Sunday, January 15, 2017

W-8 BP68

There are always a few weeks every season where it's hard to find 68 teams to fill in the bracket with. For whatever reason it's a bad week for bubble teams, and the bubble just seems awfully soft. This is one of those weeks. There is little stability down in the 11-12 seed range, and expect that area of the bracket to continue to churn.

In the end, two projected at-large bids changed this week. Northwestern and VCU came in while Ohio St and Oklahoma St dropped out. There were also two auto-bid changes, as UC-Irvine is the new Big West favorite (replacing Long Beach St) and LIU is the new NEC favorite (replacing Wagner).

This is also the first week of the "Full Bubble", which I start every year on the first bracket on or after January 15th. The Full Bubble is an exercise in mathematically eliminating teams from at-large contention with each bracket. I start with a zillion of teams, and then each bracket will narrow it down until we just have the Selection Sunday bubble. This year, the Full Bubble starts with 82 teams, and by Selection Sunday we'll be down close to 10.

Capitalized teams are projected automatic bid winners. Remember that this is a projection of what the bracket will look like on Selection Sunday, and not a measure of where teams would be if the season ended now:

1. KANSAS (BIG 12)

2. North Carolina
2. Virginia
2. Baylor

3. West Virginia
3. Louisville

4. Purdue
4. Butler
4. Creighton

5. Oregon
5. Florida
5. Xavier

6. Florida State
6. Notre Dame
6. Saint Mary's
6. Iowa State

7. SMU
7. South Carolina
7. Indiana
7. Clemson

8. Virginia Tech
8. USC

9. Michigan State
9. TCU
9. Maryland
9. Miami - Florida

10. Kansas State
10. Michigan
10. Houston
10. Minnesota

11. Texas Tech
11. Arkansas
11. California
11. VCU

12. Rhode Island
12. Northwestern




16. LIU (NEC)

Teams seriously considered that just missed the cut:
Wake Forest, Marquette, Seton Hall, Oklahoma St, Illinois St, Utah

Decent resumes, but not good enough:
UCF, North Carolina St, Pittsburgh, Syracuse, Illinois, Nebraska, Ohio St, Georgia, Tennessee, Texas A&M, BYU

Long shots, but still in the at-large discussion:
Memphis, Temple, Davidson, La Salle, St. Bonaventure, Georgetown, Providence, Iowa, Penn St, Oklahoma, Texas, Charleston, Northeastern, Valparaiso, Boise St, San Diego St, Arizona St, Colorado, Stanford, Alabama, Auburn, Mississippi, Mississippi St, Vanderbilt, Chattanooga

Still alive, but pretty much need a miracle:
UConn, Tulsa, Boston College, Georgia Tech, George Mason, George Washington, UMass, Richmond, St. Joseph's, DePaul, St. John's, Rutgers, William & Mary, Louisiana Tech, Marshall, Green Bay, Harvard, Canisius, Iona, Eastern Michigan, Ohio, Toledo, Loyola-Chicago, Missouri St, Southern Illinois, Colorado St, Fresno St, New Mexico, Wyoming, Tennessee St, Oregon St, Washington, Washington St, LSU, Missouri, North Dakota St, Arkansas St, Georgia Southern, Louisiana-Lafayette, San Francisco,


Anonymous said...

Cute you still think Duke will win the ACC when they're projected to finish 6th in latest T-Rank projections:

Jeff said...

As the bracket states clearly, the capitalized teams are the projected auto-bid winners. Duke is still the most talented roster in the country if they ever get the full team on the court and healthy at the same time.

Anonymous said...

Yeah, and right now they're projected to finish 6th. They're sitting at 2-3 having played the 2nd easiest schedule in the conference so far and have the third hardest schedule remaining in the conference. No reasonable projection would make them the favorite to win the conference at this point. Seeding has more to do with resume than it does with potential. They are very talented, but it's debatable whether they're more talented than Kentucky. Even though the committee loves Duke and does anything they can most years to gift them an easy road to the Final Four to boost ratings (see 2010 where they didn't deserve the 3rd number 1 seed over Syracuse by any reasonable metric), they'll more than likely have too many blemishes for the committee to gift them another 1 seed this year.

Jeff said...

Conspiracy theories aside, Duke hasn't had their full roster healthy and in place in a single game all season long. So while I don't question their struggles, I am simply arguing that if they get their whole roster together that they have the most talent for March.

Anonymous said...

For someone who touts themselves as a college basketball expert you sure are pretty clueless. Seeding decisions are predicated almost entirely on a team's resume (who you beat, where you beat them, etc.), not their future potential. Even if Duke was the most talented team, would they deserve a #1 seed if multiple star players missed a significant portion of the season, resulting in them finishing barely above 0.500 in the ACC and not winning the ACC tournament, but they were suddenly healthy and ready to return for the first game of the tournament? The answer is a resounding no in case you can't figure that out on your own.

Do you know how many times the committee has selected the pre-tournament top 4 KenPom teams (presumably the best four teams going forward, not factoring in resume as you seem to think that doesn't matter) as the 4 number 1 seeds? One time in 15 years in 2002. Teams with better resumes are routinely seeded higher than teams rated above them in KenPom who presumably have a better true talent level. Here's a list in case you think I'm making this up:

2003: Oklahoma and Texas get #1 seeds despite being 7th and 8th in PTKP.
2004: Stanford gets a #1 seed despite being 14th in PTKP. Teams ahead of Stanford in pretourny KenPom were seeded as low as 6 (Wisconsin, UNC)
2005: Washington gets a #1 seed despite being 15th in PTKP. Teams ahead of Washington in PTKP were seeded as low as 5 (MSU and Nova)
2006: Memphis gets a #1 seed despite being 11th in PTKP. Kansas and Illinois were both rated higher in PTKP but were seeded as 4's.
2011: Texas was rated 4th in PTKP but received a 4 seed as they were only 6-7 in tournament quality games.
2012: Syracuse gets a #1 seed despite being 7th in PTKP. Three #2 seeds were rated higher than them.
2013: Kansas gets a #1 seed despite being 7th in PTKP. 3 teams rated above them received lower seeds.
2014: Louisville gets a #4 seed despite being second in PTKP since they were only 6-4 in tourny quality games and had a soft schedule compared to other top teams.
2015: Duke gets a #1 seed despite being only 7th in PTKP. Three teams rated above them received 2 seeds.
2016: Oregon gets a #1 seed despite being only 9th in PTKP. Five teams rated above them received worse seeds.

So yeah, Duke could very well be the most talented team when healthy, but they don't deserve a #1 seed if they finish in the 4-6 range in the ACC tournament just because they were preseason #1 and people think they're the best team when at full strength.

PTKP = Pretournament KenPom

Anonymous said...

And it's not really a conspiracy theory when Syracuse goes 7-1 versus the KenPom top 25 and Duke goes 1-3 and manages to get the 3rd number one seed over Syracuse by some magic. Even if they were the better team, their resume didn't justify being given that cake walk of a bracket. Kansas was the overall number one that year and somehow ended up with the toughest region based on the aggregate KenPom ratings of the top 4 seeds in each region.

Jeff said...

"Record vs KenPom Top 25" is not a stat that the Selection Committee uses. Nor is KenPom a rating that has any significant impact on brackets at all, as I detail each year when I break down how the various computer metrics predicted the bracket.

Anonymous said...

Out of 77 brackets on you're the only person projecting Duke as a #1 seed. Duke's average seed is 4.3. They could certainly improve on that a bit, but jumping all the way to the #1 seed line is highly unlikely, even if they do finish in the top 4 in the pretournament KenPom ratings.

Anonymous said...

It's been documented (bigbluehistory quote) that the selection committee has been using Sagarin ratings, which correlate extremely well with the KenPom ratings, since 1984. So even before KenPom existed they were already using a rating that is very similar. They just don't like to publicize it much because they're afraid teams will run up the score more to try to inflate their power ratings. But Sagarin and KenPom most certainly use a diminishing returns principle for blowouts so that's not a major issue. Also, a NYT article also has the committee chairman quoted as saying they use KenPom. So as usual you don't know what you're talking about. - The most telling were the revelations that since 1984, the committee has requested Sagarin's ratings for their deliberations. Beyond that, according to Sagarin a committee member once confided with him that the committee does indeed use and appreciates his rankings. This certainly is understandable since the Sagarin ratings are well-respected and are specifically designed to reflect team strength, exactly what is the most useful measure for determining the best team among prospective at-large squads.

According to Sagarin's encounter, the committee member added that although the committee like his ratings, "we can't say that" The reason being that Sagarin's ratings take into margin of victory. Wolff explained earlier in his article that the NCAA is adverse to anything which seems to reward margin victory, as it encourages coaches to run up the score and it smacks of the type of point spreads which are often associated with gambling, an influence which has been and remains to be a threat to college basketball and sports in general. - “The common metrics most of us use are KenPom, Sagarin, L.R.M.C., B.P.I., K.P.I.,” the committee chairman, Oklahoma Athletic Director Joe Castiglione, told reporters Wednesday, sounding like an economist rattling off the names of various federal departments.

Jeff said...

I'm shocked to learn that a Kentucky fan website is convinced that the Selection Committee is engaged in a conspiracy to inflate Duke's seed each year.

Anonymous said...

You really are fucking retarded. Duke wasn't mentioned a single time on the bigblue page I referenced. It was a reference from a Sports Illustrated article from 2003.

Jeff said...

You seem like a delight.