NCAA releases first NET rankings, and we should all just ignore it for now

AP Photo/Keith Srakocic
0 Comments

The NCAA released their NET rankings for the first time all season long, and whoooooooo boy, are these things going to cause a stir.

If you have forgotten, the NET is a metric that the NCAA created in an effort to eliminate the RPI from the NCAA tournament selection process. No one is quite sure exactly how the formula is put together, but what we do know is this: After the first three weeks of the season, it is quite clear that there is not enough data for the NET to properly evaluate these teams in a manner that makes any kind of sense.

Ohio State is No. 1 in the NET, which more or less says all you need to know about where this metric is as of today. Virginia is No. 2. Texas Tech, Michigan and Gonzaga round out the top five. Duke check in at sixth, with Michigan State, Wisconsin and Virginia Tech rounding out the top ten.

And then things get weird.

Loyola Marymount is 10th, one spot ahead of Kansas. Belmont is 12th, one spot ahead of Nevada. San Francisco is 25th, two spots in front of Tennessee. Kentucky is 61st, behind Abilene Christian, Florida Atlantic and Liberty.

It’s even difficult to parse through what the formula actually values. Ohio State being No. 1 makes me believe that road wins carry weight, yet Buffalo — who is undefeated and won at West Virginia and Southern Illinois — is 30th. If it is beating quality competition that matters, how is San Francisco — with wins over UC Davis, Maine, Harvard, Dartmouth, LIU-Brooklyn and a Division II school — 25th? How is Loyola Marymount 10th while Kansas, who has the best resume in college basketball to date, 11th?

The answer appears to be raw efficiency margin:

That’s not ideal, but it is not unexpected, either. When they NCAA released a graphic during the offseason detailing how the formula for NET is put together, it was clear that they were not adjusting for opponent with efficiency margin.

You can read through all the rankings here, but I really would suggest that you did not. This is nothing more than fun with small sample sizes. There’s a reason why the most well-respected metrics in college hoops (KenPom, Bart Torvik, etc.) include data from the previous season. It helps prevent situations like this from arising.

Which means that it is way too early to even think about what this data signifies.

And the truth of it is really this simple: The NCAA probably shouldn’t have released all this information now.

All that is going to happen is that people that don’t fully understand the math behind the data are going to flip out about the kinks in a metric that isn’t close to being ready to unveil.

Oh well.