March Madness Bracketology: February 18th Edition

With nine days remaining until the start of Conference Championships for NCAA Basketball, we take another look into the current state of March Madness. Recall that to get into the March Madness Tournament, a team must either win their conference championship tournament, finish first in the Ivy League, or obtain one of 32 “At-Large” bids from the NCAA Selection Committee.

Our method of selecting teams focuses on an aggregation of models after finding a covariance structure and dependencies on previous years rankings and the results. For reference, we have managed to correctly select 65 of 68 teams for the previous three consecutive years. Comparatively, we have outperformed CBS Sports (they typically are 62-64 correct) and have tied ESPN every year.

That said, here’s our current projections for conference champions:

  • America East: Vermont (24-5, 14-0)
  • American: Cincinnati (24-3, 13-1)
  • Atlantic 10: Virginia Commonwealth (22-5, 12-2)
  • Atlantic Coast: Louisville (22-5, 10-3)
  • Atlantic Sun: Florida Gulf Coast (21-7, 10-2)
  • Big 12: Kansas (24-3, 12-2)
  • Big East: Villanova (26-2, 13-2)
  • Big Sky: Weber State (16-8, 11-2)
  • Big South: UNC – Asheville (21-7, 13-2)
  • Big Ten: Maryland (22-4, 10-3)
  • Big West: UC Irvine (15-12, 8-3)
  • Colonial Athletic: UNC-Wilmington (23-5, 12-3)
  • Conference USA: Middle Tennessee (23-4, 13-1)
  • Horizon League: Valparaiso (21-6, 11-3)
  • Ivy League: Princeton (16-6, 9-0)
  • Metro Atlantic: Monmouth (23-5, 15-2)
  • Missouri Valley: Wichita State (25-4, 15-1)
  • Mountain West: Nevada (20-6, 9-4)
  • Northeast: Mount St Mary’s (14-14, 12-3)
  • Ohio Valley: Belmont (19-5, 13-1)
  • Pac 12: Arizona (24-3, 13-1)
  • Patriot League: Bucknell (20-8, 12-3)
  • Southeastern: Kentucky (21-5, 11-2)
  • Southern: East Tennessee State (21-6, 11-3)
  • Southland: Sam Houston State (18-9, 9-5)
  • SWAC: Texas Southern (14-11, 10-2)
  • Summit League: North Dakota State (17-9, 9-4)
  • Sun Belt: UT – Arlington (19-6, 9-3)
  • West Coast: Gonzaga (27-0, 15-0)
  • Western Athletic: New Mexico State (22-4, 8-2)

Here, we have a few upsets predicted in the conference championships, but nothing major out of the ordinary. Remember that predicted winners are probabilistic winners. This means that even though UC Irvine is predicted to win, it’s at a 53 head-to-head percent probability of defeating UC Davis (at 47 percent), the current conference leader.

Last Eight In:

With this year being a little crazy with top teams falling to weaker teams on a somewhat concurrent rate, we find there are more than four teams that may miss entry into the tournament. Instead of a “Final Four In” we have to put our “Final Eight In.” These teams are on current trajectory to making the tournament but have a significant chance of finding themselves in the NIT.

  1. Texas Christian (17-9, 6-7 Big 12)
  2. Michigan (17-9, 7-6 Big Ten)
  3. Michigan State (16-10, 8-5 Big Ten)
  4. Seton Hall (16-10, 6-8 Big East)
  5. Iowa State (16-9, 8-5 Big 12)
  6. Wake Forest (15-12, 6-9 ACC)
  7. Mississippi (16-10, 7-6 SEC)
  8. Clemson (14-12, 4-10 ACC)

Wake Forest and Clemson just don’t make sense, but numbers wise they are on the boundary. Wake Forest is 9-3 out of conference with ugly loss after ugly loss in ACC play. How Syracuse is out while Wake Forest is in gives us a knee-jerk reaction; that is, until we see that Syracuse doesn’t have a non-conference winning record falling to teams like UConn and St. Johns. Clemson follows a similar ugly path as Wake Forest, but owns a horrific 4-10 conference play record. Hanging their hat on a 10-2 non-conference record including a win over a ranked South Carolina team, and Clemson effectively has the same strength as Wake Forest. So take this with a grain of salt. Personal bias suggests Wake Forest and Clemson will be out, unless quality wins come their way in the next three weeks.

First Eight Out:

Looking at the first eight out, we have:

  • Kansas State (16-10, 5-8 Big 12)
  • Houston (18-7, 9-4 American)
  • Georgia (15-11, 6-7 SEC)
  • Auburn (16-10, 5-8 SEC)
  • Syracuse (16-11, 8-6 ACC)
  • Texas Tech (17-9, 5-8 Big 12)
  • Providence (16-11, 6-8 Big East)
  • Georgetown (14-12, 5-8 Big East)

With exception of Syracuse and Georgia, you can see why these teams are out while the above teams are in. These are hair splitting teams. A couple notes can be made from these lists. Marquette and Alabama are notoriously off both lists. This doesn’t mean those teams are out of contention.

In fact, Marquette is currently out and has two quality wins: Creighton and Villanova. But their losses to St. Johns, Michigan, Providence, Seton Hall, Georgetown, and Pittsburgh weight them down. Going into their final five games of the season, Marquette will need to win four and reach the conference semi-finals if they want to crack the bubble. That would mean wins over St. Johns and Providence with either a sweep of Xavier; or a split and a win over Creighton. This would push Marquette into third / fourth in conference with a showdown with Xavier / Seton Hall in the first round of the Big East Tournament. A win there places them against Villanova or Butler in the semi-finals. That will gain them a spot in the tournament.

Alabama suffers a similar situation as Marquette but lack in the quality win department (sans South Carolina). With a terrible losing record: Texas, Dayton, Valparaiso, Auburn, Arkansas, Auburn, you have the reason why Alabama is sitting deep in the bubble. Unlike Marquette, Alabama doesn’t have the schedule to build themselves back up. Alabama will complete their season against Georgia, Texas A&M, Mississippi, and Tennessee. They will have to sweep if they want to sniff the NCAA Tournament. A sweep will push them into fourth in the conference, where they will need a key upset; either Florida or Kentucky, to keep their hopes alive.

Current Projections

Now onto the bracket…

mmbracketfeb18

Current projections of the NCAA Tournament.

Thoughts?

Now that you have seen the current projections, we should note that our models keep effectively the Top 25 teams as the Coaches and AP Polls. The lone exception is West Virginia, who is much deeper in our list (28th) while they are both Top 10 teams in the polls. West Virginia has impressive wins; but also sits on five very confusing losses for being a top 10 team: Temple, Texas Tech, Oklahoma, Kansas State, Oklahoma State. None of them are ranked. One team has even lost two-thirds of their games. Our models had Xavier in the Top 25; ranked 27th and 26th by the Press and Coaches polls, respectively.

So what do you think? Feel free to argue or support anywhere you see fit! As the season rolls on, we will update the models and continue to identify the most-likely NCAA Tournament teams!

Also, if you’d like to see a conference break down, feel free to submit in the comments below! 

Advertisements

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