|
Post by Pinnum on Nov 26, 2018 13:31:21 GMT -5
The NCAA no longer compiles or uses the RPI. The membership voted to use the 'NET' ranking which is a composite of computer rankings. It is similar to the old BCS computer rankings. www.ncaa.com/rankings/basketball-men/d1/ncaa-mens-basketball-net-rankingsHere is the A10 standings in the new metric: 40 Duquesne 4-1 56 VCU 5-1 62 Saint Louis 5-1 69 Dayton 4-2 100 Davidson 5-1 127 Fordham 5-1 149 Saint Joseph's 3-3 160 Rhode Island 2-2 172 Massachusetts 4-3 257 George Mason 2-5 263 Richmond 2-4 275 George Washington 1-5 322 St. Bonaventure 1-5 336 La Salle 0-7 For comparison, this is what the RPI would have for the A10 if it were still used... (Note: The RPI was not accurate this early in the season) 15 Duquesne 4-1 35 VCU 5-1 61 Davidson 5-1 85 Saint Louis 5-1 95 Dayton 4-2 180 Fordham 5-1 183 Rhode Island 2-2 199 Saint Joseph's 3-3 220 Massachusetts 4-3 244 George Washington 1-5 247 George Mason 2-5 314 Richmond 2-4 320 La Salle 0-7 349 St. Bonaventure 1-5
|
|
|
Post by sbu79 on Nov 26, 2018 13:49:09 GMT -5
And then a committee will largely ignore it all.
|
|
|
Post by class70 on Nov 26, 2018 13:59:12 GMT -5
|
|
|
Post by Pinnum on Nov 26, 2018 14:01:25 GMT -5
As of right now, there are four teams in at-large contention.
Obviously, teams can improve their position or see it drop off. If the A10 can have 4-6 top-75 teams going into the conference slate, the conference will be in good shape as long as the conference doesn't cannibalize itself.
|
|
|
Post by fjs64 on Nov 26, 2018 14:17:34 GMT -5
|
|
|
Post by thebonafan08 on Nov 28, 2018 10:19:47 GMT -5
The NCAA no longer compiles or uses the RPI. The membership voted to use the 'NET' ranking which is a composite of computer rankings. It is similar to the old BCS computer rankings. www.ncaa.com/rankings/basketball-men/d1/ncaa-mens-basketball-net-rankingsHere is the A10 standings in the new metric: 40 Duquesne 4-1 56 VCU 5-1 62 Saint Louis 5-1 69 Dayton 4-2 100 Davidson 5-1 127 Fordham 5-1 149 Saint Joseph's 3-3 160 Rhode Island 2-2 172 Massachusetts 4-3 257 George Mason 2-5 263 Richmond 2-4 275 George Washington 1-5 322 St. Bonaventure 1-5 336 La Salle 0-7 For comparison, this is what the RPI would have for the A10 if it were still used... (Note: The RPI was not accurate this early in the season) 15 Duquesne 4-1 35 VCU 5-1 61 Davidson 5-1 85 Saint Louis 5-1 95 Dayton 4-2 180 Fordham 5-1 183 Rhode Island 2-2 199 Saint Joseph's 3-3 220 Massachusetts 4-3 244 George Washington 1-5 247 George Mason 2-5 314 Richmond 2-4 320 La Salle 0-7 349 St. Bonaventure 1-5 I understood NET as it’s own formula entirely and not a composite of rankings. One of the controversies with its initial release is that they did not release the exact calculation for rankings.
|
|
|
Post by Pinnum on Nov 28, 2018 11:07:38 GMT -5
I understood NET as it’s own formula entirely and not a composite of rankings. One of the controversies with its initial release is that they did not release the exact calculation for rankings. It is a formula all its own that has multiple computer calculations and AI components.
|
|
|
Post by thebonafan08 on Nov 28, 2018 11:10:57 GMT -5
So.....NET is not like BCS which was a compiling of published rankings and polls to come up with a ranking. NET is its own formula independent of other ranking systems.
|
|
|
Post by Pinnum on Nov 28, 2018 11:26:53 GMT -5
So.....NET is not like BCS which was a compiling of published rankings and polls to come up with a ranking. NET is its own formula independent of other ranking systems. Yes, not exactly the public rankings but they consulted with the major basketball statisticians when building it.
|
|
|
Post by kcSBU03 on Nov 28, 2018 11:43:00 GMT -5
In a nutshell, NET = RPI + offensive efficiecy, defensive efficiency, margin and some secret sauce
|
|
|
Post by trout95 on May 12, 2020 11:29:11 GMT -5
|
|
|
Post by ceharv on May 12, 2020 13:00:33 GMT -5
It’s all just talk until some Powerconf team with a winning record gets left out of the tourney because their sched had too many cupcakes at home. Wake me when that happens.
|
|
|
Post by agoo on May 12, 2020 13:10:47 GMT -5
Removing winning percentage seems to be a way to get ahead of the calls for teams that only finish above .500 in Conference play being eligible.
|
|
|
Post by fjs64 on May 12, 2020 13:39:41 GMT -5
Removing winning percentage seems to be a way to get ahead of the calls for teams that only finish above .500 in Conference play being eligible. I was thinking the same thing, especially if next season is truncated and teams play only conference opponents.
|
|
|
Post by Pinnum on May 12, 2020 13:53:25 GMT -5
The winning percentage component helped teams with a lot of wins (single bid conference champions) who might not have got a chance to play other highly rated teams.
The winning percentage component also harmed the top conferences that expanded their number of games and beat up on each other.
The Margin of Victory was criticized as being redundant as it is already in the efficiency metric.
These changes do mean that if you play to the level of your opponent you will be helped a lot if you play a schedule against top teams and you will be harmed a lot if you play a schedule of weak teams.
|
|