The hot topic of discussion right now seems to be the announcement yesterday by the Rangers that they will use Tommy Hunter in a game 4 in the ALDS, rather than bringing Cliff Lee back on short rest. That would mean that the Rangers would have Lee available for a potential game 5, if need be, on long rest.
I've gone back and forth on this, about whether the Rangers would be better getting two starts from C.J. Wilson with Lee going on short rest, or whether they'd be better going with Hunter in game 4. This falls in the category of things where there probably isn't a definitive correct answer that we can establish, and I think either choice is defensible.
Based on what I know, I'd probably lean towards going with Hunter in game 4 and Lee in game 5, although I appear to be, among LSB folks, in the minority. So I wanted to walk through the various issues involved, and look at what I think the Rangers are contemplating.
Ultimately, the question boils down to this: Do you think the difference between C.J. Wilson and Tommy Hunter is greater than the difference between Cliff Lee on short rest and Cliff Lee on long rest?
However you slice it, Wilson has been better than Hunter in 2010. That being said, I don't know if the difference is quite as great as folks here believe it is.
Compare Wilson and Hunter in four different measures:
|C.J. Wilson||Tommy Hunter|
Wilson and Hunter each have ERAs that are superior to what their peripherals would suggest -- not surprising, given the defense playing behind them. Interestingly, though, ERA, xFIP, and SIERA all have Wilson and Hunter being separate by about a half-run of ERA, whereas FIP has them separated by almost a run and a half.
So when we are determining how much difference there is between Wilson and Hunter, in terms of expected performance in a playoff game, the FIP figure throws off what would otherwise be a rather simple conclusion -- that Wilson is about a half-run better than Hunter. If we average out the four numbers, we come up with a difference between the two of .75 in earned runs per 9, a figure that I don't think is unreasonable.
So...is Cliff Lee on short rest more than .75 earned runs per 9 worse than Cliff Lee on long rest?
That's a hard question to answer in large part because Lee has never started on short rest, and that uncertainty no doubt is part of the reason the Rangers are leery of starting him on short rest.
It has also been reported that the Rangers took into account playoff performances by pitchers over the last decade, with pitchers on short rest posting a 5.01 ERA, pitchers on regular rest posting a 4.18, and pitchers on long rest posting a 3.80. That spread, if applicable to Lee, would indicate that the Rangers would be better off with the Hunter/Lee combo.
Looking at Lee specifically, his splits indicate he performs better with more rest. His career ERA on regular rest is 4.20, compared to 3.30 with 5 days rest. And the difference this season has been dramatic...Lee has a 4.05 ERA with 4 days rest, 2.34 with 5 days rest, and 1.15 with more than 5 days rest.
Given Lee's track record of having more success with long rest rather than regular rest, and his having never pitched on short rest before, it becomes hard to justify using Lee on short rest instead of Tommy Hunter. The spread between Lee on normal rest and long rest is larger, both for his career and this season, than the spread between Hunter and Wilson. Given that Lee is likely to perform worse with short rest than even on regular rest, and the expected spread becomes even larger.
There are some other factors that I think probably play into this to a lesser degree. The Rangers probably feel that Lee, who has World Series experience, is better suited than Wilson, who has never pitched in the playoffs, to handle a game 5 on the road, particularly if that game 5 is in Yankee Stadium. If New York is the opponent, Wilson's struggles against the Yankees this year is something the team may be concerned about. And Lee may be seen as someone who would have more success on the road than in TBIA, given his history.
But regardless, history shows that pitchers pitching on short rest in the playoffs lose almost a full run off of their ERA, and Lee's history has shown that he's almost a run better on long rest than on regular rest.
In light of that, and in light of what Wilson and Hunter have done this season, I have a hard time being confident that Lee would be good enough working off of short rest to make the cost involved, in terms of his likely decrease in performance, worth the trade-off.