Sunday, November 25, 2007

Self-Scheduling Day

At my high school, it was a privilege. In major conference college basketball, it's a right-- maybe even a weapon.

Stanford 73, Colorado State 53

One of the things that I find continually irritating about college sports is the number of noncompetitive games that fans have to wade through in order to get to the gems. Let's face it-- we really don't need a game to tell us that Stanford is better than Colorado State, or that Washington State is in a slightly higher class than Mississippi Valley State.

This is, I will freely admit, more of an issue in football than basketball. In football, eligibility for bowl games (and the program-sustaining revenues thereof) is explicitly linked to number of victories, with no reference paid to strength of schedule. And thanks to the way the polls work (basically, the only motivating factor in the polls is losses-- once you're in the top 10, even if you beat the #1 team, you're unlikely to move upward unless someone else loses) there's basically a loss quota for the BCS (2 losses) and the national title game (1 loss). Your strength of schedule is theoretically factored into some of the various computer formulas which comprise 1/3 of the BCS ranking system, and even more theoretically factored into the minds of the various cranks and coaching subalterns who comprise the poll respondants, but in practice it's almost always better to not lose than to win a big game (but risk losing it).

So we see a ton of crap like Michigan State scheduling Appalachian State. Granted, it's wonderful that App State was able to pull off a huge upset. But the excitement of that game was created by the non-excitement of all the various blowouts of lousy 1-AA teams over the last 10 years or so, in exactly the same manner as the individually stupid decisions by millions of people to buy Super Lotto tickets eventually create a huge jackpot.

Basketball is a different kettle of fish. In part, this is because teams are selected for the championship tournament by a group of people who mostly know what they're doing (unlike the college polls, which are just as screwy and stupid as the football polls), and in part it's because teams play enough games that at some point, you kind of have to play someone good or no one will ever notice you exist. And best of all, basketball is a lot more prone to upsets than football, at least in the context of a single game. You'll get a number of Siena-over-Stanford results in any given week. Unfortunately, it's still not immune to the cupcake phenomenon.

This year, we see a lot of Pac-10 teams playing extremely soft non-conference schedules. There's a really simple explanation for this: everyone expects the conference to be brutally difficult. The best teams in-league are likely to limp out with 4 or 5 losses, which is already around the lower bound for a #1 seed. Most teams will not come out of conference play with more than 10 wins. To even reach the 20 wins which are traditionally a benchmark for reaching the Tourney, you need to win 10 out of 11 or 12 non-conference games (13 for lucky duckies Washington and UCLA).

This naturally leads to a lot of cupcakes scheduled, which is a frank shame. Tonight's Arizona-Kansas game was excellent. I respect Lute Olsen for choosing not to dilute his team's schedule with too many Northern Colorados and UC Davises. But I think it hurts his team. Oftentimes good Arizona teams come out of the season with 18 or 19 wins and get poor seeds in the NCAA tournament despite playing a tougher schedule than any other major conference team. The committee is only human. They can't help but be affected by the number in the Wins column to a greater degree than any other number.

As long as you have a certain quota of "tough games" and potential "statement wins," a quota which has basically been filled by the Pac-10 conference this season simply by virtue of existing, there's not much advantage to scheduling more of them.

I've got a bunch more to say on scheduling, but the rest of it has to do with the RPI, which will probably take up another one of these "vaguely topical essay" posts. There's just not much to be said about the actual game here-- Stanford played well, Anthony Goods broke out of his shooting slump to post another solid game, the team took care of the ball well, and the game wasn't on TV.

Next up... more of the same, as the Cardinal takes on bottom-feeder Sacramento State Tuesday in an untelevised game.

1 comment:

Scott Bergquist said...

Paul, I think a "point count" would be a good system for both college basketball and college football.

Here's one off the top of my head:

Out of a possible ten points, you get ten if you beat a high-ranking team, soundly. The losing team gets five, for getting beat so resoundingly. Very close score, "buzzer-beater", you get eight, the losing team, seven. If you schedule an "over-matched" school and beat them resoundingly, you get seven points. "Buzzer-beater" score, you get six, loser gets six. If you are upset by an "over-matched" opponent, you get three, the over-matched school gets ten.

You always get some points for playing a game, but, cumulatively, your strength of schedule would reward you more than pure "winning". It would be a no-brainer to rank teams for the NCAA tournament.

And the "begged" question is, of course, how do teams get a "status" as "formidable" versus "wheelchair team"?? Ah, even better! Some system could be arrived at, of that I am certain. But the "status" of very good versus very bad could also be in flux, further qualifying a team's season. For instance, once Oregon lost their (Heisman candidate) QB, beating them should have lost some of possible points, because the "status" of Oregon had changed. If a team has an All-American-type "big man" and is ripping through teams, only to have him out with a season-ending injury, beating that team at the end of the year should get you less points than the lowly team that upset the team, "big man" and all, earlier in the year.