Moose vs. Palmer

Over on Futility Infielder, our pal Jay Jaffe, responding to a piece by Rob Neyer (another friend) makes a compelling case that Mike Mussina has been a more valuable player than Jim Palmer. Jay acknowledges Palmer's superiority in traditional stats (and awards), but notes that Mussina is superior in more advanced metrics like WARP, VORP, and SNLVAR. He writes "Moose's strikeout rates relative to the league (a translated rate of 6.9 EqSO/9) made him much less reliant on his defense than Palmer (5.2 EqSO/9), so he's rewarded by getting a larger share of the credit for each run saved on his watch, and thus generated more value." It's a fascinating discussion, and one that illustrates a constant challenge for baseball thinkers, that being, how to square our growing statistical abilities with historical knowledge, especially when these two seem to conflict. Old school thinkers have a sad tendency to just dismiss the new math, but that's not really a legitimate option, and one of the nice things about this argument is that both Jay and Rob are fundamentally sound thinkers, who know their sabermetrics and their history. So I raise this issue: Jim Palmer threw 211 complete games. Mike Mussina 57. I understand that some of the stats Jay raisesmeasure relative value, but I wonder if it is even possible for a numerical model to account for the change in the very idea of what it means to be a pitcher? A real head scratcher. Bottom line, you'd do well to have Jim Palmer or Mike Mussina or both pitching for your Dream Team.