Monday, February 8, 2010

Is Northwestern an NCAA tournament team?

Did you know? Kansas St. has had six coaches since 1986, and all six are
still active in Division 1 today.

Coach – Current School – Years at Kansas St.
Lon Kruger – UNLV – 1986-1990
Dana Altman – Creighton – 1990-1994
Tom Asbury – Pepperdine – 1994-2000
Jim Wooldridge - UC Riverside – 2000-2006
Bob Huggins – West Virginia – 2006-2007
Frank Martin – Kansas St. – 2007 - Present

More coaching posts in the future, but in the meantime, I’m going to steal a piece of reader email from John Gasaway today. John was trying to be kind to his reader so he didn’t say all this, but I would have been a lot harsher, as you’ll see below.

“With nine regular season games remaining, NU still has some work to do
to get off the bubble. Meaning the Wildcats’ RPI is currently 61. Win
eight of those nine and things should be great: 23-8 overall and 11-7
in-conference. Sure, that’s not likely, but if the Cats pull off the improbable
it should really boost their RPI, right? No. In such a scenario
Northwestern’s RPI will actually fall to 67 under this inane “scheduling is
destiny” system. And if NU miraculously wins all nine remaining games
(which, again, won’t actually happen) their RPI will stay about the same. These
results are predestined by the fact that the Wildcats will play Iowa, Indiana,
and Penn State twice, and Chicago State once. But I defy any of your readers to
say that Northwestern is a better team now than they would be if they went 8-1
the rest of the way.

I hope the valuable missionary work that you, Ken Pomeroy, and others
have done over the years has appropriately diminished the status of the RPI
in the eyes of the selection committee. If Northwestern is excluded from
the NCAA tournament based on their Sagarin or Pomeroy ratings, I can accept
that. And if they’re excluded because of too few good wins or too many bad
losses, I can accept that. But if Northwestern is excluded on the basis of
RPI, I will be angry and will redouble my efforts to see that a better system is
used to select tournament teams.

Ben J.”

First, this reader is pointing out something many people have noticed and which I emphasized in December. Schedule is such a large factor in the RPI that games can have a “predestined” positive or negative effect. And in Northwestern’s case this is true in the extreme. No matter what Northwestern does, their RPI is going to go down.

But I have great news for Ben. A team’s RPI is NOT a criteria used to select the teams. The RPI is used to group opponents faced during the season and that’s it. But don’t just take my word for it. If you want to read about the process, the NCAA has been trying to make it more and more transparent every year. Here are some articles I linked to a few years ago.

Maybe the NCAA should just drop the RPI because it causes too much confusion and angst among viewers. The NCAA should say teams will be grouped based on season performance and leave it at that. Because the truth is, this isn’t the BCS. There is no magic formula for NCAA tournament selection. At the end of the day a bunch of people sit in a room for a week and compare team resumes. As Jay Bilas puts it, the selection process is very simple: “Who have you beaten and where?” That’s the criteria.

So to this Northwestern fan, I have good news. Your team won’t miss the dance because of the RPI. But I also have bad news. Your team will likely miss the dance because they don’t have enough quality wins. Northwestern is currently 1-5 vs the RPI top 50 and 2-2 vs the RPI top 100. Or to drop the RPI nomenclature, Northwestern basically has three quality wins this year. They beat Purdue, Illinois, and Notre Dame and that’s it.

That’s not a profile that will earn an NCAA tournament bid. And the six remaining games against inferior competition don’t help. Those are no-win games. If they win, they are supposed to win. Truthfully, one or even two losses to the bad teams wouldn’t matter. What Northwestern really needs is some more quality wins. They need to beat Minnesota, and they desperately need to win at Wisconsin to get a second Top 50 win. Otherwise a deep Big Ten Tournament run is the only hope.

Northwestern’s record is going to improve over the next month and that will help. Performance in the last 12 is a key criteria. But Northwestern’s opportunity to improve their resume is basically over. And no less a prognosticating novice than ESPN’s Jay Williams was spouting this on Midnight Madness. Illinois is in much better position than Northwestern even though the finishing stretch is brutal. That’s because Illinois can still prove they are an NCAA tournament team. The Pomeroy and Sagarin ratings don’t look favorable, but Illinois still has multiple chances to get more quality wins. Meanwhile, Northwestern had most of its chances and failed.

Now, this reader wants to credit John with being a missionary for conference efficiency margins over the RPI. And I think we all agree, that John and Ken Pomeroy provide a tremendous service for evaluating the quality of teams and understanding how they play. But efficiency margins are NOT a criteria for NCAA tournament selection either.

The NCAA has decided in almost every sport that margin-of-victory should not be a criteria because one of the NCAA’s primary goals is to promote good sportsmanship. And the NCAA wants to lower the incentive of teams to run-up-the-score. The NCAA is willing to do a worse job picking NCAA tournament teams in order to encourage good sportsmanship.

And to some degree I agree that margin-of-victory is a poor NCAA selection criteria. Think about this:
Minnesota was a net –8 in two games against Michigan St.
Illinois was a net –5 in two against Michigan St. and in one of those games Kalin Lucas was out.
On paper, you can argue Minnesota is about as good as Illinois. But there’s a key difference.
Illinois went 1-1.
Minnesota went 0-2.

You play to win the games. I don’t want to see some 6-10 ACC team get into the NCAA tournament because it played a lot of good teams tough and the ACC had great margin-of-victory numbers in the non-conference.

But remember that the committee is made up of humans. And humans can’t help but notice how teams played. So if you think John Gasaway’s conference efficiency margin, Ken Pomeroy’s rankings, and the Jeff Sagarin’s predictor ratings aren’t important to the committee, you are dead wrong. They look at all this information and take it into consideration.

But what I would argue is that the committee has a responsibility to use these numbers, not for picking the 34 at-large teams, but for seeding the tournament. If you ignore team performance and seed outstanding teams too low, it punishes the other teams in their bracket.

If you visit the bracket project you’ll see Maryland is estimated to be a 9 seed. (This is because there aren’t many RPI top 50 wins to be had in the ACC this year.) But based on Pomeroy’s predictive measure, Maryland is the 8th based team in the nation. Can you imagine being a 1 seed and having to play the 8th best team in the nation in the second round? That just wouldn’t be fair.

Now the NCAA walks a fine line in seeding this way. Quality wins remain the main criteria. But if you think when any group of humans is comparing resumes that a 1 point win looks the same as a 20 point win, good luck winning that argument.