Which one looks better to you?
|CC Sabathia||CC Sabathia|
|Phil Hughes||Phil Hughes|
|Andy Pettitte||A.J. Burnett|
|A.J. Burnett||Ivan Nova|
|Ivan Nova||Sergio Mitre|
There is no questioning that the one on the left is the preferable option. The only unknown is of how big a difference exists between Pettitte and Mitre. An initial reaction might have the difference pegged at a few wins, but I’m not sure it’s that wide a gap. In fact, given each pitcher’s limitations I think we’re talking about a single win, maybe even less.
As Mike did yesterday, I won’t count on Pettitte for a full season’s worth of starts. In the same way, I won’t count on Mitre for that, either, since the last time he started more than 10 games was in 2007. I also assumed 6.1 IP per start for Pettitte and 5.2 per start for Mitre. That leaves Pettitte with 126.2 IP and Mitre with 113.1. That’s just 13.1 additional bullpen innings, which we’ll have to factor in somehow.
The tougher part of this exercise is projecting ERAs. Bill James forecasts Pettitte at 3.86 and Mitre at 4.57, but I think both of those are a bit aggressive. But let’s keep them in the bank, just in case. As a rough estimate of ERA, I’d peg Pettitte at 4.00 and Mitre at his career ERA, 5.27. Let’s see how the differences work out.
If we go with the James projections, Pettitte works out to 54 earned runs, or 2.7 ER/GS. Mitre works out to 58 ER, or 2.9 ER/GS. In 20 starts that amounts to a whopping four runs. Even if we go with the more conservative 4.00 and 5.27 estimates, we get Pettitte at 56 ER, or 2.8 ER/GS, and Mitre at 66 ER, or 3.3 ER/GS. That’s a 10-run difference — or roughly a single win. Now that we’ve put it in the simplest possible terms, it doesn’t seem like that big a difference, does it? We can adjust up or down, but I don’t think you’ll get an exceedingly different answer unless you think Mitre will produce something like a 7 ERA. I don’t think that particularly likely.
Where we actually get the biggest difference is with the bullpen. Those are just 13.1 innings, but they’re 13.1 innings that are already accounted for with Pettitte. This obviously can fluctuate wildly. If we have those innings filled by 2010 Chan Ho Park, that’s another 8 ER. If they’re thrown by David Robertson it’s 6 ER; with Boone Logan it’s 4 ER; with Mo it’s 3. I’d say 5 ER is a decent compromise.
That brings our difference between Mitre and Pettitte — accounting for earned runs and innings pitched — to somewhere between 9 and 15 ER over 20 starts, or between .45 and .75 runs per start. That’s going to cost the Yankees a couple of those 20 games. But the key term is a couple. It’s hard to argue that the difference would amount to much more than that.
Bringing back Petitte will clearly make the Yankees rotation better. There is no reliable measure that can say otherwise. But given both Pettitte’s and Mitre’s limitations, the difference might not be as great as we imagine. In the AL East two wins will matter plenty. But the difference between Pettitte and Mitre is not the difference between a .500 team and a 92-win team. Unfortunately, the small difference that does exist could play a large role in determining the 2011 postseason.