NCAA Net Ratings - 2019-2020

[GoMuskies]

Xavier is #71.

https://www.ncaa.com/rankings/basketball-men/d1/ncaa-mens-basketball-net-rankings?utm_campaign=trending

[noteggs]

First rankings came out today. Xavier comes in at 71.

Edit: looks like GO had same idea. So if someone can delete this thread. Thanks

[GoMuskies]

Big East:

Butler #4
Villanova #24
Georgetown #38
DePaul #43
Marquette #53
Seton Hall #55
Creighton #56
St. John's #65
Xavier
Providence #146

[Lloyd Braun]

This can’t be accurate. If it is the formula needs fixing.

[drudy23]

It doesn't seem like we should be that low, but it also doesn't seem like we should be very high.

The metrics back up the "feeling" most people have about this team. And it ain't great.

The Xavier way has proven me wrong before - I always tell myself that Xavier almost always finds a way. Hoping that's the case this year as well.

[BigMoeMusketeer]

This can’t be accurate. If it is the formula needs fixing.

Our non-conference schedule has been bad, and we don't have a single Top 50 win. Plenty of time to fix this at TCU and in Conference play, but not surprising at all that we are outside of the Top 68. What about our first 11 games has made you say "Yeah!"? A 2OT win over UCONN? Meh.

[Juice]

Mario is currently tweeting that the data is wrong.

[waggy]

Isn't the NET secret voodoo though?

[AviatorX]

Also, worth keeping in mind that Xavier's own NET ranking means very, very little. It's a ranking used to measure strength of opponent.

[waggy]

Also, worth keeping in mind that Xavier's own NET ranking means very, very little. It's a ranking used to measure strength of opponent.

Uh...

https://www.ncaa.com/video/basketball-men/2019-12-16/dan-gavitt-explains-college-basketballs-net-rankings

[D-West & PO-Z]

Mario is currently tweeting that the data is wrong.

Yeah I saw that. Wonder what the deal is?

[xubrew]

Yeah I saw that. Wonder what the deal is?

The rankings are right, but it is sorted wrong on the team sheets. Again.

[xubrew]

This can’t be accurate. If it is the formula needs fixing.

It looks about right to me.

Having said that, it's still early. If IUPUI jumps out to a 5-2 lead on Michigan with 18:30 left until halftime, the scoreboard isn't wrong. It would be silly to look at that and start tweeting and posting and shouting about how some major upset is about about to go down, but in that moment IUPUI is ahead. The NET is still kind of like that. It's right, but it's right based on a very limited amount of data compared to what it is going to be at the end. So, I wouldn't rely too much on it right now, nor would I conclude that the formulas are off. It's just early.

[xubrew]

Another for instance...

Wake Forest has been our only true road game so far. We are 0-1 on the road. That's the same as being 0-20 on the road. Had we won that game, our NET would probably be in the 20s or 30s, because on the flip side it would be the same as being 20-0 on the road.

[XU 87]

Duquesne is celebrating that they're ranked 20th and UK is a lowly 58.

[BigMoeMusketeer]

UK is a lowly 58.

Outside of Michigan State, their non-conference schedule is AWFUL...although it does get better by a lot between now and conference play.

That Evansville game, deservedly so, really stings their resume'.

[Lloyd Braun]

Another for instance...

Wake Forest has been our only true road game so far. We are 0-1 on the road. That's the same as being 0-20 on the road. Had we won that game, our NET would probably be in the 20s or 30s, because on the flip side it would be the same as being 20-0 on the road.

So to me, that means the formula is faulty that it is based on win %. Doesn’t reward good scheduling.

[OTRMUSKIE]

Well then it accurate again bc wake sucks. But I know what you’re saying.

[GoMuskies]

We're sneaking up on people in 2020!

[xubrew]

So to me, that means the formula is faulty that it is based on win %. Doesn’t reward good scheduling.

I think the teams we’ve played are a large part of why we’re ranked where we are. We have not won a road game, we haven't won a home game against anyone with a road record, and the two neutral wins came against Towson who has an overall losing record, and UConn who is decent at best. When I look at that, and think of the concept of an ELO based ranking system (which the NET isn't totally ELO, but it is partly that) I think there are about 70ish teams that have done more against their schedule than we have. So, in our case, based strictly on our merit, I don't think it's that far off.

Beating Wake Forest would have made a huge difference. It would have been a road win against a team with a winning record. Right now we don't have that.

[Lloyd Braun]

I wasn’t referring to our schedule in particular, just the formula in general. If we are using Xavier as an example- that a win over Wake (not a good team) on the road would bolster the rating by 50 spots there is some wild variance in how these numbers are calculated. Which IMO is a flawed formula.

Edit: also I should specify my initial post of “this can’t be accurate” was not meant to imply Xavier should be higher.

[xubrew]

I wasn’t referring to our schedule in particular, just the formula in general. If we are using Xavier as an example- that a win over Wake (not a good team) on the road would bolster the rating by 50 spots there is some wild variance in how these numbers are calculated. Which IMO is a flawed formula.

Well, if you’re 0-for-1, you’re batting 000. If you’re 1-for-1, you’re batting 1000. The less games that have been played, the more erraticly the rankings can change because of one game.

[Xville]

I wasn’t referring to our schedule in particular, just the formula in general. If we are using Xavier as an example- that a win over Wake (not a good team) on the road would bolster the rating by 50 spots there is some wild variance in how these numbers are calculated. Which IMO is a flawed formula.

Edit: also I should specify my initial post of “this can’t be accurate” was not meant to imply Xavier should be higher.

I think right now it is flawed, but that's because there isn't enough data yet. By february or so, there will be enough data to where a win against wake wouldnt mean a 40-50 spot difference.

Honestly releasing the bet this early is stupid, but it gives people somethjng to talk about, like us on a messageboard:)

[OTRMUSKIE]

It use to be released in November so they have delayed it a whole
Month.

[xubrew]

It use to be released in November so they have delayed it a whole
Month.

They always released the official RPI the Monday after thanksgiving. That was always real out of whack early on as well, but no one completely flipped out over it because they understood it was just too early to be a reliable indicator. Last year when the NET came out the internet almost broke. So, they decided to wait. Now people are still flipping out.

I don’t get it. They publish batting averages and league standings and all that stuff all season long. People put it in the proper context. Not with the NET though. I’m in favor of publishing it after the first week even if it isn’t all that reliable. I mean, why not? But for some reason everyone loses their minds whenever they see the NET early on, which is actually why they pushed it back.

[BigMoeMusketeer]

Can we all agree 10-3 / 9-9 / 19-12, with this non-con schedule, isn't going to be good enough for an NCAA bid? So, it is pretty clear what needs to be done in their remaining 20 games, and winning on Saturday in Forth Worth would be a great start.

I'll admit, I was WAY wrong before the year, I thought this was a 25 win basketball team...they have played, to date, way below what I was expecting, given all that they returned ("the core four").

[Masterofreality]

71 in the NET IS LOWER than we finished last year at 64, and we weren’t happy with last year.
9-2 is a mirage folks. Travis Steele has a lot of work to do.

[GoMuskies]

Travis Steele has a lot of work to do.

My fear is that the most important work is in roster construction, which unfortunately takes time. The recruiting classes would suggest that this work is well underway, but we're probably two years away from really seeing the fruits of those efforts.

[bleedXblue]

71 in the NET IS LOWER than we finished last year at 64, and we weren’t happy with last year.
9-2 is a mirage folks. Travis Steele has a lot of work to do.

Exactly. I'd give him about a C- right now. Bench development has been Freemantle and that's about it. No one else has emerged as a super solid guy off the bench. I was buying into the renewed defensive focus, but that also needs some work to be much more consistent. Offensively, we look like 5 guys whi have never played together before. If not for Naji bailing us out of 3 games, it would be really ugly right now.

[bleedXblue]

My fear is that the most important work is in roster construction, which unfortunately takes time. The recruiting classes would suggest that this work is well underway, but we're probably two years away from really seeing the fruits of those efforts.

He's got enough talent on this team that he should get us into the tourney. I get what you're saying.......but not allowing him a pass. He had that last year and should have.

Crap, look at Butler. We have more talent than those guys........

[Xavgrad08]

From a scheduling prospective I am not sure the Wake series is good for Xavier. Most years Wake is going to lose a lot of games in the Acc. Not a good win in most years, but could be a bad loss.

[OTRMUSKIE]

How much longer do we have to play wake? It was a ten game contract I believe and they switched it to every other year.

[D-West & PO-Z]

Can we all agree 10-3 / 9-9 / 19-12, with this non-con schedule, isn't going to be good enough for an NCAA bid? So, it is pretty clear what needs to be done in their remaining 20 games, and winning on Saturday in Forth Worth would be a great start.

I'll admit, I was WAY wrong before the year, I thought this was a 25 win basketball team...they have played, to date, way below what I was expecting, given all that they returned ("the core four").

I hope we perform better than that, not sure we will by much, we will see, but I would not say definitively that wouldn't get us in. Would depend on a lot.

[xukeith]

They always released the official RPI the Monday after thanksgiving. That was always real out of whack early on as well, but no one completely flipped out over it because they understood it was just too early to be a reliable indicator. Last year when the NET came out the internet almost broke. So, they decided to wait. Now people are still flipping out.

I don’t get it. They publish batting averages and league standings and all that stuff all season long. People put it in the proper context. Not with the NET though. I’m in favor of publishing it after the first week even if it isn’t all that reliable. I mean, why not? But for some reason everyone loses their minds whenever they see the NET early on, which is actually why they pushed it back.

I agree. Let everyone in the world access the rating and formulas. Be transparent. What are you afraid of?

[xukeith]

[QUOTE=xukeith;659884]I agree. Let everyone in the world access the rating and formulas. Be transparent. What are you afraid of?[/QUOT
Good news! X moved up one spot to 70 after yesterday's news.
http://warrennolan.com/basketball/2020/net

[scoscox]

From a scheduling prospective I am not sure the Wake series is good for Xavier. Most years Wake is going to lose a lot of games in the Acc. Not a good win in most years, but could be a bad loss.

it's horrible for us

[xukeith]

If i remember correctly road wins gain most points and home losses are the worst.

[GoMuskies]

And we've neither lost at home nor won on the road. So we're middling. Yay, middling!

[X-band '01]

How much longer do we have to play wake? It was a ten game contract I believe and they switched it to every other year.

They do a home-and-away followed by 2 off years in perpetuity now.

[OTRMUSKIE]

They do a home-and-away followed by 2 off years in perpetuity now.

Forever? Please make it stop!!!

[xcellentx]

They always released the official RPI the Monday after thanksgiving. That was always real out of whack early on as well, but no one completely flipped out over it because they understood it was just too early to be a reliable indicator. Last year when the NET came out the internet almost broke. So, they decided to wait. Now people are still flipping out.

I don’t get it. They publish batting averages and league standings and all that stuff all season long. People put it in the proper context. Not with the NET though. I’m in favor of publishing it after the first week even if it isn’t all that reliable. I mean, why not? But for some reason everyone loses their minds whenever they see the NET early on, which is actually why they pushed it back.

Most people don't freak at the official RPI and the NET because people could already calculate the RPI because the formula was known. I agree with you on just having it go from the start. It does seem like waiting to reveal it is causing people to take it the wrong way. The NET really does mean nothing at this point and it would be a surprise if most teams are even within 10-15 spots of this on the final rankings.

[xcellentx]

71 in the NET IS LOWER than we finished last year at 64, and we weren’t happy with last year.
9-2 is a mirage folks. Travis Steele has a lot of work to do.

To be fair, you can't compare a whole seasons worth of data to 11 games into a season. I agree that there is work to do but that doesn't also mean that the 9-2 record means nothing.

[xukeith]

To be fair, you can't compare a whole seasons worth of data to 11 games into a season. I agree that there is work to do but that doesn't also mean that the 9-2 record means nothing.
If X 's 9-2 record is reflective and fitting for a team, do we all expect Dayton, SDST, and Duquesne to stay in the top 30 ?

[GoMuskies]

I expect dayton to stay in the 20 top if not the top 10. Not sure about SD State. Certainly not on Duquesne.

[Xville]

Yep...unfortunate but Dayton is really really good. The a10 is decent this year.

[D-West & PO-Z]

Yep...unfortunate but Dayton is really really good. The a10 is decent this year.

Yeah there are 7 A10 teams in the top 71 of the NET rankings.

[xcellentx]

If X 's 9-2 record is reflective and fitting for a team, do we all expect Dayton, SDST, and Duquesne to stay in the top 30 ?

I think for showing how you shouldn't take too much stock in these ratings yet, Duquesne and Stanford are good examples. KenPom has their NCSOS's around 320. Maybe they are that good but if that bad of a schedule can get you top 10 then hold off on taking too much stock in the specific number.

Edit: Add Liberty to that mix with their 344th ranked KenPom NCSOS

[UCGRAD4X]

[QUOTE=xukeith;659884]I agree. Let everyone in the world access the rating and formulas. Be transparent. What are you afraid of?[/QUOT
Good news! X moved up one spot to 70 after yesterday's news.
http://warrennolan.com/basketball/2020/net

As Travis said in his presser, "The NCAA has to be transparent....that's clear."

[noteggs]

For what it’s worth, X moved up 9 spots to 62.

[GoMuskies]

Sneaking up!

[waggy]

X up to 51.

[GoMuskies]

San Diego State is #1! Liberty up to #17. Duquesne fades to #66.

[xukeith]

San Diego State is #1! Liberty up to #17. Duquesne fades to #66.

http://warrennolan.com/basketball/2020/conference/Big-East

#1 conference by NET ranking.

[xcellentx]

San Diego State is #1! Liberty up to #17. Duquesne fades to #66.

If you want to see why you shouldn't overreact to our opening and current rank, Duquesne fell 43 spots with just one loss.

[Xuperman]

http://warrennolan.com/basketball/2020/conference/Big-East
#1 conference by NET ranking.

Are there any X diehards that can really appreciate THIS TEAM.....Love these guys, ESPECIALLY THE SENIORS!!! GO X full and out!
:shield::shield::shield::shield:

[muskiefan82]

Are there any X diehards that can really appreciate THIS TEAM.....Love these guys, ESPECIALLY THE SENIORS!!! GO X full and out!
:shield::shield::shield::shield:

Marshall 30 points. Next game Scruggs 30 points. Next game Goodin 25 points. Next game Jones 18 and 14. Next game, 12/30, @ Villanova - do we get that effort from all 4 on the same night? My God, I hope so.

[UCGRAD4X]

Marshall 30 points. Next game Scruggs 30 points. Next game Goodin 25 points. Next game Jones 18 and 14. Next game, 12/30, @ Villanova - do we get that effort from all 4 on the same night? My God, I hope so.

Even if two of the four "go off" and the others play more toward their average and just keep from making STUPID plays...and play good defense (obviously) we come away from Philly with a W.

But I will take all four having career games.

[paulxu]

Only 3 undefeated teams left. Auburn, San Diego St and Liberty.

[whopper]

SD State beat us last year after we were up by 19 in Hawaii and it caused a lot of dissension. Then they lost their leading scorer McDaniels who was drafted and now are undefeated? May not been as bad a loss as thought at the time. Also lost to Auburn in OT(with Welage! having 17) and they made finals. What does it show? I don't know except that each game is a stand alone event these days.

[AviatorX]

SD State beat us last year after we were up by 19 in Hawaii and it caused a lot of dissension. Then they lost their leading scorer McDaniels who was drafted and now are undefeated? May not been as bad a loss as thought at the time. Also lost to Auburn in OT(with Welage! having 17) and they made finals. What does it show? I don't know except that each game is a stand alone event these days.

SDSU sucked last year. That was a quad 3 loss if memory serves, something that should literally never happen to Xavier under any circumstance.

[noteggs]

Sure some saw this from John Fanta. Makes you wonder about this whole NET stuff.

https://twitter.com/john_fanta/status/1232863499916890112?s=12

[xcellentx]

Sure some saw this from John Fanta. Makes you wonder about this whole NET stuff.

https://twitter.com/john_fanta/status/1232863499916890112?s=12

That is a weird one. There do seem to be a lot of weird things with the NET this year. It doesn't seem to penalize you as much as you would think for either playing a lot of Q3-4 games or losing one or 2. I think a few tings might come in to play here. Creighton played a non-D1 game so they don't get any benefit from that. Duke has only 2 Q4 games compared to Creighton's 6. Duke's KU win is probably helping them as well. But Duke's losses are far worse than anything Creighton has.

The better question is probably how is Arizona ahead of Creighton...

[xcellentx]

It is also interesting if you look at TTU and Houston who are about 20 spots higher than we are and compare our NET resume to theirs

[noteggs]

That is a weird one. There do seem to be a lot of weird things with the NET this year. It doesn't seem to penalize you as much as you would think for either playing a lot of Q3-4 games or losing one or 2. I think a few tings might come in to play here. Creighton played a non-D1 game so they don't get any benefit from that. Duke has only 2 Q4 games compared to Creighton's 6. Duke's KU win is probably helping them as well. But Duke's losses are far worse than anything Creighton has.

The better question is probably how is Arizona ahead of Creighton...

Yes it’s interesting Duke had played 6 more Q 3/4 games (with 5 more wins) than X. Also, they only have 1 more Q1. However, we are only 3-9 vs Q1 vs Duke’s 4-2. X is doing better with Q2 7-1 vs 6-2. Still trying figure out the Duke 6 ranking vs X’s 41. Hmm

Before people start to say X is probably ranked correctly- I agree.

[GoMuskies]

Unlike the absurd RPI, NET does allow for margin of victory to be a factor. So that probably affects this discussion.

[noteggs]

Unlike the absurd RPI, NET does allow for margin of victory to be a factor. So that probably affects this discussion.

Which I understand, but end of game dynamics is not a consideration. IMO a ten point lead with 1 minute is fool’s gold on which way the efficiency ranking can go. Example, at the end of the DePaul game away, refs just let them beat the shit out of X on their full court press at the end of game which dropped a 17 point Q1 game to the final score.

And yes, current system beats the RPI.

[GoMuskies]

OK, but better teams win by a larger margin than lesser teams. A system that doesn't take that into account is worthless.

[noteggs]

Completely agree. Just trying to figure out why Duke losing to a Q2 by 10 kept them at 6th in the NET. If I remember correctly, X dropped three spots after losing to Villanova (Q1) with a similar score. You would think the the variation/change would be greater for a higher ranked team? Especially with the amount of Q3/4 games.

[GIMMFD]

Which I understand, but end of game dynamics is not a consideration. IMO a ten point lead with 1 minute is fool’s gold on which way the efficiency ranking can go. Example, at the end of the DePaul game away, refs just let them beat the shit out of X on their full court press at the end of game which dropped a 17 point Q1 game to the final score.

And yes, current system beats the RPI.

If I remember correctly, NET only takes into account up to 10 points, I believe anything higher than that counts the same as 10, the reason being to "keep sportsmanship" in the game or something of that sort.

[scoscox]

Completely agree. Just trying to figure out why Duke losing to a Q2 by 10 kept them at 6th in the NET. If I remember correctly, X dropped three spots after losing to Villanova (Q1) with a similar score. You would think the the variation/change would be greater for a higher ranked team? Especially with the amount of Q3/4 games.

https://twitter.com/John_Fanta/status/1232863499916890112

fanta asks the same question in a pretty entertaining way in this clip from shootaround. got a chuckle from me. start it at :45

[noteggs]

If I remember correctly, NET only takes into account up to 10 points, I believe anything higher than that counts the same as 10, the reason being to "keep sportsmanship" in the game or something of that sort.

Exactly. That’s why I used the 10 point as the example. However, teams are trying to win and not sure it really matters in conference play.

[xcellentx]

Completely agree. Just trying to figure out why Duke losing to a Q2 by 10 kept them at 6th in the NET. If I remember correctly, X dropped three spots after losing to Villanova (Q1) with a similar score. You would think the the variation/change would be greater for a higher ranked team? Especially with the amount of Q3/4 games.

To make it possibly more confusing, Arizona dropped from 7 to 11 after losing an away game to the 47th ranked team. Seems like a loss to a worse ranked team would have a similar effect but it didn't.

[xcellentx]

Liberty with a 3 spot jump after a win over the second to last place team in the NET

[D-West & PO-Z]

Can't these jumps also be explained by what the other teams you have played have done that night as well? What you do alone doesnt solely affect your net right? If Depaul loses and drops to a Quad 2 win (road game) for us wouldnt that affect our net?

[xcellentx]

Can't these jumps also be explained by what the other teams you have played have done that night as well? What you do alone doesnt solely affect your net right? If Depaul loses and drops to a Quad 2 win (road game) for us wouldnt that affect our net?

It is obviously more complicated than I'm presenting because obviously you have to take into account every game and how that effects other teams NET rankings. It is just kind of weird and funny that a team like Liberty beats the second worst team in the NET and jumps 3 spots and WSU who dropped 2 spots to be behind them did so after beating Temple.

What is crazy with Duke's NET Ranking is they were ranked 6 before losing to Clemson. The next day they actually moved up to 4th in the NET, fall back down to 6th before they play Louisville, and haven't moved from 6th since. The UL loss didn't effect it nor did the loss to NC State or Wake. Just weird.

[noteggs]

What is crazy with Duke's NET Ranking is they were ranked 6 before losing to Clemson. The next day they actually moved up to 4th in the NET, fall back down to 6th before they play Louisville, and haven't moved from 6th since. The UL loss didn't effect it nor did the loss to NC State or Wake. Just weird.

This is where I’m at. 22 point loss to NC ST and 12 point loss to Wake in the last two of three games and no movement for Duke. Sure it also has something to do with my dislike for Duke (man that 2004 game still chaps me).

Oh well at end of day, it’s only one metric that they can use and they’ll be playing the first two rounds in Greensboro NC as usual.

[whopper]

Yale beat Clemson 54-45 on Dec 22 at Clemson.. It just shows the level of parity these days. Never before in history has every game seemed like a clean slate before the game. Last week Yale lost to Cornell. These NET ratings probably work over a large sample size but don't reflect the attitude of today's player who literally lives the "never say die" attitude

[paulxu]

If I'm reading it right, we didn't play yesterday but dropped from 42 to 45.

[Xville]

If I'm reading it right, we didn't play yesterday but dropped from 42 to 45.

Yep. Pretty convinced this metric is completely stupid, which makes sense because didnt the ncaa itself develop it?

[Lloyd Braun]

At quick glance- There were 4 teams that jumped us and one team that fell beneath us:

Oklahoma (W at WVU) 49–>42
Providence (W at Nova) 50–>44
UNI (W at Drake) 43–>37
USC (W vs ASU) 45–>39

Arkansas (L at UGA) 41–>46


A win today would result in a jump to around 39 depending on other games.

[Xuperman]

At quick glance- There were 4 teams that jumped us and one team that fell beneath us:

Oklahoma (W at WVU) 49–>42
Providence (W at Nova) 50–>44
UNI (W at Drake) 43–>37
USC (W vs ASU) 45–>39

Arkansas (L at UGA) 41–>46


A win today would result in a jump to around 39 depending on other games.

Yea, it's quickly becoming tight amongst some P6 teams hovering at the .500 conference mark. It does us no good when the likes of Clemson and Texas/Oklahoma keep getting big wins. I mean Clemson has 3 top 10 wins.

Not to mention that the high possibility of some serious bid theft in the conference tourneys create some additional discomfort.

[GIMMFD]

Yep. Pretty convinced this metric is completely stupid, which makes sense because didnt the ncaa itself develop it?

Probably, which would make a shit ton of sense. I almost feel like RPI was a better metric at this point, confusing as hell.

[X-band '01]

Then again, TCU beating Baylor yesterday should have been a pleasant bonus to Xavier's metrics.

[noteggs]

Wish we did know more about the NET. All we hear from the expects is this Quad stuff and you can’t turn on game this time of year only to hear it like 6 times. My last rant on this metric..........for today.

A couple observations from yesterday’s games

UNI wins at Drake. Q3 win moves them from 43 to 37.

Michigan St wins Q1 on road against Maryland (would have been Q1 regardless) goes from 13 to 7.

Maryland loses Q1 and goes from 9 up to 15.

Seton Hall wins at Marquette by 9 (1 under the max 10). Q1 win but only goes to 13 from 15

Finally the Duke thing. 2 Q1 and 1 Q2 defeats in last 4 games. Of course stays at 6.

Makes perfect sense I guess...not so much.

[HenryMuto]

The NET rankings are a joke.

Duke still at 6th after losing 3 of their last 4 games. Got smoked by 22 vs NC State a bubble team. Lost to WF a bad team. Lost to Virginia an ok team.
Arizona 19-10 was ranked 11th before losing to UCLA and still 11th they move 0 spots after a loss have 9 losses and only 3 quad 1 wins yet ranked 11th.
BYU 10th because they beat Gonzaga ?
Kentucky only 14th I guess losing that Evansville game was like 3 losses because they only have 5 and it took winning 12 out of last 13 to move from like 24th to 14th.

[Lloyd Braun]

Slow start but it may be the only time this year we have more depth. Would hope to speed it up when they get tired. Rotate some bodies and feed Tyrique.

Still not crazy about this lineup however. Tandy boost eminent.

Edit Wrong thread

[xubrew]

Wish we did know more about the NET. All we hear from the expects is this Quad stuff and you can’t turn on game this time of year only to hear it like 6 times. My last rant on this metric..........for today.

A couple observations from yesterday’s games

UNI wins at Drake. Q3 win moves them from 43 to 37.

Michigan St wins Q1 on road against Maryland (would have been Q1 regardless) goes from 13 to 7.

Maryland loses Q1 and goes from 9 up to 15.

Seton Hall wins at Marquette by 9 (1 under the max 10). Q1 win but only goes to 13 from 15

Finally the Duke thing. 2 Q1 and 1 Q2 defeats in last 4 games. Of course stays at 6.

Makes perfect sense I guess...not so much.

-Drake was 12-1 at home going into yesterday, so UNI got a little more credit for that than what it may appear on the surface

-Maryland and Michigan State don't seem TOO out of whack, at least in regards to how much they moved.

-Duke is interesting. I THINK I know why they haven't moved down. They're sort of gaming the system without even realizing it. Their offensive and defensive efficiency numbers are really good, all their recent losses have been on the road so that gets adjusted down, the scoring margin isn't taking that much of a hit. it's +177. Remember that it's capped at 10, so if you win by 10 you get 10, if you win by 50 you get 10, if you lose by 50 you lose 10, and if a game goes into overtime you either get 1 or lose 1. +177 is pretty damn good. For instance, Xavier's is +65 (I think). I don't know where that ranks, but it's probably pretty far behind Duke. The one area where Duke is probably struggling is in the TVI, but the NET factors each of the five areas differently for each team. If your rankings in the five areas are 4, 7, 2, 29, and 5, they're going to devalue the area where you're 29th because it's an outlier from the others. So if Duke's TVI is dropping (and I can't imagine that it's not) because they're losing to teams that aren't inside the bubble, it's still just one variable out of 5, and if it's way lower than the other 4 it isn't even counting as much. So....

But, just like the RPI, all the NET really is is a sorting tool. It isn't that important in and of itself. In fact had the committee used the RPI last year instead of the NET, I think the field would have looked pretty much exactly the same.

[stammina0721]

The NET rankings are a joke.

Duke still at 6th after losing 3 of their last 4 games. Got smoked by 22 vs NC State a bubble team. Lost to WF a bad team. Lost to Virginia an ok team.
Arizona 19-10 was ranked 11th before losing to UCLA and still 11th they move 0 spots after a loss have 9 losses and only 3 quad 1 wins yet ranked 11th.
BYU 10th because they beat Gonzaga ?
Kentucky only 14th I guess losing that Evansville game was like 3 losses because they only have 5 and it took winning 12 out of last 13 to move from like 24th to 14th.

There is some validity to this. It seems the NET rankings are pretty much set at the beginning of February and three or 4 game losing streaks can't change that much. There is definitely a flaw in the formula

[paulxu]

Fun with the NET:

Team A: NET 35
Record: 15-14
SOS: 45
NC SOS: 63
Quad 1& 2 = 8-13
Quad 3&4 = 7-1
RPI: 90

Team B: NET 43
Record: 19-10
SOS: 11
NC SOS: 52
Quad 1&2: 10-10
Quad 3&4: 9-0
RPI: 28

[xubrew]

Fun with the NET:

Team A: NET 35
Record: 15-14
SOS: 45
NC SOS: 63
Quad 1& 2 = 8-13
Quad 3&4 = 7-1
RPI: 90

Team B: NET 43
Record: 19-10
SOS: 11
NC SOS: 52
Quad 1&2: 10-10
Quad 3&4: 9-0
RPI: 28

Team A I know without looking it up because they're the only team with that pitiful of a record that's NET is in the 30s. Team B, I'd have to look up.

[AviatorX]

Fun with the NET:

Team A: NET 35
Record: 15-14
SOS: 45
NC SOS: 63
Quad 1& 2 = 8-13
Quad 3&4 = 7-1
RPI: 90

Team B: NET 43
Record: 19-10
SOS: 11
NC SOS: 52
Quad 1&2: 10-10
Quad 3&4: 9-0
RPI: 28

Guessing A is Purdue - they have absolutely destroyed some highly ranked teams which has inflated their computer numbers (or caused their computer numbers to reflect exactly how good they are, if you prefer). Very weird resume.

[Xville]

When Duke doesnt fall after losing 3 of 4 and arizona is in the top 10, something is seriously broken with the metric.

[xcellentx]

-Drake was 12-1 at home going into yesterday, so UNI got a little more credit for that than what it may appear on the surface

-Maryland and Michigan State don't seem TOO out of whack, at least in regards to how much they moved.

-Duke is interesting. I THINK I know why they haven't moved down. They're sort of gaming the system without even realizing it. Their offensive and defensive efficiency numbers are really good, all their recent losses have been on the road so that gets adjusted down, the scoring margin isn't taking that much of a hit. it's +177. Remember that it's capped at 10, so if you win by 10 you get 10, if you win by 50 you get 10, if you lose by 50 you lose 10, and if a game goes into overtime you either get 1 or lose 1. +177 is pretty damn good. For instance, Xavier's is +65 (I think). I don't know where that ranks, but it's probably pretty far behind Duke. The one area where Duke is probably struggling is in the TVI, but the NET factors each of the five areas differently for each team. If your rankings in the five areas are 4, 7, 2, 29, and 5, they're going to devalue the area where you're 29th because it's an outlier from the others. So if Duke's TVI is dropping (and I can't imagine that it's not) because they're losing to teams that aren't inside the bubble, it's still just one variable out of 5, and if it's way lower than the other 4 it isn't even counting as much. So....

But, just like the RPI, all the NET really is is a sorting tool. It isn't that important in and of itself. In fact had the committee used the RPI last year instead of the NET, I think the field would have looked pretty much exactly the same.

I think what you are saying about Duke makes sense, but I think a system like this that doesn't change at all when you lose 3-4 just doesn't seem right. They also didn't change after losing to Clemson and UL right in a row.

It does seem like parts of the NET can be gamed by playing a lot of Q3-4 games. Duke racked up +128 of its +117 scoring margin in 13 wins against Q3-4.

I know it is more of a grouping tool than actually saying this is the 6th best this is the 10th best team, but there does seem to be a lot of weirdness this year. I mean if you look at Duke and Seton Hall, SH has played almost double the Q1 games Duke has, has as many Q1 wins as Duke has Q1-2 and is 6-0 Q3-4 to Duke's 13-1. For a metric that is supposed to rank performance that just doesn't seem right.

[D-West & PO-Z]

Team B is X. Correct?

My question is it is nice for us to lump Q1&2 together, but is that how the committee does it. How many of the 8 wins are Q1 for team A?

[paulxu]

Team B is X. Correct?

My question is it is nice for us to lump Q1&2 together, but is that how the committee does it. How many of the 8 wins are Q1 for team A?

Team A has 4 Quad 1 wins.
Team B has 3 Quad 1 wins.

[D-West & PO-Z]

Yesterday was our 2nd best NET win of the season.

[GIMMFD]

Is Team A Clemson? If they have 4 Q1 wins, me thinking it's the Duke, Louisville, Florida State, etc. wins for Clemson, I know their record is close to .500 as well

[xubrew]

I think what you are saying about Duke makes sense, but I think a system like this that doesn't change at all when you lose 3-4 just doesn't seem right. They also didn't change after losing to Clemson and UL right in a row.

It does seem like parts of the NET can be gamed by playing a lot of Q3-4 games. Duke racked up +128 of its +117 scoring margin in 13 wins against Q3-4.

I know it is more of a grouping tool than actually saying this is the 6th best this is the 10th best team, but there does seem to be a lot of weirdness this year. I mean if you look at Duke and Seton Hall, SH has played almost double the Q1 games Duke has, has as many Q1 wins as Duke has Q1-2 and is 6-0 Q3-4 to Duke's 13-1. For a metric that is supposed to rank performance that just doesn't seem right.

I'm not a fan of the NET as a power ranking. I'm okay with it as a sorting tool so long as that's all that it is. I don't really have a preference. I had no real issues with the RPI as a sorting tool, but it was obviously flawed if you were trying to use it as a power ranking.

It's a process by committee. That's what I think most people don't fully understand. They debate, discuss, and assess all the teams all season long ad nauseum. All the NET really does is determine how the information is sorted on the team sheets. All of the info is going to be there no matter what sorting tool is used. A lot of bracketologists just look at the NET and count up the quad wins, and base their projections on that. You can do that and get it close enough for what they're doing, and that's forecasting the bracket, but that really isn't the process the committee goes through when they select and seed the teams. I'm sure the committee knows that Arizona is a crappy road team, and that they've only won one road game against a team with a winning record, so anyone who beats them at home isn't getting the same "quad 1 credit" as a team that wins at Seton Hall (like we did)

[SM#24]

Exactly, the NET and previously the RPI were just a way to sort the teams into buckets so as to show number games played within each bucket and the record against those teams. the committee uses all the information at their disposal including non-PC metrics like KenPom, Sagarin, etc. Each committee member has their own individual value that they place on all the information, whether on the team sheet or not. Teams are discussed and then it's just a series of votes, usually done in bucket of 4 teams. At least that's what I recall from the various media members that were invited to participate in a mock selection process.

[xcellentx]

I'm not a fan of the NET as a power ranking. I'm okay with it as a sorting tool so long as that's all that it is. I don't really have a preference. I had no real issues with the RPI as a sorting tool, but it was obviously flawed if you were trying to use it as a power ranking.

It's a process by committee. That's what I think most people don't fully understand. They debate, discuss, and assess all the teams all season long ad nauseum. All the NET really does is determine how the information is sorted on the team sheets. All of the info is going to be there no matter what sorting tool is used. A lot of bracketologists just look at the NET and count up the quad wins, and base their projections on that. You can do that and get it close enough for what they're doing, and that's forecasting the bracket, but that really isn't the process the committee goes through when they select and seed the teams. I'm sure the committee knows that Arizona is a crappy road team, and that they've only won one road game against a team with a winning record, so anyone who beats them at home isn't getting the same "quad 1 credit" as a team that wins at Seton Hall (like we did)

I agree with how it is used and I think it is overall better than the RPI. I'm sure if they are say evaluating between X with the win over SH and Virginia with the win over Duke for the last spot more than just the ranking between those two teams.

I guess the main complaint would be even though you use it as a bucket, you still want the buckets to be as correct as possible. In some ways I think my complaint is that the NET is a little too close to some of the Predictive metrics as opposed to being about resume and actual performance. I want to know who the best teams are, but I also want to know what the best resumes are. Duke playing great at home in 11 Q3-4 games seems to have vastly more weight than SH playing 15 Q1 games and winning 7 of those on the road.

Mainly it is fun to complain, and even more fun when it involves Duke.

[xcellentx]

Team B is X. Correct?

My question is it is nice for us to lump Q1&2 together, but is that how the committee does it. How many of the 8 wins are Q1 for team A?

I think the committee looks at it in a variety of ways, but essentially Q1-2 are good wins/losses and Q3-4 are bad wins/losses. I think the committee wants to see a certain amount of Q1-2 wins, but will obviously look more into how much better or worse each teams Q1-2 wins are. I think last year though, no team was left out with 11 or more Q1-2 wins.

[noteggs]

I agree with how it is used and I think it is overall better than the RPI. I'm sure if they are say evaluating between X with the win over SH and Virginia with the win over Duke for the last spot more than just the ranking between those two teams.

I guess the main complaint would be even though you use it as a bucket, you still want the buckets to be as correct as possible. In some ways I think my complaint is that the NET is a little too close to some of the Predictive metrics as opposed to being about resume and actual performance. I want to know who the best teams are, but I also want to know what the best resumes are. Duke playing great at home in 11 Q3-4 games seems to have vastly more weight than SH playing 15 Q1 games and winning 7 of those on the road.

Mainly it is fun to complain, and even more fun when it involves Duke.

Again I agree (probably should stop because it may hurt my nonexistent reputation lol ;). Efficiency measures “should” be easier to obtain with 14 Q3/4 games for most of the top 25 teams. Even though the NET has shown to be a better metric than RPI, getting credit for margin of victory and efficiency metrics is like double dipping. I used Duke as an example since they’ve been cited numerous times.

If they beat a Q3/4 by 25 points, obviously they get credit for a 10 point margin. Now add in the efficiencies, you’re basically giving them credit for a full 25 point victory. So yes BREW is correct, they are gaming the system. Not sure it’s intentional or not as I’ll leave that discussion for another day.

I understand why the NET uses predictive analytics, but like you said maybe to heavily. Committee already has enough KenPoms at their disposal to make those decisions. Heck Xavier has been said to be one most inconsistent teams (well pretty consistent with a bad offense), so really hard to predict the W’s and L’s. Also, based on the changing dynamics of a team throughout year is hard to use predictive. Let’s be honest, yesterday’s game with Georgetown shouldn’t have been a Q1 win.

[xu82]

Again I agree (probably should stop because it may hurt my nonexistent reputation lol ;). Efficiency measures “should” be easier to obtain with 14 Q3/4 games for most of the top 25 teams. Even though the NET has shown to be a better metric than RPI, getting credit for margin of victory and efficiency metrics is like double dipping. I used Duke as an example since they’ve been cited numerous times.

If they beat a Q3/4 by 25 points, obviously they get credit for a 10 point margin. Now add in the efficiencies, you’re basically giving them credit for a full 25 point victory. So yes BREW is correct, they are gaming the system. Not sure it’s intentional or not as I’ll leave that discussion for another day.

I understand why the NET uses predictive analytics, but like you said maybe to heavily. Committee already has enough KenPoms at their disposal to make those decisions. Heck Xavier has been said to be one most inconsistent teams (well pretty consistent with a bad offense), so really hard to predict the W’s and L’s. Also, based on the changing dynamics of a team throughout year is hard to use predictive. Let’s be honest, yesterday’s game with Georgetown shouldn’t have been a Q1 win.

Oh, don’t sell yourself short! You have a reputation, alright! We just don’t like to talk about it in front of you....... :-)

[paulxu]

How can you have a losing record (13-15) and be #46 in the NET, and theoretically inside the cut line?

[JTG]

How can you have a losing record (13-15) and be #46 in the NET, and theoretically inside the cut line?

Don't you have to be over .500 for an at large bid ?

[xubrew]

Don't you have to be over .500 for an at large bid ?

Technically no. Not anymore. But practically, yes.

[Xville]

If duke loses tonight, will they love up to 1 in the net?

[noteggs]

Oh, don’t sell yourself short! You have a reputation, alright! We just don’t like to talk about it in front of you....... :-)

Well good, my life’s journey may be complete to have a reputation on a basketball board. Wife and I are going out to celebrate my obscure goals in life lol. BTW, I’m expecting a lot I mean a lot of DM’s on who I should tread lightly around. People can be a little different behind your back...oh wait, you just said that :-)

[xcellentx]

The NET is obviously flawed because DePaul didn't move from 81 even though they were supposed to move up inside 75 so they could improve our wins.

[paulxu]

This will give you guys who understand all the complexities, something to think about until the fall. Enjoy!

https://www.ncaa.com/news/basketball-men/article/2020-05-11/di-mens-basketball-committee-announces-change-net-2020-21

[xubrew]

I think this will make the NET better, but I still don't think it makes much of a difference. Whatever sorting tool they use only really determines how information is organized onto the page.