|
Post by jdaztec on Feb 13, 2020 15:30:54 GMT -8
Good question Survalli
|
|
|
Post by Motown Monty on Feb 13, 2020 17:55:32 GMT -8
A Quad 2 road loss at this stage in the season is not too damaging.
|
|
|
Post by Motown Monty on Feb 13, 2020 18:00:52 GMT -8
My guess is if we lose to Boise we probably only drop a few spots. Likely still be in the top 3 or 4. A loss at BSU would also be a Quad 2 loss and likely have a similarly minimal affect on our NET ranking.
|
|
|
Post by Motown Monty on Feb 13, 2020 18:02:49 GMT -8
Hey, they have SDSU at #1 so they must be doing something right! BTW, I thought our sorry MWC schedule was going to brings us down on the NET ranking...seems that as long as we keep winning that won't be the case? I'll admit I made that assumption but our undefeated record is buoying the ranking.
|
|
|
Post by fisherville on Feb 13, 2020 20:44:00 GMT -8
There are a lot of factors that go into net and it doesn't show the difference team by team. I'm sure the overall rating dropped a little but losing a 6 point game against a top 100 team on the road won't have a big impact at this point.
|
|
|
Post by TheSanDiegan on Feb 13, 2020 23:30:18 GMT -8
A Quad 2 road loss at this stage in the season is not too damaging. This is true. However, Arizona has no business being ranked above the three teams below it in NET (and possibly more). Arizona has a Q1 record of 2-5 (.286 WP), and a combined Q1-Q2 record of 8-7. Referring to AztecBill 's thread of the last week or so ("A Little Math"), the probability of this outcome (assuming a 50-50 chance of winning a given game) is .19638 (19.64%). By contrast, WVU, Michigan State, and Butler all have better combined Q1-Q2 records with significantly lower associated probabilities of occurrence. For instance, Butler, which currently sits three places behind Arizona, has a Q1 record of 8-5 and a combined Q1-Q2 record of 13-6. The probability of amassing that record is .05175 (5.175%), much lower than Arizona's comparably pedestrian 19.64% chance of amassing their record. Similarly, WVU and Mich. St. both have lower probabilities associated with their respective records. All three also have better combined road and neutral court records. All three should be ranked above Arizona.
|
|
|
Post by moctezumaii on Feb 14, 2020 4:42:55 GMT -8
A Quad 2 road loss at this stage in the season is not too damaging. This is true. However, Arizona has no business being ranked above the three teams below it in NET (and possibly more). Arizona has a Q1 record of 2-5 (.286 WP), and a combined Q1-Q2 record of 8-7. Referring to AztecBill 's thread of the last week or so ("A Little Math"), the probability of this outcome (assuming a 50-50 chance of winning a given game) is .19638 (19.64%). By contrast, WVU, Michigan State, and Butler all have better combined Q1-Q2 records with significantly lower associated probabilities of occurrence. For instance, Butler, which currently sits three places behind Arizona, has a Q1 record of 8-5 and a combined Q1-Q2 record of 13-6. The probability of amassing that record is .05175 (5.175%), much lower than Arizona's comparably pedestrian 19.64% chance of amassing their record. Similarly, WVU and Mich. St. both have lower probabilities associated with their respective records. All three also have better combined road and neutral court records. All three should be ranked above Arizona. Yeah, but did you see the Iowa Caucuses? : )
|
|
|
Post by Motown Monty on Feb 14, 2020 5:29:00 GMT -8
A Quad 2 road loss at this stage in the season is not too damaging. This is true. However, Arizona has no business being ranked above the three teams below it in NET (and possibly more). Arizona has a Q1 record of 2-5 (.286 WP), and a combined Q1-Q2 record of 8-7. Referring to AztecBill 's thread of the last week or so ("A Little Math"), the probability of this outcome (assuming a 50-50 chance of winning a given game) is .19638 (19.64%). By contrast, WVU, Michigan State, and Butler all have better combined Q1-Q2 records with significantly lower associated probabilities of occurrence. For instance, Butler, which currently sits three places behind Arizona, has a Q1 record of 8-5 and a combined Q1-Q2 record of 13-6. The probability of amassing that record is .05175 (5.175%), much lower than Arizona's comparably pedestrian 19.64% chance of amassing their record. Similarly, WVU and Mich. St. both have lower probabilities associated with their respective records. All three also have better combined road and neutral court records. All three should be ranked above Arizona. And there’s a reason Arizona is a 6 Seed in most bracket projections. NET is only a tool the committee uses and not the end all when it comes to seed lists.
|
|
|
Post by DeeMoney on Feb 14, 2020 8:42:36 GMT -8
Louisville lost by 6 on the road to a good team, the NET metrics only probably had them favored by 5 or so going in. So in regards to expected outcome vs actual outcome its not that big of a swing (11 points) when you take into account the other 25 games worth of data.
Now scoring margin is only one part of NET, as has been mentioned the others also point to LVille not moving that much due to a single digit road loss to a good team. There is a lot of data out there to balance out vs a single event.
|
|
|
Post by TheSanDiegan on Feb 14, 2020 9:06:44 GMT -8
This is true. However, Arizona has no business being ranked above the three teams below it in NET (and possibly more). Arizona has a Q1 record of 2-5 (.286 WP), and a combined Q1-Q2 record of 8-7. Referring to AztecBill 's thread of the last week or so ("A Little Math"), the probability of this outcome (assuming a 50-50 chance of winning a given game) is .19638 (19.64%). By contrast, WVU, Michigan State, and Butler all have better combined Q1-Q2 records with significantly lower associated probabilities of occurrence. For instance, Butler, which currently sits three places behind Arizona, has a Q1 record of 8-5 and a combined Q1-Q2 record of 13-6. The probability of amassing that record is .05175 (5.175%), much lower than Arizona's comparably pedestrian 19.64% chance of amassing their record. Similarly, WVU and Mich. St. both have lower probabilities associated with their respective records. All three also have better combined road and neutral court records. All three should be ranked above Arizona. And there’s a reason Arizona is a 6 Seed in most bracket projections. NET is only a tool the committee uses and not the end all when it comes to seed lists. What you say is true, but by what metric then do they deserve to be 9th?? They've sh¡t on the road (2-4) and have the a horrible record against Q1 teams (2-5). Maybe on the douchey fanbase scale they should rank top 10, but not in any MBB metric that gauges success based on performance. Furthermore, when you see a disparity in respective movements that reflects a lack of consistency in adjustments to the model, it is indicative of either i) serious flaws in the model (which need to be addressed) or manipulation of the results. Either possibility merits further discussion/investigation.
|
|
|
Post by DeeMoney on Feb 14, 2020 10:14:12 GMT -8
Regarding Arizona, once again you have to look at Scoring Margins. I know, I know wins and losses are what matter (and when it comes to ranking teams I mostly agree), but the best predictive metrics use scoring margin to take into account how good a team is.
Consider Arizona's losses: 5pt loss AT Baylor, 4pt vs Gonzaga (right there thats 4 possesions difference between beating the top 2 teams). 3pt neutral court v St Johns, 1 pt loss at Oregon, 1 pt loss at Az St. Those are three VERY GOOD games with a 5 pt. difference.
Mix in blowout wins against Colorado, at Wazzu, and neutral vs Wake. You can start to see why AZ is where they are.
|
|
|
Post by moctezumaii on Feb 14, 2020 11:46:44 GMT -8
And there’s a reason Arizona is a 6 Seed in most bracket projections. NET is only a tool the committee uses and not the end all when it comes to seed lists. What you say is true, but by what metric then do they deserve to be 9th?? They've sh¡t on the road (2-4) and have the a horrible record against Q1 teams (2-5). Maybe on the douchey fanbase scale they should rank top 10, but not in any MBB metric that gauges success based on performance. Furthermore, when you see a disparity in respective movements that reflects a lack of consistency in adjustments to the model, it is indicative of either i) serious flaws in the model (which need to be addressed) or manipulation of the results. Either possibility merits further discussion/investigation. Definitely sounds like Iowa peeps moonlighting on the NET to me.
|
|
|
Post by TheSanDiegan on Feb 14, 2020 12:11:39 GMT -8
What you say is true, but by what metric then do they deserve to be 9th?? They've sh¡t on the road (2-4) and have the a horrible record against Q1 teams (2-5). Maybe on the douchey fanbase scale they should rank top 10, but not in any MBB metric that gauges success based on performance. Furthermore, when you see a disparity in respective movements that reflects a lack of consistency in adjustments to the model, it is indicative of either i) serious flaws in the model (which need to be addressed) or manipulation of the results. Either possibility merits further discussion/investigation. Definitely sounds like Iowa peeps moonlighting on the NET to me. I'm totally down with the idea of a Hawkeye troll farm. They have a vested interest in seeing our opponents ranked above Gonzaga's.
|
|
|
Post by tonatiuh on Feb 14, 2020 15:58:08 GMT -8
A very close friend and a good source from Kansas told me that the NET is run by the Wizard of Oz. And, his assistant is ToTo, but he won't say a thing about what is going on!
|
|
|
Post by aztecandrew on Feb 16, 2020 16:57:56 GMT -8
The BIG question will be: does the committee have enough faith in its own tool to give the #1 overall tournament seed to the team that has the highest NET?
If they don’t, they’re admitting that the tool that they developed isn’t worth crap.
|
|
|
Post by Motown Monty on Feb 16, 2020 17:13:01 GMT -8
The BIG question will be: does the committee have enough faith in its own tool to give the #1 overall tournament seed to the team that has the highest NET? If they don’t, they’re admitting that the tool that they developed isn’t worth crap. Again, the NET is one of the tools the committee uses to develop the seed list, ONE. It essentially replaced the oft maligned RPI. RPI wasn't used exclusively (thank goodness) by the committee either.
|
|