« No Parking | Main | The Travelogue Continues »

April 30, 2004

The Power of Faith

A wise man once told me that there are two kinds of people in the world: those that divide everyone into two types, and those that don't. I'm in the former camp, evidently, as I think there are two kinds of people: those who believe what they believe regardless of the evidence, and those who allow their beliefs to be effected by evidence. ESPN's Buster Olney clearly falls in the second group.

One of the big controversies in baseball over the past few decades has been over sabrmetrics, or the study of baseball statistics. Begun by men like Bill James, sabrmetrics attempts to look at the sport and try to answer questions like, does the bunt pay off, or is stealing bases very valuable? That may not sound like much, but until James and his comrades began applying systemic methods to the questions, there were a phenomonal number of misapprehensions about the sport. In fact, there still are today: just look at how much attention is still paid to batting average over on base percentage. On base percentage has a pretty strong correlation with winning, while batting average has very poor correlation to winning. Still, if you were to ask who were more valuable, a guy hitting .330 with a .360 on base percentage and a guy hitting .250 with a .400 on base percentage, a large number of fans (and people who actually make their living in the sport) would take the first guy.

But general managers like Billy Beane are starting to change that. Beane is the most well-known of the new breed, thanks to the book Moneyball. Moneyball chronicles how Beane has kept Oakland competitive for the past five years despite being under some pretty tight fiscal constraints. Beane has done it by throwing out what 'everyone knows' and using what has actually been proven in its stead. While this has led to success, it has also led to backlash against Beane and those like him, because he's threatening beloved shibboleths. So every once in a while we see an article like Olney's latest.

Olney tries to extol the glories of a new statistic called productive out percentage, a stat that is supposed to tell us which teams are using their outs productively and which are wasting them. Unsurprisingly, Olney zeroes in on Oakland as one of the bad teams that doesn't use outs well, and tells us that this is why Oakland hasn't done well in the playoffs despite such impressive regular season records. Unfortunately, Olney's examples don't even stand up to a cursory inspection.

Unfortunately, I couldn't find the complete Elias listing of POP, so I just used what I could find in the article. I popped the numbers into a spreadsheet along with the winning percentage for the teams in question as of yesterday. Then I told Excel to correlate the numbers. As those familiar with statistics know, a correlation close to one indicates a high degree or correlation, while a correlation number close to zero indicates little or no correlation between the two numbers. The POP result: 0.07, or basically no correlation whatever.

Of course, Olney's own article should have told him that much, since he noted that the Red Sox (currently holding the best record in the majors) and the Yankees (often considered to have done well over the past few years) have very poor POP numbers relative to the rest of the league, while the best POP teams are Detroit, Pittsburgh, and Arizona, two of whom are off to terrible starts and the third of which lost 119 games last year.

That doesn't matter to Olney, though, because he knows that playing baseball the old fashioned way is the way to win. It doesn't matter what facts we bring up to note that maybe the old ways don't work so well, because Olney just knows it the way so many people know that God does or doesn't exist: it's a question of faith.

I suppose it's an easier way to live than actually testing your beliefs. But it does tend to mean you end up constructing more and more elaborate frameworks to support your clearly inaccurate beliefs.