Statistics beat the Committee


You have heard me complain countless number of times about the committee.  There are plenty of places where you can look at the seeds and totally get that they are off.   But one of the things that frustrated me is that it felt that many of the teams that played won Sunday did not get a fair assessment which included their championship game.

I could give tons of reasons why Kentucky belonged ahead of Texas A&M – when you have two teams that are co-champs, they both finish with the same record, they both were able to beat some good teams out of conference, and they both play in the conference championship, the team that wins that game should get ranked better.  I am sure some arguments can be made.  Before Sunday’s game, I had Texas A&M slightly ahead.  But after Kentucky won on a neutral court, I felt you have to move them up.

And everyone knows that I think Michigan State and Purdue were unfairly treated – but I am biased towards the Big 10 and it doesn’t help that both of them lost their games, so I will leave them out.

But there was one other team that was punished for what appears to be simply playing on Sunday afternoon.   One mid-major conference played their conference championship on Sunday – the Southland conference.  And the team that won that game was Stephen F Austin, who finished the season with a victory over Texas A&M – Corpus Christi to go 27-5.  The Lumberjacks were rewarded with a 14 seed.   And so I thought it would be interesting to look at how the statistical ratings like KenPom, Sagarin, ESPN BPI, and Adjusted Scoring Margin would rank the teams compared to the rankings.

Team Committee RPI Pomeroy BPI Sagarin ASM
Chattanooga 1 5 10 7 10 9
Ark – Little Rock 2 2 3 2 4 5
Yale 3 3 2 1 1 2
South Dakota St 4 1 7 6 5 6
UNC Wilmington 5 4 6 4 6 7
Hawaii 6 11 4 3 3 3
Stony Brook 7 6 8 8 7 4
Iona 8 9 5 9 8 8
Green Bay 9 14 13 13 13 16
Buffalo 10 12 16 14 16 15
Fresno St 11 8 11 10 9 14
Stephen F Austin 12 7 1 5 2 1
Middle Tennessee 13 10 14 16 12 11
CSU Bakersfield 14 15 9 12 11 10
UNC Asheville 15 16 12 11 15 12
Weber St 16 13 15 15 14 13
Austin Peay 17 19 18 19 18 20
Hampton 18 17 20 20 19 19
Florida Gulf Coast 19 21 17 17 17 17
Fairleigh Dickinson 20 20 21 21 21 21
Southern 21 18 19 18 20 18
Holy Cross 22 22 22 22 22 22

You will notice that in all of the statistical ratings, 4 of the top 5 teams in each of those rankings were Yale, Little Rock, Hawaii and Stephen F Austin – 4 of the 5 teams that pulled upsets from the mid-major conference champions.    That’s right, both Pomeroy and Sagarin would have had the 4 12 seeds as the 4 teams that pulled upsets, and BPI would have only missed on UNC-Wilmington, the team that was beating Duke at halftime.  Adjusted Scoring Margin sometimes is off because of who teams play – but even it would have ranked those 4 teams as the top 5, switching out Little Rock for Stony Brook.

So, how could Stephen F Austin, who was either the best or second best mid-major according to two of the more respected computer rankings of college basketball, be ranked as a 14 seed instead of a 12 seed (and not just a 14 seed – the worst ranked 14 seed).

Their RPI was a little higher than the 12 seeds, but if RPI matters that much, why weren’t teams like Valparaiso, St. Mary’s and St. Bonaventure in the tournament over teams that were 30 RPI rankings below them.  By the way, Valparaiso just beat St. Mary’s tonight to make it to Madison Square Garden and the NIT semi-finals – where their run included all double-digit victories including one against the ACC’s Florida State.   Now, I admit that the Seminoles aren’t Duke or Miami – but if Valparaiso could beat FSU by double digits, I have to imagine they could have held their own in the tourney.

But back to the Lumberjacks, since even RPI would have given them a 13 seed.  I think the reason is the committee didn’t want to come up with multiple brackets.  And I think that if Texas A&M – Corpus Christi had upset SFA to win the Southland’s bid, their profile would have been a 14 seed.  So, my gut opinion is that they filled the bracket with a placeholder as a 14 seed for the Southland winner.  It is the most reasonable explanation to me of how the Lumberjacks could have a better record than most of the mid-majors, have better statistical rankings and still be ranked so far behind multiple teams.

And then the committee’s best 3 seed in West Virginia ended up getting this team that clearly should have been a couple seeds better.  And while the 3 seed has to win that game, regardless, it is not a coincidence that this tournament saw a record number of upsets.  I guess I should be happy – it has led to a really fun tournament.  I like seeing so many double digit seeds succeed.   But it doesn’t stop me from thinking that we might be losing some really good teams earlier than necessary in the tournament because the tournament wasn’t seeded appropriately.  OK – I will try to make that my final rant about the committee.  We have 16 incredible teams remaining – and 1 of them is going to deserve to be the NCAA National Champion – bring on the Sweet 16!!!!


Leave a Reply

Your email address will not be published. Required fields are marked *