Net Rankings Please Explain

Madabs

Sophomore
Joined
Dec 29, 2018
Messages
1,204
Obviously, Quad 1 and 2 Wins are quality. But what other factors contribute to the NET rankings that could have Virginia Tech Ranked 38th? They have 0 Quad 1 wins and they only have 1 less Quad 2 loss 4-3 vs 4-4. Does someone know what the deal is?

CurrentPreviousTeam ConferenceRecordRoadNeutralHomeQuad 1Quad 2Quad 3Quad 4
1111DukeACC21-46-22-013-25-13-26-17-0
3738North CarolinaACC18-75-30-313-10-74-08-06-0
3839Virginia TechACC15-105-51-29-30-54-35-26-0
4040Wake ForestACC20-65-32-113-21-34-36-09-0
5656Notre DameACC17-77-30-310-12-53-18-14-0
6867Miami (FL)ACC18-76-22-210-34-14-46-24-0
7777VirginiaACC16-94-52-010-42-44-13-47-0
8585SyracuseACC13-123-61-39-30-63-16-54-0
8888ClemsonACC12-132-62-28-50-41-66-25-1
9797Florida St.ACC13-113-62-18-42-42-34-35-1
126126LouisvilleACC11-132-72-07-60-33-63-45-0
137136NC StateACC10-162-51-37-81-61-43-55-1
151153Georgia TechACC9-142-50-27-70-70-54-15-1
168169Boston CollegeACC9-141-70-28-50-43-40-56-1
180180PittsburghACC10-162-80-08-80-32-65-53-2
 
Advertisement
Obviously, Quad 1 and 2 Wins are quality. But what other factors contribute to the NET rankings that could have Virginia Tech Ranked 38th? They have 0 Quad 1 wins and they only have 1 less Quad 2 loss 4-3 vs 4-4. Does someone know what the deal is?

CurrentPreviousTeamConferenceRecordRoadNeutralHomeQuad 1Quad 2Quad 3Quad 4
1111DukeACC21-46-22-013-25-13-26-17-0
3738North CarolinaACC18-75-30-313-10-74-08-06-0
3839Virginia TechACC15-105-51-29-30-54-35-26-0
4040Wake ForestACC20-65-32-113-21-34-36-09-0
5656Notre DameACC17-77-30-310-12-53-18-14-0
6867Miami (FL)ACC18-76-22-210-34-14-46-24-0
7777VirginiaACC16-94-52-010-42-44-13-47-0
8585SyracuseACC13-123-61-39-30-63-16-54-0
8888ClemsonACC12-132-62-28-50-41-66-25-1
9797Florida St.ACC13-113-62-18-42-42-34-35-1
126126LouisvilleACC11-132-72-07-60-33-63-45-0
137136NC StateACC10-162-51-37-81-61-43-55-1
151153Georgia TechACC9-142-50-27-70-70-54-15-1
168169Boston CollegeACC9-141-70-28-50-43-40-56-1
180180PittsburghACC10-162-80-08-80-32-65-53-2
Makes zero sense to me either. We swept a Wake team that has just 1 Quad 1 wins (we have 4), yet they're nearly 30 spots above us. We blew out a UNC team that has zero Quad 1 wins, and they're also 30 spots ahead.
 
Makes zero sense to me either. We swept a Wake team that has just 1 Quad 1 wins (we have 4), yet they're nearly 30 spots above us. We blew out a UNC team that has zero Quad 1 wins, and they're also 30 spots ahead.
The algorithm makes no sense and no one truly knows what it is. Brilliant.
 
Advertisement
In the "traditonal" polls, tied for 37th in the AP and 32nd (one behind Wake) in the Coaches Poll.
 
The 2021-22 men's basketball season marks the fourth season of the NCAA Evaluation Tool (NET) rankings, which replaced the RPI prior to the 2018-19 season as the primary sorting tool for evaluating teams. In May 2020, the NCAA announced there will be changes made to the NCAA Evaluation Tool to increase accuracy and simplify it by reducing a five-component metric to just two.

The remaining factors include the Team Value Index (TVI), which is a result-based feature that rewards teams for beating quality opponents, particularly away from home, as well as an adjusted net efficiency rating. The adjusted efficiency is a team’s net efficiency, adjusted for strength of opponent and location (home/away/neutral) across all games played. For example, a given efficiency value (net points per 100 possessions) against stronger opposition rates higher than the same efficiency against lesser opponents and having a certain efficiency on the road rates higher than the same efficiency at home.
 
Last edited:
Advertisement
Is there any notable data not included in the NET?

Game date and game order were not included in the NET rankings so a team's first game counts the same as its 30th.

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.

“When we adopted the NET in 2018, we had reviewed several seasons worth of data and we insisted that we would continue to evaluate the metric,” said Dan Gavitt, the NCAA’s senior vice president of basketball. “We’ve been very satisfied with its performance thus far, but it became evident after two seasons of use that this change would be an improvement. While we will continue to monitor the metric, I don’t anticipate any additional adjustments for several years. We believe this change will result in more precision throughout the season and will be easier for our membership and the public to understand."

The updated NET is consistent with the women’s basketball NET, which was revealed after the Division I Women’s Basketball Committee worked with a team from Google Cloud to evaluate women’s basketball statistical data for a 10-year period.

In addition, the overall and non-conference strength of schedule has been modernized to reflect a truer measure for how hard it is to defeat opponents. The strength of schedule is based on rating every game on a team's schedule for how hard it would be for an NCAA tournament-caliber team to win. It considers opponent strength and site of each game, assigning each game a difficulty score. Aggregating these across all games results in an overall expected win percentage versus a team's schedule, which can be ranked to get a better measure of the strength of schedule.
 
Last edited:
How are the NET rankings used?

Since the NET rankings serve as the primary sorting tool for Division I men's basketball, they play an important role in establishing a team's resume. The men’s and women’s basketball NET rankings and team sheets will be provided publicly on a daily basis on NCAA.com and NCAA.org starting in December.

Using the quadrant system, which was in its fourth season in 2020-21, the quality of wins and losses will be organized based on game location and the opponent's NET ranking.

Quadrant 1: Home 1-30, Neutral 1-50, Away 1-75
Quadrant 2: Home 31-75, Neutral 51-100, Away 76-135
Quadrant 3: Home 76-160, Neutral 101-200, Away 135-240
Quadrant 4: Home 161-353, Neutral 201-353, Away 241-353
The number of Quadrant 1 wins and Quadrant 3/4 losses will be incredibly important when it comes time for NCAA tournament selection and seeding.
 
I posted this two weeks ago, so some of the data is outdated, but it still largely fits:

Miami is still as low in the NET/Kenpom/Torvik as it is because of the non-conference. It's not just that we lost to Alabama, it's that we lost to Alabama by 32 points. It's not just that we lost to Dayton, it's that we lost to Dayton by 16 points. The only 'good' wins are a 6 points neutral court win over North Texas and a 5 point win at Penn State - neither of which are likely tourney teams. And then there are very few absolute blowouts against our cupcakes. FAU is actually technically a Q2 game at this moment (will probably be a Q3 game at the end of the year), but we only won that game by 2 points. Fordham is a below average A10 team, and we only beat them by 6. Stetson and Canisius are bad this season, but those were only 10 point wins. The only legitimate blow out games were FAMU and Lipscomb - two more terrible teams.

You can separate a team's conference and non-conference performance on Torvik. On Torvik, Miami is currently rated 62nd for the season. But if you only take into account conference play, Miami has been the 33rd best team in the country. On the flipside, just looking at the non-conference portion, Miami rated as the 106th best team in the country. This team is certainly closer to the former than the latter, but it is taking time to overcome where we started.

That also explains why Virginia Tech is as high in the NET ratings as they are. They rated as the 15th best team in NC play per Torvik (beat a top 100 St. Bonaventure team by 27 on a neutral court, a top 150-ish Navy team on the road by 20), but 106th in the nation so far solely off of their performance in conference.
 
I posted this two weeks ago, so some of the data is outdated, but it still largely fits:

Exactly, style points matter, and outside of UNC, we simply don't have any. As you detailed, that's especially true when playing non-con cupcakes like Stetson or Canisuis. You have to blow those teams out and we didn't.
 
Last edited:
Advertisement
Exactly, style points matter, and outside of UNC, we simply don't have any. As you detailed, that's especially true when playing non-con cupcakes like Stetson or Canisuis. You have to blow those teams out and we didn't.
Yes, and it is bull**** for paying the price for games in November and December. Just keep winning at this point and rest will take care of itself.
 


Reading between the lines, I think the problem is the non conference schedule.


NO. It’s really simple. The problem is that our defensive efficiency and overall efficiency is bad because we win close games (we give up more points per possession - including against bad teams).

Our strength of schedule is good. There is no weighting for OOC vs. conference. We let bad teams score points so our defensive efficiency is bad (outside the top 100), our offensive efficiency is good but not terrific (top 30)… so our overall ranking on an efficiency based ranking system is worse than our W-L record.
 
Advertisement
I posted this two weeks ago, so some of the data is outdated, but it still largely fits:


All of that stuff is true (statistically), but it further points out that the stats should be a starting point for the evaluation, not to mention what the limitations are on the stats themselves.

For instance, why should losing "by 32" be any worse than losing "by 16"? Maybe break losses down into "close games" (9 or fewer points) and "bad losses" (10 or more points) without piling on (statistically). Both of those games were at a tournament, and both of those games involved COVID "injuries". Now I know, I know, people will say that "injuries are a part of the game", and they are when you lose a player for an extended period of time. But the COVID situation has created scenarios in which a team on one night is fundamentally different than the same team on a different night and for no actual basketball reason, such as an injury.

Anyhow, I get the point that we had two double-digit losses at a neutral site tourney (though the worst loss was to a RANKED TEAM), but we also have 18 wins, one of which was on the road against a RANKED TEAM. We cannot control the fact that Duke has been the only consistently-ranked team in the ACC. ****, some other ACC teams might be ranked right now IF MIAMI HADN'T BEATEN THEM. At the end of the day, we have 18 wins, and I don't really care if all 18 were by 1 point each. We beat the teams. And they are not bad teams. Maybe the ACC isn't as stacked as in prior years, but we are beating quality teams on the regular.

Look, if the voters had ranked us when we beat Duke, and then dropped us down when we lost to F$U twice, maybe I would be more forgiving. But our remaining 6 games are against teams with worse conference records than us. We are tied for 3rd in the ACC and we don't have any games left against Duke, Notre Dame, or North Carolina (and we beat two of those three). We can never erase the Alabama and Dayton games, and we don't have any opportunities in the next 6 games to change minds either.

Ultimately, it's all some sort of bizarre joke. If you weren't expected to do well, then you do well, who "above you" can you beat to prove your own legitimacy? I also have a problem with other teams getting slobbered on because "oh, look at the talent they added in the portal", yet when Miami brings in some talented transfers and wins 18-24 regular season games (even if our margin of victory is single digits 18-24 times), we are just "slightly above average expectations".

Watch the games. Nearly every one of our 18 wins involved Miami absorbing the other team's best shot in the first half, followed by halftime adjustments and superior performance in the second half.
 
All of that stuff is true (statistically), but it further points out that the stats should be a starting point for the evaluation, not to mention what the limitations are on the stats themselves.

For instance, why should losing "by 32" be any worse than losing "by 16"? Maybe break losses down into "close games" (9 or fewer points) and "bad losses" (10 or more points) without piling on (statistically). Both of those games were at a tournament, and both of those games involved COVID "injuries". Now I know, I know, people will say that "injuries are a part of the game", and they are when you lose a player for an extended period of time. But the COVID situation has created scenarios in which a team on one night is fundamentally different than the same team on a different night and for no actual basketball reason, such as an injury.

Anyhow, I get the point that we had two double-digit losses at a neutral site tourney (though the worst loss was to a RANKED TEAM), but we also have 18 wins, one of which was on the road against a RANKED TEAM. We cannot control the fact that Duke has been the only consistently-ranked team in the ACC. ****, some other ACC teams might be ranked right now IF MIAMI HADN'T BEATEN THEM. At the end of the day, we have 18 wins, and I don't really care if all 18 were by 1 point each. We beat the teams. And they are not bad teams. Maybe the ACC isn't as stacked as in prior years, but we are beating quality teams on the regular.

Look, if the voters had ranked us when we beat Duke, and then dropped us down when we lost to F$U twice, maybe I would be more forgiving. But our remaining 6 games are against teams with worse conference records than us. We are tied for 3rd in the ACC and we don't have any games left against Duke, Notre Dame, or North Carolina (and we beat two of those three). We can never erase the Alabama and Dayton games, and we don't have any opportunities in the next 6 games to change minds either.

Ultimately, it's all some sort of bizarre joke. If you weren't expected to do well, then you do well, who "above you" can you beat to prove your own legitimacy? I also have a problem with other teams getting slobbered on because "oh, look at the talent they added in the portal", yet when Miami brings in some talented transfers and wins 18-24 regular season games (even if our margin of victory is single digits 18-24 times), we are just "slightly above average expectations".

Watch the games. Nearly every one of our 18 wins involved Miami absorbing the other team's best shot in the first half, followed by halftime adjustments and superior performance in the second half.

It’s not about how much you win or lose by. It’s about your average points per possession and your opponents average points per possession adjusted by quality of the individual opponent. Of course getting blown out is fewer points/poss for you and more for your opponent… hurting your efficiency metrics. Close games with bad teams do the same thing.

Points/poss on O and points/poss on D. Adjusted for quality of opponent. That’s it. That’s all.

It’s stupid because predictive metrics don’t sufficiently factor actual outcomes… so they make a poor basis for a performance based ranking system.
 
There's plenty of rankings out there for various metrics, including NET, but I don't see PPP (for O or D) anywhere. It would presumably be non-secretive, wonder why no one has published (an easy to find) list of the PPP standings.

I think we improved (PPP) last game because we had so many points off turnovers, compared to our usual sad "give it right back" prowess. Why this should have such a bearing on bracketing is of course debatable: if we steal 10 times and score on 5, we are much worse off PPP than a team that only steals twice, and converts each time. But it is what it is.

I agree SOS and opponents in general (or non-conf) are somehow less influential than metrics like PPP. But if there is some solidity to that being a major factor, it was nice to see Penn State (who we beat) upset ranked MSU last night!
 
All of that stuff is true (statistically), but it further points out that the stats should be a starting point for the evaluation, not to mention what the limitations are on the stats themselves.

For instance, why should losing "by 32" be any worse than losing "by 16"? Maybe break losses down into "close games" (9 or fewer points) and "bad losses" (10 or more points) without piling on (statistically). Both of those games were at a tournament, and both of those games involved COVID "injuries". Now I know, I know, people will say that "injuries are a part of the game", and they are when you lose a player for an extended period of time. But the COVID situation has created scenarios in which a team on one night is fundamentally different than the same team on a different night and for no actual basketball reason, such as an injury.

Anyhow, I get the point that we had two double-digit losses at a neutral site tourney (though the worst loss was to a RANKED TEAM), but we also have 18 wins, one of which was on the road against a RANKED TEAM. We cannot control the fact that Duke has been the only consistently-ranked team in the ACC. ****, some other ACC teams might be ranked right now IF MIAMI HADN'T BEATEN THEM. At the end of the day, we have 18 wins, and I don't really care if all 18 were by 1 point each. We beat the teams. And they are not bad teams. Maybe the ACC isn't as stacked as in prior years, but we are beating quality teams on the regular.

Look, if the voters had ranked us when we beat Duke, and then dropped us down when we lost to F$U twice, maybe I would be more forgiving. But our remaining 6 games are against teams with worse conference records than us. We are tied for 3rd in the ACC and we don't have any games left against Duke, Notre Dame, or North Carolina (and we beat two of those three). We can never erase the Alabama and Dayton games, and we don't have any opportunities in the next 6 games to change minds either.

Ultimately, it's all some sort of bizarre joke. If you weren't expected to do well, then you do well, who "above you" can you beat to prove your own legitimacy? I also have a problem with other teams getting slobbered on because "oh, look at the talent they added in the portal", yet when Miami brings in some talented transfers and wins 18-24 regular season games (even if our margin of victory is single digits 18-24 times), we are just "slightly above average expectations".

Watch the games. Nearly every one of our 18 wins involved Miami absorbing the other team's best shot in the first half, followed by halftime adjustments and superior performance in the second half.
First of all, we had our whole lineup for the Orlando tournament, so I don't know why you keep saying we didn't. The boxscore for the Alabama game shows that we played everyone. Dayton was the same, and I even looked back to the FAMU game before the tournament, and we had our full lineup there (so it wouldn't have been a case of someone having Covid and having to get back into shape during the tournament).

Of our top 8 players, only Waardenburg has missed multiple games (the home games against Wake and NC State). Wooga also missed the NC State game, while Jordan Miller missed the most recent Georgia Tech game and Bensley missed the Canisius game. (It's worth noting that all of those games were wins) The group of Kam, Wong, Moore, and Walker have played in every game. Obviously there are games were a player may be playing while he is banged up, but that's true of every team. Frankly, we've had more Covid/injury luck than most teams in the country this season.

What you are advocating more for rankings that are based purely on a resume, and I 100% agree that those metrics should matter more when it comes to considering who should be ranked, who should make the NCAA tournament, etc... Metrics such as strength of record or Torvik's Wins Above Bubble (which I think is the best measurement I've seen for pure resumes) I think should carry more weight than a team's individual NET rating. But you still need to rate the teams in some way to be able to use those, and in the aggregate, I prefer metrics like the NET, Kenpom, Torvik which do take into account things like offensive/defensive efficiency, victory margin, etc. over what we had in the RPI to do the rankings. All of those systems can be manipulated, but the RPI is the easiest of those groups to manipulate.

Will there be outliers with the NET where a team's quality is not accurately reflected? Sure, and Miami presents an argument about that. Providence this season provides an even stronger argument for that. Iowa and Virginia Tech may reflect teams that are on the opposite end of the spectrum, where they are highly rated because they blasted inferior competition but don't really have the quality wins to back it up.

For what it's worth, on Torvik's Wins Above Bubble metric, Miami is +2.1, which is good for 26th in the country (this includes the Louisville game, which gave us a boost of +0.5). In theory, a team that is right on the cut line would be right around 0 with this metric. That's a good place to be, but it also isn't a number that says this team must be ranked. Receiving votes but being one of the teams just outside of the top 25 is probably the right spot for Miami to be at this point of the season.
 
Last edited:
Advertisement
Back
Top