Colorado Belongs in the NIT: Who Was Really Snubbed by the Selection Committee

Berry Tramel of The Oklahoman writes an interesting article about scheduling.  In that article, he addresses a lesson to be learned from Colorado’s alleged “snub” from the 2011 NCAA Tournament — schedule better.

Go back to the criticism leveled towards the Boise State and TCU football programs.  The crowded that hated those two teams united behind one critique — the Broncos and Horned Frogs did not play anyone!  They argued that both teams have one really tough game and should not be rewarded for winning one tough game.

Now, I have laid it out here before that I do not buy into that argument; good teams are good teams regardless of their conference.  But, if you want to make that argument in football, then why is it not applied to basketball.  Colorado beat Kansas State three times and has a win over Texas.  Nice.  But what about the rest of the schedule.  What about games against Longwood and Western New Mexico?  The latter of those two is a Division II basketball team!!!!!  I know the Mustangs are great and all, but what good does it do to play a D-II school, especially one that went 14-16 this year.  That alone should eliminate a team from even being in the discussion for an at-large bid.

[NOTE: I know Chaminade is a D-II school, but they host a tournament that attracts quality opponents.  That case is different].

Additionally, the Buffaloes played Cal State Bakersfield (9-19), Maryland-Eastern Shore (9-22), Texas-Pan American (6-25), The Citadel (10-22), Idaho State (9-20), and Alcorn State (4-24)!!  Add in Longwood (12-19) and those seven teams are a combined 59-151!!!  Add to that losses to San Francisco (17-14), Oklahoma (14-18), and lowly Iowa State (16-16) and there is a reason Colorado is playing in the NIT.  I am not even knocking them for their scheduled games against Indiana, Oregon State and Colorado State.  But when you have a non-conference schedule that bad, you do not belong in the NCAA tournament; you belong in the NIT!

Virginia Tech is another team that people try to argue should be in the tournament.  Sure, games against Kansas State (loss) and Purdue (loss) are good schedules.  But, Campbell (12-19), UNC-Greensboro (7-24), CSU-Northridge (14-18), St. Bonaventure (16-14), South Carolina Upstate (5-25), and Mount St. Mary’s (11-21) is just a bit too much on the “sure win” column.  The Hokies also played Longwood.  Those six “sure wins” combined to go 77-140 on the season!!!

Additionally, VT’s only signature win was against Duke and maybe Penn State.  And while they had no bad losses like Colorado, the Hokies were swept by Virginia, a team that went 16-15, lost to Seattle (an 11-20 independent), and only win against an NCAA tourney team was a two-point victory over Clemson (a marginal NCAA tourney team at best).  Although not as pathetic as Colorado’s schedule, Virginia Tech’s overall non-conference schedule — coupled with only one signature win — keeps them from “dancing.”

Yeah, I know.  Colorado head coach Tad Boyle tried to make the argument that they went on the road to Georgia (loss), San Francisco (loss), Harvard (loss), Indiana (win) and CSU-Bakersfield (win) and that is why they deserved to win.  Seth Greenberg complained that he beefed up his schedule and that is why they deserved to get into the tourney.  Problem is, Greenberg did not win any of those games; for Boyle 2-3 against those road games — 0-3 versus the “good” teams; 2-0 versus the bad ones.

So, let’s say that Colorado and Virginia Tech made the field; which teams do they replace?  UAB?  Virginia Commonwealth?  How about Southern California and Marquette?

You see, there is no clear determinant as to why UAB or VCU should be removed and Colorado and Virginia Tech should be added except for one thing — the former two are from mid-tier conferences while the latter two are from the so-called “Big Six.”  Strength of schedule?  That is a bogus measure.  First off, Virginia Tech is only slightly ahead of UAB (77 for VT versus 78 for the Blazers); VCU is 84…Colorado is better at 49.  But if you are going to target a team for its weak strength of schedule, how about Cincinnati (83), Florida State (84), George Mason (91), or Temple (92).

Strength of schedule is not a good specific measure.  Sure, it is a decent general measure, but think about this — the top overall seed Ohio State has a SOS of 23.  The top SOS is Georgetown, which is a sixth seed.  The team with the 14th SOS — Seton Hall — is not even in the tournament.  Ergo, strength of schedule is a poor measure.

Polls are also a poor measure because those do not rank every team.  The RPI is the only measure that ranks every team on a seemingly objective basis.  I have argued the use of the RPI as a way to fill in the at-large bids before.  So, let’s use that and see where Colorado and Virginia Tech fall.

Out of the 68 seeds, 31 are guaranteed to winners of specific conferences; the remaining 37 are up for anyone.  Of the 31 conference champions, it is likely that some will fall beyond an RPI ranking of 68 — this year 15 teams had an RPI higher than 68, ranging from 75th (Long Island) to 322nd (UNC Asheville).

Now, of the 37 at-large teams, all 37 have an RPI of less than 68; the two lowest are Marquette (64) and Clemson (57).  Thus, if Colorado and Virginia Tech were to replace anyone, would it not be those two teams instead of UAB or VCU?

Ah, but there is more.  Let’s say that after the 31 automatic qualifies that the remaining 37 will be given a bid based on their RPI.  In other words, the remainder of the field will be made up of teams with the highest RPIs amongst the non-champions.  Through this method, teams like BYU (RPI 5), North Carolina (6) and Pittsburgh (10) will no doubt be in the field.

But where is the cut off?  Because 15 automatic qualifiers fall outside the RPI Top 68, only the top 53 in the RPI would get in.  However, that numbers is altered a bit because Gonzaga (56th) and Oakland (53rd) also made it in as AQs.  Belmont is the 51st rated team in the RPI, therefore the at-large pool would only go to number 50.

Based off of that criteria, the following snubbed teams would get into the NCAA tournament:

  • Harvard (RPI 35; seeded 9th)
  • Cleveland State (RPI 42; seeded 11th)
  • Missouri State (RPI 43; seeded 11th)
  • St. Mary’s (RPI 46; seeded 12th)
  • Colorado State (RPI 50; seeded 12th)

The five teams that would be knocked out based off of this criteria:

  • Michigan (RPI 52; NCAA seed 8th)
  • Florida State (RPI 55; NCAA seed 10th)
  • Clemson (RPI 57; NCAA seed 12th)
  • Marquette (RPI 64; NCAA seed 11th)
  • Southern California (RPI 67; NCAA seed 11th)

Now, the seedings for the five that would get in under this format is based on ranking the teams 1-68 in order of RPI.  Therefore, Kansas would be the number one overall seed and UNC-Asheville would be the 68th seed (so to speak).  In terms of the First Four games, it would set up as follows:

  • 16th seed – UNC-Asheville v. UC Santa Barbara
  • 16th seed – Alabama State v. Arkansas-Little Rock
  • 12th seed – Colorado State v. Virginia Commonwealth
  • 12th seed – Illinois v. Georgia

At this point it should be obvious who is still not in — Colorado (RPI 65) and Virginia Tech (RPI 63).  And, notice who still is in the tourney — VCU (49) and UAB (31).  In fact, based on the criteria presented, the Blazers would be an eight seed.

Furthermore, not only would the Buffaloes and Hokies still be left out under the RPI criteria, but there would still be nine teams ahead of both of those (eleven ahead of Colorado — Marquette and Virginia Tech).  Of those nine, six are other snubbed teams — Marshall, Boston College, UTEP, Wichita State, Oklahoma State, and Southern Miss — and three are teams that got in — Michigan, Florida State and Clemson.  So, why does Colorado or Virginia Tech deserve to be in over any of these teams!?

The point is that by an objective measure, Virginia Tech and Colorado do not belong in the NCAA tournament.  Both teams are exactly where they need to be — the NIT.  Beyond that, both teams have to do better than Longwood.

Besides, let’s be real — how many teams have a legitimate shot at winning the tournament?  Five?  Maybe ten?  So the other 58 teams are just filler.  Is that what Colorado and Virginia Tech really want to be — filler?  It is like a 6-6 team making it to a bowl game.  Sure you went to a bowl game, but will anyone really remember it beyond the universities involved?  Same here.  Would anyone really remember that Colorado went as a 12 seed and maybe had Sweet Sixteen run, at best?  Exactly.

Enjoy the National Invitational Tournament!

Advertisements

4 thoughts on “Colorado Belongs in the NIT: Who Was Really Snubbed by the Selection Committee

  1. Hey, found your article through a similar tag as mine and really enjoyed it, it had some great stuff! My only argument with this, while I think you are right about who really got snubbed, is that the RPI is still a poor measure of a team’s abilities. The RPI is extremely overrated and a little outdated. The formula involves 25% of the team’s win percentage, 50% of their opponent’s win percentage, and 25% of their opponent’s opponent’s win percentage. With this as the formula, it’s easy to jack up your RPI by playing poor teams that play great teams. For example, Miami(OH), who was not good this year, played at Duke and at Kansas to name a few. By scheduling the Redhawks, you boost your RPI. So to use your idea, it would be better to use something like the Sagarin computer rankings, which will probably still prove your point about Va. Tech and Colorado. But I very much agree with your overall point and thought you had some great stuff. Well done!

  2. i agree with you — the RPI cannot be taken alone. but even the Sagarin has issues — Southern California has the worst Sagarin rating among at-large teams, yet are discussed the least. under a Sagarin-based model, UAB would be removed, to be sure, but so would the Trojans. VCU would remain. VT and CU would get in under that measure, but there were other teams that got in with lower ratings (Georgia; Tennessee).

    additionally, there still exist a seeding mismatch based off of the Sagarin — SDSU should be a one; Utah State a six; Florida a five.

    nevertheless, i appreciate your comment and the read. certainly the RPI has its flaws; i will admit to that.

  3. I am not a fan of computerizing the process to be honest. I just think within the selection process the Sagarin is a more effective measure than the RPI. I think the human element makes College Basketball and the entire tournament that much better. While it is completely subjective, I think the eye-test has a place in the process too.

    As for VT and CU, even though their schedules were pretty weak, they actually beat someone. Colorado beat K-State three times. Despite a poor SOS that should carry a lot of weight. And Va. Tech beat Duke when the Blue Devils were 1 in the country. Again, even though the schedule wasn’t tough they had some good wins. VCU and UAB really didn’t beat anyone (VCU beat George Mason, I believe).

    But regardless of who got snubbed and who didn’t, this just proves that 64 teams, or even 65, was the right number and 68 is just a waste. The only good thing that came out of this was that the opening round games are still meaningless. With the exception of maybe Clemson, none of those teams should move on past the next round.

  4. actually, i believe 64 is too many; i’d rather see it drop back down to 48. if Colorado were to be in it, would they really have a legitimate shot at winning it? my answer is no. that is why i related the “just making the tournament” mentality to celebrating a 6-6 football team making it to a bowl game.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s