Monday, January 16, 2017

1/16 S-CURVE

The 1 line:  Villanova (17-1), Kansas (16-1), UCLA (18-1), Kentucky (15-2)
The 2 line:  Baylor (15-1), Louisville (15-3), Gonzaga (17-0), Butler (15-3)
The 3 line:  North Carolina (15-3), Florida St (16-2), Duke (14-4), West Virginia (15-2)
The 4 line:  Virginia (13-3), Creighton (16-1), Notre Dame (16-2), Oregon (15-2)
The 5 line:  Florida (14-3), Xavier (13-4), Arizona (16-2), Cincinnati (15-2)
The 6 line:  St Mary's (15-2), Wisconsin (14-3), Maryland (15-2), PUrdue (14-4)
The 7 line:  Minnesota (15-4), USC (16-3), South Carolina (13-3), TCU (13-3)
The 8 line:  Arkansas (13-4), SMU (15-4), Northwestern (15-4), Miami (12-4)
The 9 line:  Dayton (12-4), Seton Hall (12-5), Iowa St (11-5), VCU (14-4)
The 10 line:  Michigan St (12-7), Pittsburgh (12-6), California (13-5), Middle Tennessee (14-3)
The 11 line:  Kansas St (13-4), Virginia Tech (13-4), Clemson (11-6), Indiana (12-6), Rhode Island (11-6), Nevada (15-3)
The 12 line:  UNC-Wilmington (15-2), Illinois St (13-4), Akron (13-3), Richmond (11-6)
The 13 line:  Monmouth (13-5), Georgia Southern (9-6), New Mexico St (15-2), Vermont (13-5)
The 14 line:  North Dakota St (10-6), Belmont (11-4), UNC-Greensboro (11-5), Green Bay (9-6)
The 15 line:  Princeton (9-6), UNC-Asheville (11-6), FGCU (11-5), Texas Southern (8-9)
The 16 line:  Boston (9-7), UC-Irvine (9-9), New Orleans (6-7), Weber St (7-6), Mount St Mary's (7-12), Morgan St (5-10)

Next 4 in:
Michigan St
Pittsburgh
California
Kansas St

Last 4 in:
Virginia Tech
Clemson
Indiana
Rhode Island

Last 4 out:
Illinois (11-6)
Georgia (10-6)
North Carolina St (12-6)
Houston (13-5)

Next 4 out:
Texas-Arlington (10-5)
Texas Tech (13-4)
Wichita St (14-4)
Ole Miss (10-7)

5 comments:

Anonymous said...

Andrew,

Let me start by saying I've been following your blog for about 2 years now and I've always valued your reasoning and logic. I look forward to your posts daily.

I am on the Bracket Matrix as well and hold one of the computer generated rankings. I am always trying to improve it, but it is a tricky business since it is so difficult to program the parsing of all kinds of different team resume highlights and weaknesses. Yet, humans somehow seem to do it almost effortlessly and somehow all seem to come to nearly the same conclusion.

I was compelled to post today because many respected bracketologists, yourself included, have UCLA as a 1 seed, and I just can't figure it out. There is no metric I track which has them deserving being anywhere near a 1. The only exception is the AP/USA Today polls, which are also a mystery to me. Let's compare to Baylor (which my system has a 1 seed, vs UCLA a 4 seed):

RECORD: Ucla 18-1, Baylor 16-1 wash
RPI: Ucla 20, Baylor 1 BAYLOR
RPI SOS: Ucla 90, Baylor 7 BAYLOR
Top 25 wins: Ucla 1, Baylor 3 BAYLOR
Top 50 Record: Ucla 3-1, Baylor 7-1 BAYLOR
Best win: Ucla at Kentucky, Baylor N Louisville UCLA
2nd best win: Ucla N Nebraska, Baylor vs Oregon BAYLOR
Loss: Ucla at Oregon, Baylor at West Virginia wash
Common Opp: Ucla lost to oregon by 2, Baylor beat oregon by 16 BAYLOR
Computers: Kenpom has Ucla 12, Baylor 8. Sagarin has Ucla 12, Baylor 11. BAYLOR

Can you please explain to me in your eyes what merits UCLA on the top line? I'm interested in the though process. From what I can tell it comes from two things: the win over Kentucky, and the several wins over usually good programs (Michigan, Ohio State, Texas A&M) which looked good at the time but have now lost some luster. Do people need to re-evaluate UCLA's resume? Is owning the best true road win worth 3 seed lines over an otherwise 4 or 5 seed deserving resume?

Thanks

Andrew said...

I'm banking on some committee history on that one. In the past they've bypassed better resumes for conference champions when picking 1 seeds. I'd have to dig, but I'm sure I have a couple rants from the last couple years talking about that posted.

I think it's as simple as Baylor not playing a road game in the non-con, plus having Kansas in the way of a conference title, while UCLA has the one signature road win and a relatively clear path to the Pac-12 title.

The committee has a history of leaning on polls, eye test, and conference championships for 1 seeds and I don't trust them to do the "right" thing and put Baylor there. At least not until more results come through.

Indydoug said...
This comment has been removed by the author.
Indydoug said...

Given Cincinnati's few remaining quality win opportunities (@SMU, Hou., XU) what is their upside if they finish 28-3 and win conference tourney?

Andrew said...

My gut says 2 or 3 seed, maybe an equal chance for either. I can't see a 1 seed, just with Villanova, Kansas/Baylor, Kentucky, plus whoever wins the ACC is going to rack up a boatload of signature wins...not enough room up there.

My rule of thumb for conference champs from non-power conferences (Top 6) is that their seed is almost more dependent on what the other teams around them do than what they do. Cincy's seed in that scenario might depend more on how many ACC teams can hang around on the top couple of lines, if Gonzaga can stay undefeated, if WVU can stay close to KU/BU in the B12, and if the B1G champion can separate from the rest of the conference.