Cliff Lee vs. league-average

Cliff Lee of the Phillies
Cliff Lee of the Phillies

A Facebook friend asked, in light of last week’s post headlined “For the want of league average, greatness was missed,” what the signing of Cliff Lee means to the Phillies. That is, “I would be interested to see if the numbers project them as a rotation for the ages,” he wrote.

Well, I think yeah. The numbers clearly project them as a rotation for the ages. Whether they’ll be one, who knows. Joe Posnanski talked to Bill James about that question. Jayson Stark huddled with Leo Mazzone and Davey Johnson and compared expectations of the 2011 Phillies staff to the greatest rotations since World War II.

Go read them if you want to read something smart about the 2011 Phillies starting rotation.

But what I wondered — and what I thought my Facebook friend asked on first reading his question — is whether adding Lee to the 2010 Phillies rotation would have made it one for the ages.

You’ll recall that, with Roy Halladay, Cole Hamels and half a season of Roy Oswalt, the Phillies’ rotation was quite good, a close third in the National League by ERA. The problem was the rest of the starting staff: Kyle Kendrick, Joe Blanton and Jamie Moyer, all of whom pitched poorly, plus a few decent spot starts from J.A. Happ, Vance Worley and Nelson Figueroa.

Replacing Kendrick, Blanton and Moyer’s 78 starts with league-average pitching would have dropped the Phillies starters’ combined ERA from 3.55 to 3.22, which would have been the lowest in the N.L. since the already legendary 1998 Atlanta Braves starters put up a 3.06. But what if we just replaced one of those guys with Cliff Lee? Would that be enough? How does adding one great pitcher to a staff that’s half elite and half poor compare to replacing the poor half with average pitchers?

Kyle Kendrick would have been the odd man out if Lee had signed with the Phillies before 2010, so what happens if we replace his 177 and two-thirds innings and 4.81 ERA in his 31 starts with Cliff Lee?

But what numbers do we use for Cliff Lee? We can use his actual 2010 numbers with Seattle and Texas. He made three fewer starts than Kendrick but tossed 34 and two thirds more innings, thanks to his astonishing average of 7.58 innings per start. We can add three starts of similar pitching to Lee’s total to match Kendrick’s 31 starts. That would bring him up from 212.1 to 235 innings. Either way, because the difference is so small, substituting Lee’s numbers for Kendrick’s we’re left with an ERA of 3.26 for the Phillies starters.

Cliff Lee of the Rangers
Cliff Lee of the Rangers

That’s for the ages, all right, also the best since the ’98 Braves, but not as good as the 3.22 we came up with by using all league-average starters in place of the three Phillies’ non-aces.

Adjusting for the American League’s higher run-scoring environment wouldn’t change much because the A.L. didn’t score that much more than the N.L. this year. The N.L. ERA of 4.02 was 97 percent of the A.L.’s 4.14. For starting pitchers the N.L. ERA of 4.07 was 95 percent of the A.L.’s 4.27. Using either multiplier to take a few earned runs away from Lee, we get the Phillies starters’ ERA down to 3.23.

What about adjusting for ballpark? Lee made six home starts in Seattle, where Safeco Field is a pretty extreme pitcher’s park, and seven home starts in Texas, where the Ballpark is favorable to hitters, though not to the same extreme. Despite its reputation and early history as a hitter’s haven, Citizen’s Bank Park in Philadelphia has played as roughly neutral the last few years, according to both Baseball-Reference.com and “The Bill James Handbook.”

Happily for me, because I wouldn’t be able to figure it out anyway, it doesn’t look like there’s any way transferring Lee’s 2010 numbers from Seattle and Texas to Philadelphia would lower his ERA by much because of park factors.

Maybe we should look at the ever-so-brief time when Lee actually pitched in Philadelphia, in the second half of 2009. He started 12 games and had an ERA of 3.18 in 79 and two-thirds innings. If we take those same numbers and extrapolate them out to Kendrick’s 31 starts, we’d have a Phillies starting ERA of 3.30.

Well, listen, it looks like any way I know how to slice it — which, let’s be clear, probably leaves out a lot of really smart ways of slicing it — replacing Kyle Kendrick with Cliff Lee would have had pretty much the same effect on the starting rotation’s ERA as replacing Kendrick, Blanton and Moyer’s starts with league-average performance.

That’s interesting, isn’t it?

Cliff Lee of the Indians
Cliff Lee of the Indians

But what about cost? Lee is awfully expensive. Which costs more, picking up an elite player to go with two poor ones or replacing three poor ones with average players? That’s an easy one to answer. I don’t know.

But we can compare the actual guys we’re talking about. Lee’s new contract will reportedly pay him an average of $24 million a year, so let’s use that figure. With Lee as a starter for the 2010 Phillies — making his 2011 salary — the other full-season starter would have been Blanton, with Moyer probably filling a spot for half the year till the Oswalt trade.

Cot’s Contracts says Blanton signed a three-year, $24 million deal before the 2010 season, with annual salaries of $1 million in 2010, then two years of $8.5 million, plus a $6 million signing bonus. I’d call that $7 million for 2010. Moyer made $8 million, various online resources tell us.

So Lee, Blanton and Moyer would cost $39 million in our imaginary world, $35 million if we could jettison Moyer and his contract upon getting Oswalt.

The four starting pitchers who straddled league average ERA in the N.L. this year were Bronson Arroyo and Derek Lowe below it — that is, better than average — and Barry Zito and Randy Wolf above, all veteran guys, meaning they’re expensive. Zito’s got a ridiculous contract, so let’s use the other three as our league average guys to replace Blanton, Moyer and Kendrick. Arroyo made $11 million in 2010, Wolf $9.25 million and Lowe $15 million. The three of them cost $34.25 million together.

We only need two and a half of them, so if we could get away with only paying half to one of them our total would be somewhere between $27.8 million and $30.6 million. Less, but not a lot less, than the $35-$39 million we’d have paid for Lee, Blanton and half a year of Moyer.

Cliff Lee of the Mariners
Cliff Lee of the Mariners

Close enough that I’d rather have Lee, even if the overall ERA is going to be similar. Having four elite starters would come in pretty damn handy in the postseason.

I realize this is some whack cipherin’ I’m doing here. I have no idea, really, how adding Cliff Lee to the 2010 Phillies would have affected that staff. And there are all sorts of ways to get both poor pitching and league average pitching a lot cheaper than by paying Blanton, Moyer, Arroyo, Wolf and Lowe to do it — the next two ERAs above Wolf’s, for example, belonged to Jonathan Niese and Randy Wells, neither of whom made $430,000

But it’s fun to think about, and I do think that while I certainly haven’t proved anything, it’s a reasonable assertion that replacing one poor pitcher with an elite pitcher like Cliff Lee has about the same effect on a staff’s overall performance as replacing three poor pitchers with league-average performers. I’d love to see someone who can really make a spreadsheet sing study that.

What I think that person would find: Those league-average guys, they’re pretty good.

* * *
<p><i><b>Photo credits</b>
Phillies: <a href=”http://www.flickr.com/photos/artolog/3776568591/”>Art Siegel</a>
Mariners: <a href=”http://www.flickr.com/photos/macberns/4715005608/>John Berner</a>
Rangers: <a href=”http://www.flickr.com/photos/sidehike/5152160871/”>Bo Nash</a>
Indians: <a href=”http://www.flickr.com/photos/marksobba/1472917568/”>Mark Sobba</a>
All via Flickr/Creative Commons</i>

* * *

Photo credits
Phillies: Art Siegel
Rangers: Bo Nash
Indians: Mark Sobba
Mariners: Bo Nash
All via Flickr/Creative Commons