Offical Net ranking thread

STrib: Gophers' low NET ranking puzzles coach Richard Pitino

The Gophers are one of the best examples in college hoops of the difference between NET and RPI. They are 16-5 overall and 6-4 in the Big Ten, but their NET after Wednesday’s 86-75 victory over Illinois was 50th. That’s 10th in the Big Ten, even behind Indiana (No. 48) after the Hoosiers dropped to 12-9 and lost their seventh game in a row Wednesday at Rutgers. Minnesota’s RPI is No. 33, while Indiana is No. 66.

Those RPI rankings seem more accurate to what actually is going on, at least right now. One team is playing well and beat some quality opponents (four wins vs. top 35 NET teams). The other is really on a nose dive.

“I’m not saying we’re the best team in the country, but with the résumé we have right now,” Gophers coach Richard Pitino said Wednesday, “[Minnesota’s NET] makes no sense to me.”

So why is this happening? Well, NET appears to focus more on scoring margin and efficiency numbers than RPI did, according to analysts. The NCAA hasn’t said that, but it revealed the NET’s five factors: game results, offensive and defensive efficiency, winning percentage, adjusted win percentage and scoring margin.

http://www.startribune.com/gophers-low-net-ranking-puzzles-coach-richard-pitino/505176182/

Go Gophers!!
 

STrib: Gophers' low NET ranking puzzles coach Richard Pitino

The Gophers are one of the best examples in college hoops of the difference between NET and RPI. They are 16-5 overall and 6-4 in the Big Ten, but their NET after Wednesday’s 86-75 victory over Illinois was 50th. That’s 10th in the Big Ten, even behind Indiana (No. 48) after the Hoosiers dropped to 12-9 and lost their seventh game in a row Wednesday at Rutgers. Minnesota’s RPI is No. 33, while Indiana is No. 66.

Those RPI rankings seem more accurate to what actually is going on, at least right now. One team is playing well and beat some quality opponents (four wins vs. top 35 NET teams). The other is really on a nose dive.

“I’m not saying we’re the best team in the country, but with the résumé we have right now,” Gophers coach Richard Pitino said Wednesday, “[Minnesota’s NET] makes no sense to me.”

So why is this happening? Well, NET appears to focus more on scoring margin and efficiency numbers than RPI did, according to analysts. The NCAA hasn’t said that, but it revealed the NET’s five factors: game results, offensive and defensive efficiency, winning percentage, adjusted win percentage and scoring margin.

http://www.startribune.com/gophers-low-net-ranking-puzzles-coach-richard-pitino/505176182/

Go Gophers!!

From the Article that Bleed linked:
Pitino admitted after the 11-point victory over Illinois on Wednesday that he kept his starters in the game to make sure the final margin was at least 10 points. The NET caps point differential at 10, but the efficiency numbers can look better the more you distance yourself in points.

“I’m never trying to show up another coach,” Pitino said. “But if they’re going to tell me a win by more than 10 points means more, then if we can get it, I’m going to get it. It’s hard to win games in this league by double digits.”


This is confirmation that coaches are paying attention to it and it is affecting their decision making. That sucks for the guys at the end of the bench when a coach is forced to leave in their starters to increase the margin of victory in order to improve their team's ranking for the computers.

Also, it is a joke that we would be behind Indiana in any metric right now. That fact alone tells you there is something off with it.
 

Yes, pretty obvious when looking at the NET rankings that increasing the value of sos would need to be a tweak for next season. Has been mentioned before, but one main problem is this whole thing is difficult because the sample space for each team is so different. Also, when NET was rolled out the NCAA did not present it as a minor, somewhat insignificant new metric, but rather an important new tool. Therefore, very understandable that coaches, broadcasters, and bracketologists (see bracket matrix) are giving it a lot of weight.

The other tweek needs to be removal of the "margin of victory" portion. It's being double-counted right now. One as a "margin of victory up to 10 points" and the other category of "team efficiency" which is very largely a "How wide of a scoring margin do you usually have".
 

The Gophers should get in the tournament if they win at least 10 conference games. Go 11-9 or better and they are a lock. In my opinion, NET ranking will only come into play for seeding and/or if they go 9-11 in conference play but maybe win a few in the BT tourney.
 

Maybe I'm not understanding this, but based on the few articles I've seen recently this does not make any sense. Scoring margin should not be factored in to computer rankings. Making a team continue to try to score like we did against Illinois so we can meet some random committee-determined threshold is just stupid.

Which is more impressive? Being behind for most of the game but pulling ahead and winning by 10 in the final minutes because the other team won't stop fouling you in the final 5 seconds when you're already up by 8 points? Or being up by 20 points for most of the game, pulling your starters and then winning by only 9?

Can their algorithms account for this? This seems like something that needs to be a human evaluation, and is why you would have a human committee, to determine these kinds of intricacies. I hope these NET rankings are not taken so literally when it comes to be selection time.
 


My favorite example is Iowa State at #12 Kenpom and #14 NET with an identical 16-5 record as Minnesota. Kenpom ranks ISU SoS at 45, Minnesota at 43. Several common opponents like Iowa (ISU lost to them), Illinois, Ok State.

With other variables that similar as Minnesota, how you win has to be a huge part of NET.
 

https://twitter.com/KevinPauga/status/1091042022465638400

This is a pretty reasonable explanation of the NET rankings coming from a guy who was part of the conversation to create them, but still has his own metric to evaluate teams.

I especially appreciate this part: Teams will total about 2,000-3,000 offensive and defensive possessions in a season. A small sample of late possessions isn't moving your NET really at all (maybe a spot or two in multiple extreme circumstances). I understand the rightful concern, but let's not overthink this.
 

https://twitter.com/KevinPauga/status/1091042022465638400

This is a pretty reasonable explanation of the NET rankings coming from a guy who was part of the conversation to create them, but still has his own metric to evaluate teams.

I especially appreciate this part: Teams will total about 2,000-3,000 offensive and defensive possessions in a season. A small sample of late possessions isn't moving your NET really at all (maybe a spot or two in multiple extreme circumstances). I understand the rightful concern, but let's not overthink this.

I agree the not overthinking part. No way IU or Nebraska should be ahead of the Gophers right now in any ranking. There...took no thinking at all.


Sent from my iPhone using Tapatalk
 

The new NET evaluation systems gives your team bonus points if your players are smiling when they score. Also, putting someone on a poster is frowned upon as being "intent to make another look bad".

And in deeper investigation, "If players on your team stick out their knees to impede driving players or flop at a high percentage you activate double bonus points."

I took these statements directly from the NET guide. I swear.
 



https://twitter.com/KevinPauga/status/1091042022465638400

This is a pretty reasonable explanation of the NET rankings coming from a guy who was part of the conversation to create them, but still has his own metric to evaluate teams.

I especially appreciate this part: Teams will total about 2,000-3,000 offensive and defensive possessions in a season. A small sample of late possessions isn't moving your NET really at all (maybe a spot or two in multiple extreme circumstances). I understand the rightful concern, but let's not overthink this.

No issue. Just heading into the homestretch of the college basketball season and the NET rankings are a joke. What do you say? Let's not overthink this.....?
 

To me, this points to scheduling your non-conference cupcakes as weak as possible and then absolutely drilling them. Is that what the NCAA is trying to promote?
 

https://twitter.com/KevinPauga/status/1091042022465638400

This is a pretty reasonable explanation of the NET rankings coming from a guy who was part of the conversation to create them, but still has his own metric to evaluate teams.

I especially appreciate this part: Teams will total about 2,000-3,000 offensive and defensive possessions in a season. A small sample of late possessions isn't moving your NET really at all (maybe a spot or two in multiple extreme circumstances). I understand the rightful concern, but let's not overthink this.

He's not addressing the "margin of victory" component in that phrase. Only the efficiency standard.

Also, with efficiency, he's down talking the last possession or two, but it's equally important in the metric as the first possession of the game or any possession in the 2nd half.
So to say it's not important is disingenuous.

Essentially, if you want to play your subs for two minutes, you may as well play them in the first half or early 2nd half because the possessions all mean the same regardless of whether you are winning or losing.
 

https://twitter.com/KevinPauga/status/1091042022465638400

This is a pretty reasonable explanation of the NET rankings coming from a guy who was part of the conversation to create them, but still has his own metric to evaluate teams.

I especially appreciate this part: Teams will total about 2,000-3,000 offensive and defensive possessions in a season. A small sample of late possessions isn't moving your NET really at all (maybe a spot or two in multiple extreme circumstances). I understand the rightful concern, but let's not overthink this.

Impossible to reconcile his statement with the rankings. See Iowa State example above. The only way his statement makes sense is if scoring margin is a huge consideration, which is another problem.
 



He's not addressing the "margin of victory" component in that phrase. Only the efficiency standard.

Also, with efficiency, he's down talking the last possession or two, but it's equally important in the metric as the first possession of the game or any possession in the 2nd half.
So to say it's not important is disingenuous.

Essentially, if you want to play your subs for two minutes, you may as well play them in the first half or early 2nd half because the possessions all mean the same regardless of whether you are winning or losing.

He's just saying it doesn't have much of an effect. Which is true. Even if you are talking about the last two minutes of every blowout game -- you are still only looking at maybe 6-7 possessions in that time of each game that was enough of a blowout to put in reserves.

We've had 11 games where the winning/losing margin was more than 10 points.
Figure we have another 5 to end the season.
Figure there are 7 possessions in the final two minutes of each of those games.
16 x 7 = 112
So 112 out of 2,000 (if you take the minimum) possessions in the season is less than 6%.
Plus the other team likely doesn't have their starters in at that point either.

That's a pretty small thing to be worried about. That's his point.
 
Last edited:

Impossible to reconcile his statement with the rankings. See Iowa State example above. The only way his statement makes sense is if scoring margin is a huge consideration, which is another problem.

Why is that a problem? Are you saying that a 2 pt win should be counted the same as a 30 pt win?

And scoring margin is not a HUGE consideration, it's just one of many factors.
 

Why is that a problem? Are you saying that a 2 pt win should be counted the same as a 30 pt win?

And scoring margin is not a HUGE consideration, it's just one of many factors.

Hia quote indicates efficiency is a very minor consideratiin. How does this square with the Iowa State, among others, example?
 

From the Article that Bleed linked:
Pitino admitted after the 11-point victory over Illinois on Wednesday that he kept his starters in the game to make sure the final margin was at least 10 points. The NET caps point differential at 10, but the efficiency numbers can look better the more you distance yourself in points.

“I’m never trying to show up another coach,” Pitino said. “But if they’re going to tell me a win by more than 10 points means more, then if we can get it, I’m going to get it. It’s hard to win games in this league by double digits.”


This is confirmation that coaches are paying attention to it and it is affecting their decision making. That sucks for the guys at the end of the bench when a coach is forced to leave in their starters to increase the margin of victory in order to improve their team's ranking for the computers.

Also, it is a joke that we would be behind Indiana in any metric right now. That fact alone tells you there is something off with it.
That's exactly my concern. Is this becoming coach to the metric, a bit like teach to the test?
If followed, I think this will interrupt player development. Someone brought up scrubs, but did/will this keep Omersa out of some real time minutes?
 

That's exactly my concern. Is this becoming coach to the metric, a bit like teach to the test?
If followed, I think this will interrupt player development. Someone brought up scrubs, but did/will this keep Omersa out of some real time minutes?

It's called "posturing"

He's mad because it doesn't rank us as well as he thinks it should. If it ranked us #5 he'd be talking about how great it is.
 

<script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>


Rick Pitino chimes in. I'm thinking he may still be talking to Richard.
 

Hia quote indicates efficiency is a very minor consideratiin. How does this square with the Iowa State, among others, example?

His quote indicates that the last two minutes of a game (when a team might have their reserves in the game if it is a blowout) is a very minor thing. He didn't say efficiency as a whole isn't a big consideration.
 

He's just saying it doesn't have much of an effect. Which is true. Even if you are talking about the last two minutes of every blowout game -- you are still only looking at maybe 6-7 possessions in that time of each game that was enough of a blowout to put in reserves.

We've had 11 games where the winning/losing margin was more than 10 points.
Figure we have another 5 to end the season.
Figure there are 7 possessions in the final two minutes of each of those games.
16 x 7 = 112
So 112 out of 2,000 (if you take the minimum) possessions in the season is less than 6%.
Plus the other team likely doesn't have their starters in at that point either.

That's a pretty small thing to be worried about. That's his point.

Okay, so let's hypothetically concede your point. Then are we to believe that the current NET rankings are in fact a solid representation of who is good and who is not? And by extension that the old rankings were incorrect and that we were putting the wrong teams in the tournament? And considering the wide variations we see between, let's say the RPI and NET, that not only were we putting in the wrong teams, but completely the wrong teams?
 

Another concern: the game has oft been abandoned during the telecast to discuss the NET ranking. Going forward, will we have a prayer of knowing how many fouls there are?
 

His quote indicates that the last two minutes of a game (when a team might have their reserves in the game if it is a blowout) is a very minor thing. He didn't say efficiency as a whole isn't a big consideration.

My bad, read his quote too quickly. Problem I could see would be situation like Illinois game. Pitino had to consider not only efficiency but also scoring margin. Even though for just a few possessions, this would be impactful. Can see his point, though, as well. Would probably be affected as well by how many times a team finds itself in this situation.
 

Okay, so let's hypothetically concede your point. Then are we to believe that the current NET rankings are in fact a solid representation of who is good and who is not? And by extension that the old rankings were incorrect and that we were putting the wrong teams in the tournament? And considering the wide variations we see between, let's say the RPI and NET, that not only were we putting in the wrong teams, but completely the wrong teams?

1. It's a working model, not a finished product. I don't think it's perfect by any means. Tweaks will be made in the coming years. I like the idea of combining game results with efficiency measures. The premise behind it is in the right place in my opinion, though I don't think it's perfect yet. It probably never will be, some fan base will always be mad.

2. RPI was a horrible metric, and there is a reason that everyone from coaches/fans/administrators/etc wanted to see it go.

3. NET and RPI are the same in one way -- they each are just one of the metrics used. I do the bracketology thing each year, and I include every metric that the selection committee considers in the formula. There are a ton of things that have nothing to do with NET or RPI that are considered. KenPom, KPI, Sagarin, ESPN BPI, ESPN SOR, road record are all considered. Heck, the biggest factor in the model I use is winning percentage. None of those things are based on the NET rankings.

I guess my point is that there are so many factors considered, and I sort of get it why NET is singled out (because it is directly from the NCAA), but I think it's a little funny that people seem to have zeroed in on NET when other ranking metrics have us rated in a similar way. NET doesn't even have us rated the lowest out of all the different metrics used.

And last but not least -- let's wait to see what it looks like at the end of the season.
 
Last edited:

Net ranking is not the sole answer to the at large field correct?

It's just a comparison metric and not even THE comparison metric, but one of several comparison metrics?

In that light, while flawed, it's probably a decent indicator of what a team COULD do or WOULD do in a game against an unnamed opponent.
 

Net ranking is not the sole answer to the at large field correct?

It's just a comparison metric and not even THE comparison metric, but one of several comparison metrics?

In that light, while flawed, it's probably a decent indicator of what a team COULD do or WOULD do in a game against an unnamed opponent.

Correct. It is one of six metrics used and other factors are considered as well.
 




Or scheduling good teams and beating them all. Then your metrics will be awesome. Like if your 20-1 you will have great metrics. Simply can not play poorly or even average and be 20-1. Beating good teams means playing well.
 




Top Bottom