Wednesday, March 18, 2009

On "Bracketology"

Technically, I guess the term "bracketology" is accurate, as "-ology" means "study of a branch of knowledge." "Bracketologists" do study the bracket, its history, its patterns, and its rules and regs (more accurately, its principles and precedures). But, in our culture, it also connotes "science" or a an academic discipline, and this is certainly not what "bracketology" is or does. Major media types build up their in-house bracketologists to be some sort of mathematics guru with access to formulas and complex equations that mere mortals cannot comprehend. It's just not true. Truth be told, "bracketology" is more art than science. It's more like picking the right job candidate from resumes/interviews than solving algebraic equations or using the scientific method to discover which teams are best.

Even the RPI is basic math. No one has to "solve" anything. It's a simple formula: win % x .25 [road wins weighted value is 1.4, home wins are 0.6], opponents' win % x .50, opponents' opponents' win % x .25. Add those three values. That's it. Beyond that, there is no math to be done. And the RPI is available in umpteen (technical math term) different forms online and is updated by the minute, or at least every day. Bracketologists do not even have to do the math that is involved.

When Joe Fan asks me, "How do you get the teams right?" my answer is this: 1) I've been doing this awhile, 2) I read the selection critieria, and 3) I've done the committee process five times. I understand what the committee aims to do. But, scientific it's not. Individual bracketologists are much more about trying to guess what a group of ten humans will do and there are always unknowns and quirks. Each committee differs a little in where it places its emphasis. Some years, finishing strong seems to matter more than in other years. This year, committee chair Mike Slive's magic phrase was "full body of work." This helps explain how Arizona got in (good early wins). It also explains how they were picked over Creighton and San Diego State, who were red hot teams before losing in their conference tourneys. Every committee is different. Knowing this also allows me to stay calm on Selection Sunday when the committee does not do what I think it "should."

If we must stick with the term "bracketology," then we must at least understand that this "science" is more like a TV meteorologist or economics columnist than chemist or mathemetician. Meteorology studies the "rules" of weather, and can get decent results predicting it, but it's far from exact. There is too much variation in natural weather patterns. Economics can use past and current trends and history to forecast the rise and fall of economic markets, but it is nowhere near exact in its prognostications. "Bracketology" is like that. We can study history and patterns of past committees and there are some constants, but projections and predictions are attempting to speak for a long, complex process performed by ten humans. I do think it's valuable because it helps me keep up with all conferences and the more I participate in the process, the better I understand it.

But, I have never mistaken myself for a scientist. Bracketology is a lot more about learning the process than it is "getting the bracket right." That's a fun game, but it's not the primary reason for doing it. The primary reason should be to better understand how our national championship bracket is birthed.