You don't want to pitch CC Sabathia against Cliff Lee. So the Yankees found out in the first game of the World Series, which they lost, 6-1 (trailing 2-1 on Sabathia's watch).
This is the reason why the strategy of pitching Sabathia (and others) on three days' rest appears to be working. This, in spite of the fact that starters on three days' rest are 12-36 in playoff games since 1999, according to Fox News.
But after the Yankees' win in game three, Sabathia pitched game four against the Phillies' fourth starter, Joe Blanton. The Yankees won (even if Sabathia didn't). Pitching on short rest in game five, A.J. Burnett was just a sacrificial lamb against the fully-rested Lee. The Yankees won two out of three. In Philadelphia.
It was PHILADELPHIA that should have gambled on another Sabathia-Lee matchup, after the loss in the third game. Lee gave up only two runs in seven, yesterday (three of his runs were in the top of the eighth, meaning that he would probably have been relieved earlier in a close game).
Match those two runs, and assume that the bullpen gives up two more (instead of the one it actually surrendered), for a total of four against Sabathia et.al. in game four, and the Phillies might have won it 5-4, even against Sabathia.
Then pitch Blanton against short-rested Burnett in game five, and transpose the seven runs allowed in game four into game five, and the Phillies can win 8-7. In that case, they'd be 3-2, and the odds-on favorite, even going back to Yankees Stadium.
Instead, it is the Yankees who are 3-2, going home, where they have to win only ONE game. And they have CC Sabathia available for game seven if they lose game six. Given his short game five tenure, even Burnett might be available as a "long reliever" by then.
As Yogi Berra supposedly said, "it isn't over until it's over." It's barely possible that the Phillies will win the last two in enemy territory and thus, the World Series, after dropping two out of three at home.