So, in good fun and with respect, Geoff and I have continued to disagree on this subject. For the record, I enjoy Geoff’s blog and have been having a wildly good time of it. I suppose I even got a beat writer to read my blog, which can’t hurt.
I’ll start by saying that, while FIP is an excellent predictor of future performance and future ERA, it doesn’t exactly tell us what happened. Now if we take the FIP’s Geoff gave us on Twitter, i.e. the ’10 FIP’s of the Phillies 2011 top-4 and the top-4 of the ’97 Braves, we get the following: Braves – Greg Maddux (2.43), John Smoltz (3.04) Denny Neagle (3.34) and Tom Glavine (3.96), and the Phillies – Cliff Lee (2.58), Roy Halladay (3.01) Roy Oswalt (3.13) and Cole Hamels (3.67). If we average them, the hypothetical 2011 Phillies rotation has them beat by just .095 or close to a wash. Please note, however, that Tom Glavine beat his FIP sizably in basically each of his peak years. Because of this, despite pitching several seasons in which his skills eroded at the end of his career, his career ERA is still 41 points lower than his FIP.
Why don’t we attack this question in another way, using Wins Above Replacement (WAR) this time? It’s very difficult to know exactly how much better the AL is than the NL now, though it’s probably significant. It’s going to be next to impossible to know how much more difficult it is to pitch in the current AL than the NL of the 90s. So comparing the two rotations in a variety of ways is probably our best bet.
In 1997, the rotation of Maddux, Glavine, Smoltz and Neagle posted WAR of 7.3, 5, 4.5 and 4.1 for a total WAR of 20.9. If we take the WAR of the 2011 Phillies, and use their 2010 WAR for Halladay, Lee, Hamels and Oswalt, we get 6.9, 4.3, 4.7 and 5.3 for a total of 21.2, a surplus of .3 wins – or a wash.
Let’s try it another way; let’s compare each pitcher’s very best seasons and add those up. Halladay’s best season came in 2003 when he posted a 7.3 and won the Cy Young. Lee’s came in 2008 when he threw down a 7.3 and won the Cy Young. Oswalt’s came in 2002 and his best WAR was 6.2. Finally, Hamels’ WAR of 4.7 in 2010 was his career best. In 2011, Hamels will be one year removed from his best WAR, Oswalt (9), Halladay (8) and Lee (3). That’s twenty one (21) total years removed. Also, Hamels will be the only pitcher still in his prime at 27, while Oswalt (33), Lee (32) and Halladay (34) will not be.
In 1997, Maddux was two years removed from his best season by WAR (8.8), which came in 1995. Glavine was six years removed from his 1991 Cy Young season with a 7.4 WAR. Neagle was just one year removed from his 1996 best WAR of 5.1, and so was Smoltz with his 6.1 1996 WAR. All totaled, the 1997 Braves’ rotation was a total of ten (10) years removed from their very best seasons, i.e. their peaks, and likely three of the four men will someday enter Cooperstown. Age-wise, Maddux and Glavine were each 31, and Smoltz was 30 while Neagle was 28.
If we add their career best WAR, we find that the Braves’ rotation has the edge here with a 27.4, the 2011 Phillies rotation with an excellent 25.7. The Braves’ rotation was also 1.5 years younger.
What many people believe is that Lee’s numbers might jump when he’s given a full season in the National League. Will they? It’s tough to say. But we do know this: in 2009, Cliff Lee threw 80 innings in the NL for the Phillies and his ERA+ was 124. Over the rest of his season in Cleveland, his ERA+ was 135. While Lee was pitching in Cleveland, he was in the AL, but not the AL East where the greatest teams reside.
Roy Halladay, on the other hand, had pitched his entire career in the AL East. One would expect a large increase given all that thump and then going to the lowly NL, right? His ERA+ in the AL East in 2009 was 159. When he went to the NL and won his unanimous Cy Young award, it was 165, or an increase of just under 4%.
Look, the 2011 Phillies have a wonderful shot at dethroning the 1997 Braves, the rotation that landed on Bill James’ list of the best 32 rotations of all time. But it won’t be easy, and it’s far from a forgone conclusion. The question really is, will they or won’t they (exceed the ’97 Braves in brilliance, which is our standard). It’s a wonderful question to ask, ponder and try to answer. Unfortunately, we can’t answer it; that’s up to the players. But when I hear people speculate that this is the most likely outcome, it sort of makes me cringe. I sort of think the incumbent deserves the edge, seeing as how, you know, they’ve already done it. Just a little more respect, that’s all. Call me when the Phillies trade for Zack Greinke.
Stats come from Baseball-reference
Insightful baseball analysis for the casual crowd, Moneyball masses and you, the Third and King crew
The problem with FIP (or xFIP, which some people think is better or tERA, which a few thinks is better) is that a key underpinning of that is the assumption that pitchers will regress to a BABIP of roughly .300.
ReplyDeleteHowever, as Tom Tippett showed in his landmark study of DIPS theory, there are pitchers in the history of baseball who can control their BABIP, starting with knuckleballers but also including a category he labeled, "Crafty Lefties".
Thus FIP, while working for the vast majority of pitchers, and well, it fails for pitchers who truly have a good skill for a pitcher, avoiding hits. Like Cain and Zito. (I know Cain is RHP, but I guess he's a crafty righty because his BABIP has been much below the .300 mean everyone is suppose to regress to).
And like Glavine.
TangoTiger, on his The Book blog/website noted that it takes around 6-7 full seasons worth of starts/IP for there to be enough BIPs to say that a BABIP is significantly below the .300 mean. Cain is way below and nearing 6 seasons, Zito is way past that, though he appears to be less effective in preventing hits while with the Giants.
Nice debate, thanks for sharing.