Basketball Prospectus: Unfiltered Everything Else is Fluff.

March 25, 2010

Applying a New Mexico correction

Filed under: Uncategorized — John Gasaway @ 12:26 am

New Mexico won 15 straight games this year, and under the “Did you lose? No? OK,” heuristic applied by polls and conventional wisdom alike, that was good enough for the Lobos to earn a three-seed in the NCAA tournament. At the time, I thought that was generous. Per-possession performance in a conference with perfect round-robin scheduling is the nearest thing we have to the Platonic ideal of accuracy, and on that basis the Lobos were effectively the third-best team in a league without much in the way of recent NCAA tournament success. Nevertheless, Steve Alford‘s team earned a three, whereupon they beat Montana by five in the first round before falling to Washington by 18. 

I’m not saying the Lobos’ early exit was inevitable. Worse teams than the 2010 version of New Mexico have made runs deep into March. Anything can happen, whether it’s the third-best team in the Mountain West going on an NCAA run or Northern Iowa beating Kansas. Anything can happen.

My point is simply that we knew prior to the event that probability was the enemy of the Lobos. And what we owe to the entire field is seeding that is based on the best information we have at the time. Nevertheless, New Mexico was overseeded. An overseed (and, of course, New Mexico’s not the only example) is unfair to the teams in the three brackets where it doesn’t occur. It gives an advantage to the team that gets to play the overseed.

We can avoid this in the future by applying a New Mexico correction in the seeding process.

As vividly demonstrated annually by Wisconsin and the Mountain West Conference, a gaudy efficiency margin is no guarantor of March success. On the other hand a middling per-possession figure in conference play has proven to be a handy mark of doom for a simple reason. The odds always favor doom. All teams, but one, are doomed, eventually. It’s been six years since a major-conference team with a middling per-possession figure made the Final Four. It’s been five years since a major-conference team with a negative efficiency margin made the second weekend.

The other day I remarked half in jest that there should be one committee, using existing means, that selects the NCAA tournament field and another entirely separate committee, using reality, that seeds the field. Actually it doesn’t matter how many committees there are, of course, as long as reality intercedes during the seeding process. Using per-possession performance as a uni-directional correction to bump seeds down but not up would quiet fears that a margin-aware seeding process will encourage teams to run up the score. A New Mexico correction will apply only to middling teams. If you can only beat Air Force at your place by three, we probably don’t need to stay up nights worrying about you intentionally running up the score. 

None of which is to take anything away from this particular New Mexico team, which won the Mountain West regular season title and did, after all, outscore its conference opponents by a healthy 0.11 points per trip. Alford has built a consistent winner in Albuquerque (note that the Lobos were underrated last year), and that’s exactly my point: New Mexico didn’t do anything to find themselves in the position of Easily Foreseeable Overseed poster child. Being in this position is something that simply happens to a team. You win an unusually and unsustainably large number of close games. It happens every year to a small number of teams.

And you know what? As long as doing so doesn’t give you a higher seed than you should have, there’s no harm there whatsoever. I see a day not far off where we merely slap such a team on the back and say, Way to go! Keep it going! instead of What? A three-seed? Are you crazy? 

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress