philadelphia in the 99th percentile of recent history, which bears out their dominance. 11th all-time in adjusted ERA, and 12th all-time in DRA is an accomplishment that takes anyone's breath away.
When it comes to cFIP, things get a little bit trickier. Recall that the Phillies had the best FIP- in the modern era, at 76. Their cFIP of 84 is marginally less impressive, as it knocks them down to 24th:
cFIP extends three years beyond DRA, to 1950; this broadens our sample to 1,622 teams. Even that expanded competition can't explain the moderate dropoff from Philadelphia. What's to blame?
The "c" in cFIP stands for contextual. This means, in simple terms, that it takes into account things that normal FIP ignores. Two of the bigger elements are park factors (which FIP- doesn't customize per individual park) and opponent quality (which FIP- doesn't include whatsoever). In each of these areas, the seven Philadelphia starters benefited from some good fortune:
(Innings pitched as starters; opposing TAv and park factors overall.)
The MLB average for TAv is always .260, meaning every starter save Worley faced easy competition. As for the park factors, they likely benefited from easy trips on the road — keep in mind that the Phillies played nine games in pre-renovation Citi Field, in addition to nine contests in Turner Field. FanGraphs doesn't go into specifics on the road games that teams play, instead cutting each park factor in half. Thus, because Citizens Bank Park inflated offense by two percent in 2011, each Philadelphia pitcher received a park factor of 101, even though their actual experience was much easier than that. In both environment and adversaries, the Phillies got pretty lucky, which inflated their FIP- and which cFIP corrected for.
In the end, the 2011 Phillies still possessed one of the best rotations in recent memory. Their results, as measured by DRA, were incredible, and even with the greater context of cFIP, their peripherals made them elite as well. Altogether, this makes a pretty good case for the more advanced metrics: not only do they usually confirmwhat we already know, they can also provide more background and more accuracy than their cruder counterparts. Philadelphia had a truly great rotation five years ago, and now we can more fully appreciate it.
. . .