I tend to follow up a reasoned debate like the one that inspired me to write about the value of league average performance by diving into stats to find facts to back up my view. And if I find them, I obnoxiously pepper my interlocutor with them via e-mail.
This time, searching for ways to get across the idea that league-average performance is pretty damn good, I stumbled across something that I think illustrates the point beautifully.
Before I get to that, I want to make clear that this league-average is pretty good business is nothing like an original thought of mine. It’s sabermetric gospel. As Jay Jaffe pointed out on Twitter, it’s fundamental Bill James observation #2. In the 1988 “Baseball Abstract,” James wrote: “Talent in baseball is not normally distributed. It is a pyramid. For every player who is 10 percent above the average player, there are probably twenty players who are 10 percent below average.” I don’t think you can really understand roster construction without understanding this point, and a whole lot of people understand roster construction just fine.
But a lot don’t, even among some pretty hardcore baseball fans — and writers.
So I decided to look at the pitching of the Philadelphia Phillies and the San Francisco Giants, the National League’s best regular-season team and its champion, both of which had very good starting pitching. One of the things that got me thinking about these two teams was a line in a national story about the Giants late in the season that mentioned them having “the best rotation this side of Philadelphia.”
As a Giants fan, this struck me as odd because there’s no way I would have traded the Giants’ five-deep rotation for the Phillies’ three stars and pray for rain. At the time, the Giants’ worst pitcher was Barry Zito, who had been struggling since early August after pitching well for the first four months of the season.
And this is where I’ll tell you that the league-average guy my friends and I were arguing about was Barry Zito. Except for his terrible 2008, when his ERA ballooned to 5.15, and he really was legitimately that bad, Zito has been a roughly league-average pitcher for his four years in San Francisco.
In 2007, the National League ERA was 4.44, but for starting pitchers it was 4.64. Zito’s was 4.53. In 2009, N.L. starters had a 4.30 ERA, the league ERA was 4.20 and Zito’s was 4.03. This year, Zito’s 4.15 ERA was a bit higher than the league ERA of 4.03. Starting pitchers’ ERA was unusually close to the overall figure, 4.05.
The problem is, Zito is a terrible guy to use to illustrate the idea that league average is pretty good because Zito makes a ridiculous $18 million a year on a seven-year contract. Of course he’s a terrible disappointment. He hasn’t lived up to that deal, but then, almost no one could have. It would have been an insane deal even if Zito were a better pitcher than he’s ever been.
Also, Zito got to a little worse than league average in 2010 by pitching beautifully for four months, then pitching horribly for two. He had a 3.35 ERA through his first 23 starts, and then for the rest of the year, starting Aug. 11, he had a 6.66 in 11 starts and a relief appearance. He was dreadful. Not the best poster boy for “pretty good.”
But you have to count the good four months. Overall, Zito was an important part of a terrific staff. The Giants’ starters combined for an ERA of 3.54, second best in the league behind the Cardinals’ 3.50. The Phillies were right behind the Giants at 3.55. But look how they got there.
The Giants got 34 starts from Jonathan Sanchez, 33 each from Zito, Matt Cain and Tim Lincecum, 18 from Madison Bumgarner and 11 from Todd Wellemeyer, the only one of the bunch who didn’t pitch well. Here were their ERAs as starters — Zito had the one relief appearance and Wellemeyer had two:
Bumgarner: 3.00
Sanchez: 3.07
Cain: 3.14
Lincecum: 3.43
Zito: 4.13
Wellemeyer: 5.82
The Phillies got 33 starts each from aces Roy Halladay and Cole Hamels and 12 from Roy Oswalt, another elite guy who came over in a midseason trade from Houston. The rest of the starts were Kyle Kendrick 31, Joe Blanton 28, Jamie Moyer 19, J.A. Happ 3, Vance Worley 2 and Nelson Figueroa 1. Their ERAs as Phillies starters:
Oswalt: 1.65
Happ: 1.76
Worley: 1.80
Halladay: 2.44
Hamels: 3.06
Figueroa: 3.60
Blanton: 4.74
Kendrick: 4.81
Moyer: 4.84
OK, you want to see the value of league average?
The Giants got 44 starts from pitchers whose ERA was below the league average of 4.02, although 33 of those were by Zito, who had the lowest below-average ERA in the league. In other words, Zito was the best below-average pitcher in the N.L.
The Phillies, though, got a whopping 78 starts — almost half their season — from Kendrick, Blanton and Moyer, none of whom were as bad as the Giants’ Wellemeyer, but all of whom were well below league average. And the Phillies still almost matched the Giants and were third in the league in starting pitcher ERA.
But look what would happen if you replace that trio’s 464 innings with league average pitching: The Phillies starting ERA would drop to 3.22. There hasn’t been a team anywhere near that figure this century.
And what was standing between the Phillies, with their trio of elite starters, and that historically great rotation? It was the lack of two and a half league-average starters, three for half the season before they got Oswalt, and then two for the second half.
If the Phillies had managed to have, say, Bronson Arroyo, Derek Lowe and Zito, the three pitchers who straddled the league-average ERA, instead of Blanton, Kendrick and Moyer, they’d have had a starting rotation for the ages.
And all the stories about that great rotation — and there would have been a lot of them — would have focused on Halladay, Hamels and the midseason pickup, Oswalt. Arroyo and Co. would have gotten a mention, but they’d be lesser characters, just keeping the mound warm for those two days between Oswalt and Halladay.
But we know better, don’t we? It would have been Arroyo, Lowe and Zito, three league-average guys, who made the group historically great. After all, Halladay, Hamels and Oswalt were actually there in 2010, and the Phillies rotation wasn’t great. It was only very good — third best in the league.
With league-average starters the rest of the time, instead of poor but not Wellemeyer-ishly terrible starters, they would have been the best rotation since the mid-’90s Maddux-Glavine-Smoltz Braves.
* * *
Moyer photo by Hounddiggity/Creative Commons license