POLL: Should the Selection Committee's Discretion Have Limits?

[visitingcock]

When we first learned of the NET, we said "Oh Boy! Objectivity in choosing/seeding teams for Tournament.

Then, we learned that the selection committee has the discretion to depart from those rankings. Uh, okay I guess that's needed.

Then, we learned about all the things that they could consider in departing from those rankings. Uh, wait a minute. That looks like any subjective belief can be justified. A UConn can have an objective NET rank of 2 but be dropped to the tenth selection? A Southern Cal can be ranked number 10 but be selected number 2 (I'm being told) ? This is different from coaches/AP/ESPN/Fox ratings in what way?

Should a limit be placed on the committees "discretion".

If you think it's fine as is with seemingly unlimited discretion vested in the committee, chose selection 1.

If you think there should be a numeric limitation on haw many places a team can be moved up or down by the committee, chose the number you think is correct

 

[LETTERL]

The whole NET thing is stupid. Teams who consistently blow out opponents and have large leads for long periods of time benefit in the NET; teams that play close games, and win close games, don't fare as well, and sometimes get "punished", in the NET. It needs to be tweaked, or eliminated and another metric developed in its place.

It really shouldn't matter if a team wins by 1 point every game; the important thing is the victories...not the margin. But with the NET, winning by 1 or two points consistently hurts those teams in the NET ratings.

I know there are other things making up the NET but as the ACC season played out, I noticed NC State would often slip, even after winning, if the game was close and the team they were playing was much lower in the NET. Then, last week, during the ACCT, our NET actually dropped one place after we had beaten a Duke team that was #20. Then, we moved up 3 places after a comfortable win over much-lower rated FSU.

It makes no sense. We won the frigging Duke game and that should have counted as a quality win and an upward movement in the NET. But it didn't.

 

[Geew1992]

Or an LSU team the has 1 good win on there resume while beating all the D3 schools be a #2 seed????

 

[Plebe]

It makes no sense. We won the frigging Duke game and that should have counted as a quality win and an upward movement in the NET. But it didn't.

But NC State was already ranked higher than Duke. Why would a team be expected to move up in the NET for barely beating a lower-ranked team on a neutral court? It's not even supposed to work that way.

 

[Vowelguy]

Once we get a foolproof ranking system then the committee should have limited discretion.

 

[DefenseBB]

Ok, this thread started off half-cocked and has gone completely kaboom.

The NET is simply a tool that replaced RPI as one of a set of criteria the committee considers. The NET shows how well a team did (predictive) against it's opponents.

Let's break it down to a basic example: SC is #1 in NET which means their calculated score in the NET algorithm (black box) was the highest of all 356 D1 teams. The NET calculated their algorithm score of their 32 (32-0) games. The score UConn earned against their schedule ranked #2 in NET. Note that SoS can and is a component of the NET.

This does not mean UConn is #2 in the country. It just means the algorithm (a tool) calculated their predicted results against 33 (28-5) opponents had the second highest number.

Texas is #3 in NET. The algorithm calculated their 33 (29-4) opponent results into a score that was better than 354 D1 schools.

@LETTERL mentions a concern that the committee addressed in 2020 and here is the direct linguistic from the NCAA website:
"With the changes announced in May 2020, the NET will no longer use winning percentage, adjusted winning percentage and scoring margin. The change was made after the committee consulted with Google Cloud Professional Services, which worked with the NCAA to develop the original NET."

So the committee is a carpenter, the tools they use are NET (incorporates SoS, results, location of games, offensive and defensive efficiency), profile good wins, Tournament champ, reg season Champ, profile bad losses, injuries to key players, trend of team as other examples.

UConn being number 2 means simply they performed better against their schedule than other teams did against their schedule. With 5 losses, UConn will be a 9-12 overall rank (#3 seed) in my estimation. The Committee releases are showing how they have valued these other items.1-SC, 2-Iowa, 3-Stanford, 4-USC, 5-Texas, 6-LSU, 7-UCLA, 8-tOSU, 9-OreSt, 10-UConn, 11-NC State, 12-ND.

Frankly as long as UConn is in Albany, who only BID on hosting knowing that the UCONN fans would buy the tickets, I will be excited. Or if Iowa and SC get placed there, then I can charge a huge premium on my seats.