Thursday, February 11, 2021

Another administrative note

 As you may have guessed, my plans have had a modest change to them this season.  A sequence of real-life emergencies (yes, plural) have more or less taken me away from all free time (which includes even following the sport on a day-to-day basis).  As it turns out, family health and employment end up being more important than college basketball, although it's debatable.  I thought I could fake my way through November and then pick up steam in the blog in December; real life decided naaaaaaah.

I've more or less had to abandon sports consumption, and only now can start playing catch-up.  Literally as I look right now, Kentucky is 5-13?  The hell happened there?  Duke is under .500?  Why is there a toothpaste ranked #13 in NET?  This is the level of knowledge I am currently operating at.

It's fully my intention to publish the annual bracket prediction on Selection Sunday (a personal streak dating back to 1997 and a blog streak of 2014).  But I'm going to be coming from a very sideways angle this time around.  This is going to have to be much more haphazard.  It might make an interesting social project, to see if intense knowledge of the sport is required for Bracketology.

I dunno if there'll be daily recaps.  Although it'd be hilarious to go back to November, with February eyes.  Might be some intrinsic value there.  I'm probably going to try and start with conference deep dives, and compare the eyes of a person who has just the numbers with the eyes of people who've seen the season play out.

God willing, I can carve out some time this weekend to do this.


HenryMuto said...

Welcome back. As a Kentucky fan I rather not talk about it !!!!!!!!

As an Ohio State fan giddy up !

Bryan Wilson said...

Selection Committee top 16 reveal is tomorrow. I will post my predictions tomorrow morning.

Bryan Wilson said...

OK Here's my guesses:

1: Gonzaga, Baylor, Michigan, Ohio State
2: Illinois, Alabama, Villanova, Houston
3: Virginia, West Virginia, Iowa, Missouri
4: Tennessee, Wisconsin, USC, Texas,
5: Texas Tech, Oklahoma, Colorado, Kansas

To summarize my hot takes:

- Top 2 lines I think are basically set. I almost got Virginia in over Houston but couldn't manage it.
- Missouri is tough. They have fantastic wins and no bad losses but metrics are really bad. I am guessing the wins will be rewarded so 3 seed.
- USC in the top 4. Good computer numbers, 3-1 in G1, 5-1 road record.
- No Texas Tech in top 4. Lots of cupcakes leading to so-so SOS, lots of losses, not great computer metrics.

My computer system has:

1: Gonzaga, Baylor, Michigan, Ohio State,
2: Illinois, Virginia, Alabama, Houston,
3: Villanova, West Virginia, Iowa, Wisconsin,
4: Southern California, Florida State, Kansas, Tennessee,
5: Missouri, Colorado, Creighton, Oklahoma,

It does project a bit based on team power rating so some of my comments above about computer numbers are exaggerated.

HenryMuto said...

I think Ohio State gets a #1 seed today based on all those huge wins and quad 1 wins but their schedule is so brutal having to play all these teams they beat already again so a few losses will be left and as many as 3-4 more losses. So I doubt they can hang onto it.

Indiana today should win but never know but after today 5 straight games they could lose all of them.

@ Penn St
@ Michigan St

Good lord.

HenryMuto said...

You missed Oklahoma and Texas Tech is seems you undervalued the Big 12.

HenryMuto said...

Loyola Chicago very impressive today

HenryMuto said...

Not sure who CBR is on bracket matrix but he had Ohio State as a 3 seed lol

Bryan Wilson said...

LOL after yesterday's games Colgate goes UP 2 more spots to #11 in NET

HenryMuto said...

I keep waiting for Colgate to lose to get out of the top 30

HenryMuto said...

Such a big difference in NET and RPI for some teams. Take Cleveland State for example.

CSU is 163 in the NET and 61 in RPI

Wright State who is tied with CSU for 1st and most consider best in the Horizon.
They are 59 in NET and 81 in RPI

This makes a huge difference in seeding for the auto bid winners.

CSU would be a 15/16 seed this year using NET and if they still used RPI would be a 13 seed range. RPI is 61 !

Wright State crushes CSU in NET but is 20 spots behind in RPI.

I think the NET weights EFF too much and doesn't account for winning and losing enough.

Andrew said...

I was thinking about a stand-alone post talking about NET.

Forget basketball, just think about the numerical system that NET is. In order for it to work, you need a lot of feedback. You need a lot of non-conference results to have a reliable estimate of how good conferences are. Because once you're in conference play, all the results have a circular feedback problem.

Colgate being that high up the board is the perfect example, because the Patriot League is basically a closed feedback loop this year.

NET is going to be a mess this year.