The Problem of Prognostication

Some time back, I offered the right to pick a post topic to anyone who managed to name one of the Physics Nobel laureates for 2006. Tom Renbarger won, and picked his topic:

OK, with Midnight Madness on the horizon, I've decided to request a sort of season preview of two (trying to press my advantage since I got two names) of the following three conferences: A-10, plus one of the Big East or ACC. Or, if you get on a roll, all three. If you're pressed for time, the A-10 would suffice, and maybe something about Maryland. :-)

I'm going to try to do all three leagues (though I know basically nothing about the Atlantic 10), and have the first post (on the ACC) scheduled already. I don't normally do pre-season predictions for college basketball, though, and I thought I'd do an introductory post to explain why.

The basic reason is simple: college basketball is the hardest sport to predict of any of the major sports in this country. (Explanation below the fold.)

First of all, basketball in general is a tough game to predict. Of the major team sports (football, basketball, baseball), it's the one most susceptible to individual domination, just because there are so few guys on the court at any given time. One really great player can carry a team in basketball in a way that just doesn't happen in football or baseball.

College basketball also has more turnover than any other major sport. In the professional ranks, other than the occasional financially motivated fire sale (see the Florida Marlins), teams stay fairly constant from one year to the next. You might pick up a few new guys, but the backbone of a given team usually stays more or less the same for several years at a time.

In college basketball, on the other hand, the very best you can hope for is to keep a player around for four years. And the best players almost never stick around that long. The most talented and greediest players usually leave early for the NBA, at higher rates than even in college football (where players have to stick around for three years before heading to the NFL). It's not unheard of for a team to start four guys who weren't with the team the previous year.

Finally, recruiting is a total crapshoot. You're getting new players in all the time, and you know less about what you're getting than in any other sport. Professional teams acquire rookies from college, true, but most of the major changes to any given roster are usually other pros, who have an established track record in the league. You get the occasional player who really blossoms or totally collapses in a new location, but for the most part, you know what you're getting when your team picks up a new player.

In college basketball, the newcomers were mostly playing high school the previous year, and probably not playing against competition that was anywhere near as good as what they'll face in college. The collegiate ranks are full of 30-point-a-game scorers from high school who struggle to reach double figures in college. Between made-for-tv high school games, big-name camps, and the ever-expanding AAU system, players are better scouted today than they were ten years ago, but you still can't really be sure that the highly-regarded recruit your team is picking up will be worth a damn.

And even beyond that, there are always a few guys who slips past the scouting system. Obviously, I'm partial to guys who played for my teams, but look at players like Juan Dixon and Lonny Baxter at Maryland, who weren't supposed to be much of anything, and wound up anchoring two Final Four teams and winning a national title. Or Joe Smith, who was an afterthought in Maryland's 1993 recruiting class, and wound up the national player of the year as a sophomore. And that's not even counting guys like Tim Duncan, who come from oddball places that the scouts don't look-- Duncan started out as a swimmer in the Virgin Islands, and has ended up one of the all-time greats. Nobody saw that coming.

Jim Boeheim at Syracuse has made a Hall of Fame career out of finding guys like that. The loss of Pearl Washington was supposed to be crippling, but Sherman Douglas turned out to be even better than Washington. They got slapped with probation in the early 90's, and were supposed to have nothing, and Lawrence Moten came out of nowhere to be a great player. Even a guy like Gerry McNamara was supposed to be a role player, not a key contributor on two Big East championships and a national title. Good coaches get the top recruits to come play for them, but great coaches find the guys that the recruiting gurus missed, and buid teams around them.

So, predicting the results of a college basketball season before it starts is a fool's game. There are just too many variables to do a sensible job of it. I'm usually ok with making predictions about conference play after watching the non-conference season, but that's about as far as I'm willing to go.

I'll write up my thoughts on the upcoming season for the ACC, Big East, and Atlantic 10 in the coming days, but I'm not going to try to predict the results in detail. I'll give some general comments, based on the games I saw last year (which weren't all that many-- it was a bad hoops-watching year for me), and the pre-season magazine (Athlon Sports's College Basketball preview) that I bought to get the recruiting reports. You'd have to be out of your mind to use these predictions as the basis for any kind of wagering, but if I actually get anything right, I reserve the right to gloat about it come March...