Manipulating the NET

[upperdeck]

Yup.. the game was always to beat bad teams that are less bad. Any team in the top 50 in reality should be beating those 110-300 teams almost 95% of the time at home.

 

[Cusefan0307]

The NET was invented to fkkkk mid-majors. College sports are a money grabbing joke and will only get worse.

With that said, how many days until conference tourneys start?

There's quite a few mid majors doing well for themselves in the NET. The Mountain West specifically is doing much better than the ACC.

 

[upperdeck]

Money to be made picking against them in the tourney then

That said, I'm not picking against Houston and their D

the problem with the tourney and the results is that seeding creates so much reward for the top teams.. for many years often the first 2 games were pretty easy for the top teams. Now not so much

Imagine a seeding that was blind draw. would the top 5-10 teams get thru to the sweet 16 so often if they got a harder schedule? these days the diff between a 3 seed and a 10 seed is almost nothing more than the schedule they played not how well they played.

 

[jncuse]

The NET was invented to fkkkk mid-majors. College sports are a money grabbing joke and will only get worse.

With that said, how many days until conference tourneys start?

I would be interested in the theory behind that. The MWC has played well in OOC and seemed to navigate the NET world pretty well.

 

[Cusefan0307]

Just want to comment on a few things on the RPI.

a) The RPI is still about W/L's. You do need to win games. The gaming / punishing came as a result of who you played, I don't think the biggest weakness was losing to great teams, but better to lose to #5 than #50 obviously.
b) The RPI punishes/destroys you when you play sub 280/sub 300 teams, especially more than a few. That is because 50% of the formula is the opponent's win%.
c) The weakness of the RPI, and where it has been gamed, is teams (and specifically leagues) instructed their teams to try to find sweet spot wins and avoid the punishers. More in the 100-200 category at home. All basically as winnable as sub 300 games (all 95%+ chance of winning), but would be rewarded better.

But I highlighted b) above because this is the natural hedge to the NET manipulation by the Big12. Its a simple solution. You want to stop teams doing what the Big12 has been doing this year, make the "NEW NET" = 50% current NET + 50% RPI.

I did a calculation maybe 10 days ago, where I found the average NET of Big12 teams to be 47, and the average RPI to be 67.

They are getting rewarded in one system, and getting punished in the other. Their true worth is probably somewhere in the middle of the two.

Just a minor quibble, but the RPI counted home games as .6 per win and road games 1.4 per win. And a win against a team with a bad RPI was worse than a loss against a team with a good RPI.

 

[djcon57]

the problem with the tourney and the results is that seeding creates so much reward for the top teams.. for many years often the first 2 games were pretty easy for the top teams. Now not so much

Imagine a seeding that was blind draw. would the top 5-10 teams get thru to the sweet 16 so often if they got a harder schedule? these days the diff between a 3 seed and a 10 seed is almost nothing more than the schedule they played not how well they played.

For sure...and that reflects when you see like 12 seeds as favorites. Will be interested if there's any value in the 1st round this year

 

[Cusefan0307]

the problem with the tourney and the results is that seeding creates so much reward for the top teams.. for many years often the first 2 games were pretty easy for the top teams. Now not so much

Imagine a seeding that was blind draw. would the top 5-10 teams get thru to the sweet 16 so often if they got a harder schedule? these days the diff between a 3 seed and a 10 seed is almost nothing more than the schedule they played not how well they played.

The seeding when RPI used was horrible. If you look at KenPom some of those upsets weren't really upsets, but moronic seeding based on a metric that had nothing to do with what happened on the court.

 

[ScrabbCuse]

Imagine a seeding that was blind draw. would the top 5-10 teams get thru to the sweet 16 so often if they got a harder schedule? these days the diff between a 3 seed and a 10 seed is almost nothing more than the schedule they played not how well they played.

Sounds like what tennis does. Seed the top 16 and randomly place the other 48 teams. It would be chaos but possibly more enjoyable than what they come up with now.

 

[pearl31]

make the "NEW NET" = 50% current NET + 50% RPI.

yeah I'm starting to think this might be the most viable common sense approach available, or at least something to the effect of a rpi/net convergence

 

[Orangezoo]

yeah I'm starting to think this might be the most viable common sense approach available, or at least something to the effect of a rpi/net convergence

A simple math exercise would put us in the 60s going this route which honestly is a very fair spot to be and in the bubble conversation. We scheduled well with some good wins but have struggled our fair share too. Using this approach you would see where going 4-0 heading into the tourney would have us in many more brackets.

The other side of this is to go with a larger ncaa tourney and NIT and allow first round losers to move to the NIT bracket.. you then ultimately work towards 1/3 of all teams playing in the post season more aligned with other sports even if still a bit less.

 

[HRE Otto IV]

Just want to comment on a few things on the RPI.

a) The RPI is still about W/L's. You do need to win games. The gaming / punishing came as a result of who you played, I don't think the biggest weakness was losing to great teams, but better to lose to #5 than #50 obviously.
b) The RPI punishes/destroys you when you play sub 280/sub 300 teams, especially more than a few. That is because 50% of the formula is the opponent's win%.
c) The weakness of the RPI, and where it has been gamed, is teams (and specifically leagues) instructed their teams to try to find sweet spot wins and avoid the punishers. More in the 100-200 category at home. All basically as winnable as sub 300 games (all 95%+ chance of winning), but would be rewarded better.

But I highlighted b) above because this is the natural hedge to the NET manipulation by the Big12. Its a simple solution. You want to stop teams doing what the Big12 has been doing this year, make the "NEW NET" = 50% current NET + 50% RPI.

I did a calculation maybe 10 days ago, where I found the average NET of Big12 teams to be 47, and the average RPI to be 67.

They are getting rewarded in one system, and getting punished in the other. Their true worth is probably somewhere in the middle of the two.

But is C really "gaming the system"? Playing a decent mid major vs a complete crap mid major should be a good thing. There should be incentive to do so. For the top teams it doesn't really matter. But for the bubble teams, they could actually lose those games.

Also wasn't there an original RPI that seemed more accurate and an RPI 2.0 that was crap (but still better than the NET)?

Some people will say the NET isn't a big deal that it isn't really used, but it is used. Whenever people bring up record vs the Top 50 how do they think the Top 50 was composed in the first place? Yes they might not look at your actual ranking but it is used when they look at who you played, without actually looking at who you played.

It really shouldn't be that hard. On March 11th come up with your locks and your under considerations. Compile each teams records against only those teams. Then add in any results from Championship week. That makes it easy to see who you played, who you beat, and how many bad Ls (outside this group) that you had. You don't need an RPI or a NET.

Just doing a quick look I would say that there are about 80 teams right now under consideration (including auto bids). We still have 17 days left in the season so it is possible a few more can be added to that list. You don't need a NET to separate those last 25 or so teams.

 

[CU44SE]

I'm no math/stat mind but it just seems like common sense:
it ought to be as simple as who you beat, regardless of score, and where (i.e. road wins should weigh more than neutral wins which should weigh more than home wins)... you shouldn't get credit for close losses nor should you get (more) credit for blowout wins...
like, who tf cares about style and efficiency? that should have no part in the formula. again, who you beat and where. are we really rewarding teams losing games efficiently?
am I oversimplifying?

one major issue. You need to absolutely consider strength of schedule and victory. If you win 24 games and play cupcakes and a down confrence and I win 18 but played the hardest schedule in the country. You use numbers to quantify that.

Bottom line and this is very generic.

Road win 1 point
Neutral win .75 ponts
Home win 0.5 points.

Now wins are wins. But you should deffinetly weigh the victory or loss based on their schedule and who they play.

I don’t like the average margin of victory into this algorithm though. It doesn’t account for bad games. It doesn’t account for when you’re down 8 with 6 mins left and everything just falls apart and you lose by 20. Not every 20 point victory/loss is the same. Against Tennessee this year I think it was like a 2 point game or 4 point game with 3 minutes left. We ended up losing by 15 I think. Computers see that as a 15 point loss. We were within 1 possession under 4 timeout. But numbers don’t see that as a close game.

Bottom line. There is a lot of good to be used from numbers. But margin of victory should never be utilized to weigh scores or results. This will always skew the results because numbers only look at the end result. If you are going to utilize numbers for margin of victory and so on. The bare minimum would be to utilize the mean or standard deviation of the score for every minute of the game. That way you could see how tightly contested the game was throughout and not just the end result.

 

[upperdeck]

one major issue. You need to absolutely consider strength of schedule and victory. If you win 24 games and play cupcakes and a down confrence and I win 18 but played the hardest schedule in the country. You use numbers to quantify that.

Bottom line and this is very generic.

Road win 1 point
Neutral win .75 ponts
Home win 0.5 points.

Now wins are wins. But you should deffinetly weigh the victory or loss based on their schedule and who they play.

I don’t like the average margin of victory into this algorithm though. It doesn’t account for bad games. It doesn’t account for when you’re down 8 with 6 mins left and everything just falls apart and you lose by 20. Not every 20 point victory is the same and not every 20 point victory is the same. Tennessee this year I think it was like a 2 point game or 4 point game with 3 minutes left. We ended up losing by 15 I think. Computers see that as a 15 point loss. We were within 1 possession under 4 timeout. But numbers don’t see that as a close game.

Bottom line. There is a lot of good to be used from numbers. But margin of victory should never be utilized to weigh scores or results. This will always skew the results because numbers only look at the end result. If you are going to utilize numbers for margin of victory and so on. The bare minimum would be to utilize the mean or standard deviation of the score for every minute of the game. That way you could see how tightly contested the game was throughout and not just the end result.

The numbers could see it though.. They have the data to use much more than we do.

A team is down 8 with 1 min to go.. Some teams press it out, foul and make a few 3s and lose by 4, some teams press it out, foul miss a few 3s and lose by16, some teams let it finish and lose by 8,

all three games are the same game unless a team actually comes back and wins, but how does the computer deal with it

Some teams throw the walkons in on a 15 pt game who knows where it goes. lose by 7 or 25.

you could in theory add some logic to see where the game was at every TO break. game flow does matter.

You could be in a close game with 45 secs left, foul go down 6, miss a 3 foul again go down 8, and a walkon throws in a 30fter at the buffer and you lose by 11..

you lose by 20 in fb and you more often than not were not in the game for a good chunk of it.. Bball is vastly different you can lose 5-10 pts in 30 secs all the time. commit a foul on a 3 get a T throw a ball away and lose 6-7 pts in 10 secs.

If a team goes 6-7 deep and plays out the string they will probably not win as many blowouts as a team that goes 8-10 deep. It doesn't mean they didn't have the game under control they just don't have the depth when they take the starters out in a game they have won.

 

[UpstateSM]

There's quite a few mid majors doing well for themselves in the NET. The Mountain West specifically is doing much better than the ACC.

A Bonnies Twitter account did some really good research on this topic last year. I will link it if I find it tonight. Just saying because there is data that backs up my opinion.

Mountain West teams are so smart in their noncon scheduling. Three Man Weave spent some time in a recent pod talking about them. I am really interested to see how they do this march. I'm also just sick of seeing ACC fans whine on Twitter. The league stinks.

 

[two3zone]

 

[M2MPrinciples]

And Wazzu *probably* only scheduled poorly because they didn't anticipate being as good as they've been. Lord knows they aren't getting help from their league. Kyle Smith is going to be a hot name come carousel season, deservedly or not.

 

[pearl31]

And Wazzu *probably* only scheduled poorly because they didn't anticipate being as good as they've been. Lord knows they aren't getting help from their league. Kyle Smith is going to be a hot name come carousel season, deservedly or not.

yeah geez they won at Arizona last night

 

[Fly Rodder]

The best teams will identify themselves IN THE TOURNAMENT ITSELF. Buy winning and advancing.

It's still a single elimination tournament. Poop happens. If that was the best way to crown a champion the pro sports would do that instead of 5 and 7 game series.

 

[Cusefan0307]

A Bonnies Twitter account did some really good research on this topic last year. I will link it if I find it tonight. Just saying because there is data that backs up my opinion.

Mountain West teams are so smart in their noncon scheduling. Three Man Weave spent some time in a recent pod talking about them. I am really interested to see how they do this march. I'm also just sick of seeing ACC fans whine on Twitter. The league stinks.

What constitutes stinking can be debated.

Obviously UNC with Duke just behind them are the class of the league. Happy we got to the heels this year.

Clemson beat a projected 3, 7, 8, and 10 seed in the OOC and they are a game over .500 in conference.

Florida is a top 25 team and easily in. They are because they have had a good SEC season. They beat no one OOC outside of Pitt. UVA beat Florida and went 9-2 OOC. Wake beat Florida and went 8-3 OOC.

Pitt had a terrible schedule. They probably thought WVU would be good, but then Huggins got fired and that win meant nothing. They lost to Florida and played a bunch of directional schools.

No need to get into us.

FSU and NC state both sucked early and then suddenly got competitive during the conference season. It would have been better for the top 6-8 teams had they continued to suck because they are dragging down other schools when they win.

Virginia Tech is a strange team. They’ve been bad on the road in conference. They did beat the number 6 team in the country in December and also beat Boise who‘s also in. Like us they’ve been hurt by some terrible scoring margins.

Miami completely collapsing has hurt the league. BC played a terrible schedule and is dragging down the conference when they play anyone including us.

Georgia Tech and Notre Dane are rebuilding, but competitive. That drags down league metrics.

Louisville is borderline the ACC version of DePaul. They kill the leagues metrics too.

If you look at the conference as a whole the middle to top teams all have some really good wins in November and December or at least took care of business, but the bottom teams kill the league on the whole compared to the Big 12 or SEC. The league on the whole can’t afford teams at the bottom being this bad going forward.

 

[UpstateSM]

What constitutes stinking can be debated.

Obviously UNC with Duke just behind them are the class of the league. Happy we got to the heels this year.

Clemson beat a projected 3, 7, 8, and 10 seed in the OOC and they are a game over .500 in conference.

Florida is a top 25 team and easily in. They are because they have had a good SEC season. They beat no one OOC outside of Pitt. UVA beat Florida and went 9-2 OOC. Wake beat Florida and went 8-3 OOC.

Pitt had a terrible schedule. They probably thought WVU would be good, but then Huggins got fired and that win meant nothing. They lost to Florida and played a bunch of directional schools.

No need to get into us.

FSU and NC state both sucked early and then suddenly got competitive during the conference season. It would have been better for the top 6-8 teams had they continued to suck because they are dragging down other schools when they win.

Virginia Tech is a strange team. They’ve been bad on the road in conference. They did beat the number 6 team in the country in December and also beat Boise who‘s also in. Like us they’ve been hurt by some terrible scoring margins.

Miami completely collapsing has hurt the league. BC played a terrible schedule and is dragging down the conference when they play anyone including us.

Georgia Tech and Notre Dane are rebuilding, but competitive. That drags down league metrics.

Louisville is borderline the ACC version of DePaul. They kill the leagues metrics too.

If you look at the conference as a whole the middle to top teams all have some really good wins in November and December or at least took care of business, but the bottom teams kill the league on the whole compared to the Big 12 or SEC. The league on the whole can’t afford teams at the bottom being this bad going forward.

It's not the ACC we signed up for and we deserve a chunk of the blame for that.

I also hate it with a passion and will never say anything nice about the league regardless of the logic and facts presented to me.

 

[pokercuse08]

It seems like the way to game NET is to play teams ranked 100-135 on the road and blow them out. You grab a Q2 win and run up the margin. Then just load up bottom of the basket Q4 home games and run them out of the gym.

Like beating #135 on the road is Q2 and so is beating #31 at home. That's wild. Meanwhile, if you play team 240 or lower, it doesn't matter if its home or away.

Of course, you also need the opponents you schedule to stay healthy and finish where they're supposed to, so trying to pick the 100-135th teams is going to involve a lot of luck. Scheduling cupcakes at home and smoking them is the way to go.

Then given the way it works, each game is sort of zero-sum because efficiency is so key. So having all of your conference opponents juice up their NET ranking is key. Then you lock it in and nobody can take it from your conference. You especially want the bottom of your conference to have a juiced up rating, because upsets happen and you don't want them messing up everyone's score. Also when teams get better as the year goes on, they can only take from conference opponents and it weakens the whole conference.

It's like the non-conference schedule is a game of hungry hungry hippos, and whichever conferences get the most marbles are basically locked into having good NET rankings. Then each conference resets its own game of hungry hungry hippos, with schools fighting over the marbles their conference won in the non-conference.

 

[Red2003]

It seems like the way to game NET is to play teams ranked 100-135 on the road and blow them out. You grab a Q2 win and run up the margin. Then just load up bottom of the basket Q4 home games and run them out of the gym.

Like beating #135 on the road is Q2 and so is beating #31 at home. That's wild. Meanwhile, if you play team 240 or lower, it doesn't matter if its home or away.

Of course, you also need the opponents you schedule to stay healthy and finish where they're supposed to, so trying to pick the 100-135th teams is going to involve a lot of luck. Scheduling cupcakes at home and smoking them is the way to go.

Then given the way it works, each game is sort of zero-sum because efficiency is so key. So having all of your conference opponents juice up their NET ranking is key. Then you lock it in and nobody can take it from your conference. You especially want the bottom of your conference to have a juiced up rating, because upsets happen and you don't want them messing up everyone's score. Also when teams get better as the year goes on, they can only take from conference opponents and it weakens the whole conference.

It's like the non-conference schedule is a game of hungry hungry hippos, and whichever conferences get the most marbles are basically locked into having good NET rankings. Then each conference resets its own game of hungry hungry hippos, with schools fighting over the marbles their conference won in the non-conference.

This is exactly right. I said before there should be a Q5. Q5 would be a 200+ net team at home and it just doesn’t count. Teams would still schedule 2-3 or of these but not 6-8 of them.

 

[orangemass]

Derelict and Manipulate my balls. Zoolander

 

[CardiacCuse]

It seems like the way to game NET is to play teams ranked 100-135 on the road and blow them out. You grab a Q2 win and run up the margin. Then just load up bottom of the basket Q4 home games and run them out of the gym.

Like beating #135 on the road is Q2 and so is beating #31 at home. That's wild. Meanwhile, if you play team 240 or lower, it doesn't matter if its home or away.

Of course, you also need the opponents you schedule to stay healthy and finish where they're supposed to, so trying to pick the 100-135th teams is going to involve a lot of luck. Scheduling cupcakes at home and smoking them is the way to go.

Then given the way it works, each game is sort of zero-sum because efficiency is so key. So having all of your conference opponents juice up their NET ranking is key. Then you lock it in and nobody can take it from your conference. You especially want the bottom of your conference to have a juiced up rating, because upsets happen and you don't want them messing up everyone's score. Also when teams get better as the year goes on, they can only take from conference opponents and it weakens the whole conference.

It's like the non-conference schedule is a game of hungry hungry hippos, and whichever conferences get the most marbles are basically locked into having good NET rankings. Then each conference resets its own game of hungry hungry hippos, with schools fighting over the marbles their conference won in the non-conference.

yes.

its ridiculous.

 

[FireballPhil]

Money to be made picking against them in the tourney then

That said, I'm not picking against Houston and their D

That all depends on the refs. They get the wrong crew and they will all foul out. There are at least 3 fouls every possession.