SABRmetrics- pro or con?

SABR is the Society of American Baseball research (http://www.sabr.org/). SABR is comprised of a wide array of folks who study the history and stats of the sport. SABRmetrics is the term used to describe the mathematical analysis that SABR members have devised for measuring baseball statistics. Bill James, Keith Olbermann, and Bob Costas are all members, to name some of the more famous. Some members are math professors who crunch numbers for a living.

There are extremes here. Some say that baseball is best interpreted and strategized using statistical analysis. Billy Beane (Oakland GM) uses SABRmetrics to help decide which players to sign and trade. Bill James is probably most notably tied to the movement. The development of adjusted batting averages, Pythagorean Win-Loss records, etc., stems from SABRmetrics.

Others, dubbing themselves as purists, believe baseball is not played on a computer, thus any attempt to take away from what goes on between the white lines is folly. Beyond the basic statistics, these folks tend to scoff at park adjustments, W/L record adjustments ("they could've won 100 games, but they actually won 88" would be a common refrain) and the like. A player who hit .250 hit .250, they would say.

So where are you on the spectrum?
Any SABR members here?

This came up on a talk show, so I thought I'd bring it here. I'll wait a bit before I give my take. I know you're waiting with great anticipation. :laugh:

 

What, no takers? This usually generates some discussion either way.

 

I strike a bit of a middle approach. At the end of the day, as the manager in Bull Durham said, you hit the ball, throw the ball, and catch the ball. :laugh: I'm not big on trying to mathematically explain everything. The whole Pyth W/L is suspect to me. Park adjusted averages are too, to a lesser extent. But I find much of what SABRmetrics proffers up to be very interesting. VERY interesting. James has made some good contributions (His HOFM badly needed tweaking, and has been.) All true purists have to love numbers because they do tell a story. But let's be careful not to try to create a story; just use the numbers to tell the story and paint the picture of what happened, not what might've happened if the games were played in a vaccum.

So I appreciate many contributions of SABR and, though not a member, am not ruling out joining. I appreciate a lot of what James proffers and I reject some of his less compelling work. I'm a purist (not in the "let's take the game back a hundred years" ilk - I support the DH after all) who appreciates the good that SABRmetrics offers, while rejecting it all as gospel. The only thing about baseball that's gospel is what goes on between the lines.

 

Has anyone ever read "Covering the Bases: Making Sense of Bill James' Statistical Nonsense" by Walsh and McFall? It's supposed to be a critical evaluation of James's work, and paints him in a bad light. I'm looking to pick it up and read it if I can find some free time.

 

Well, James is certainly not infallable. Remember that people rejected James outright. To reject a review of his work outright is not fair.

 

Well, I don't know the thinking. That's why I want to read it. And even if it is as you posit, it's still worth considering. Maybe they make the case James is an idiot and SABRmetrics is fallacy, and maybe they're right. Or maybe not. I'll find out once I get it and let you know :thumbs: I take it you wouldn't be interested in reading it because it doesn't agree with your bias....er, you don't have time? (Teasing with you, friend)

 

I hear you.