In 17 days the NCAA tournament selection committee will announce the brackets for this year's 65-team field. The criteria that the committee will use to select and seed the field are numerous, but in large measure they come down to three elements: Wins, losses, and opponents (albeit with a due sensitivity for game-location underpinning all of the above). In other words, the game is the thing--where it happened, which team you played, and whether you won or lost.
Of course if you've been reading along with us for a while here at Prospectus, you know that in addition to looking at games we also like to consider what can be learned from individual possessions. Mind you, these aren't antithetical approaches, or at least they're not to me. After all, games are comprised of possessions. For example right now Kansas is perhaps the best team in the country on a per-possession basis and, lo and behold, the Jayhawks look just as good on a per-game basis (27-1).
It's just that any game has much more to tell us than simply who had more points when time ran out, and at Prospectus we're more than happy to listen. That willingness gives us better and more detailed information than what the selection committee chooses to employ. In fact, I like to think of each possession as something to be "won" or "lost," based on the difference between how many points you scored and allowed, on average, on each trip down the floor (a quantity otherwise known as efficiency margin).
So why doesn't the selection committee avail itself of this "better and more detailed information" when selecting and seeding the field? In theory the committee does indeed consider a very wide range of information, even up to and including an efficiency-margin-based rating system like Ken Pomeroy's. But in practice the standing objection to a proper acknowledgment of efficiency margin's value can be boiled down to one word. Sportsmanship. It is feared that if the selection committee starts explicitly giving teams credit for beating the heck out of their opponents, then really good teams will start running up the score intentionally and wantonly.
I can't help feeling that this fear is overblown. First off, I'm not advocating a selection process that simply parrots Ken's rankings or Jeff Sagarin's Predictor or my listings of efficiency margins step by mindless step. No one would advocate that, just as no one would suggest that such ratings should banned from the committee room outright. But in between these two extremes there's a sweet spot where efficiency margin is along for the evaluative ride without driving the bus.
I actually think a margin-aware bracket would look surprisingly similar to what we get now. Kansas and Kentucky would still be Kansas and Kentucky, automatic qualifiers would still populate the bottom three lines in each regional, etc. What I envision is a process that merely refines the information that's already being used, one that's informed by a thousand or so possessions instead of just a few dozen games.
Second, even if the perfect worst-case bad-sport situation arises, one where an unsportsmanlike coach finds himself with a really good team that can dominate opponents, I think said coach would find that intentionally running up the score in conference play is very hard to do, much harder than it is in college football. (I say in conference play. I take it as axiomatic that there's little if anything to be gained from measuring how badly a given major-conference team can beat opponents that they themselves chose to schedule.) Even the most talented and least sportsmanlike teams will miss their shots between a third of and half the time against even the least talented opponents. Yes, you can always shoot that last shot as time expires instead of merely holding on to the ball. That's not going to send your efficiency margin soaring skyward. Yes, you can always leave your starters in the game even when the outcome's been decided. Then you'll simply get yelled at by your fans for tiring out your stars.
Third, even if that perfect worst-case situation arises, the occasions when that evil coach will be able to walk on the dark side will be few and far between. I write this having just watched Kansas play a home game against Oklahoma. If not the best team in the nation, the Jayhawks are certainly one of the two or three best by any reasonable measure. And while the Sooners aren't Nebraska-bad, they are reliably and decidedly weak on the road and, anyway, they played this particular game without Willie Warren. Yet KU won this game by just 13 points.
Lastly, and perhaps most importantly, running up the score is self-defeating. Past a certain lopsided point the undeniably impressive victory takes on the quality of a mere fluke. Or, if it happens against a struggling opponent, it wasn't that impressive to begin with. Here's what I said earlier this year when Marshall beat East Carolina by 51 points. (And, mind you, I'm not implying in any way that Marshall ran up the score. I trust they did no such thing. But this is the kind of result people fear would result from a more margin-aware tournament selection procedure.)
Marshall attained their gaudy efficiency margin in large part by beating East Carolina to a pulp this weekend. Remove that single 51-point blowout from their resume and their in-conference number drops to a much more descriptive +0.04. Save your praise for UTEP.
Perhaps we could make qualitative corrections like that under a new tournament selection procedure as well.
In sum, I doubt that the factors that guard against running up the score have really been as artificial and tenuous as a selection committee's procedures. I would instead hazard to guess that they've been as fundamental and durable as the essential nature of the game itself and the sportsmanship norms that impose themselves as a matter of course on any member of the coaching fraternity.
The stated ideal of the NCAA tournament selection committee is to choose the best teams. Being an ideal, it is of course unattainable, and we'd do well to remember that every Selection Sunday when the committee's handiwork is yet again being torn limb from limb. Nevertheless, making this selection with so little consideration for something that we know from experience matters a great deal can't help but keep us at arm's length from that ideal. We didn't always have this efficiency margin stuff, but we do now. It should be added to the mix.
True, we could well decide that a tournament populated by at-large teams with the highest number of "good" wins and the lowest number of "bad" losses is the way to go. If so, then the status quo or something close to it is perfectly acceptable. But if we really want a tournament with the best teams, then it's time for a more thoroughly and avowedly margin-aware selection process.
John's all about good sportsmanship on Twitter: @JohnGasaway. College Basketball Prospectus 2009-10 is now available on Amazon.
John Gasaway is an author of Basketball Prospectus.
You can contact John by clicking here or click here to see John's other articles.