Saturday, October 4, 2008

Wave of the Future Hits Geology, Baseball

One of my new geoscience friends was recently explaining to me a paradigm shift in the way people think about paleoclimate, from the way "The Old Guard" does science to the way "The New Guard" is doing it. Broadly speaking, The Old Guard sees a qualitative trend in some paleo data, and uses scientific intuition, developed over years spent in the field, to make some sweeping claims about the mechanism causing the observed effect, and that's that. But according to the New Guard, that's not enough anymore-- mechanistic hypotheses must be rigorously and quantitatively validated, or at least constrained. Climate modeling must be used to test one's claim to see to what degree it is consistent with the evidence in the data.

I couldn't help but automatically compare this New and Old Guard of geoscience to the evolution of statistical thinking in baseball (as cultivated by Billy Beane, and explained in Michael Lewis' book Moneyball). The old-school managers and scouts look for players with qualities like "a great body" and "good face" who embody their vision, from their years of experience, of a great ball player. Managers who espouse the new, sabermetrically-inclined view discard these traditional values. Instead, they do studies of the statistics of ball players to see which ones actually translate into wins for the team. Then, if they can find a fat player (and thus undervalued by the old guard) with a high OBP (which would manufacture more runs), then they can get this guy on the cheap to make their team stronger!

Now, this ain't a deep comparison by any means... I'm just excited to see people in all fields improving their field with some serious mathematics! Math is the wave of the future!