NET Rankings / Bracketology

Status
Not open for further replies.
We’re going to be ranked Monday, and - unless we lose to anyone not named OSU or Iowa - we’ll likely finish the year ranked. Stare at those rankings and computer formulas all you like, color me skeptical that a ranked team from the Big Ten will be seeded below a 6 or 7 at worst for the simple reason that it would SCREW some 23-win team from a weaker conference to match them up with us. Human decision making will simply be way more important than most are counting on, IMO.
 
       
  • Like
Reactions: IndyILL
Loda, IL
We’re going to be ranked Monday, and - unless we lose to anyone not named OSU or Iowa - we’ll likely finish the year ranked. Stare at those rankings and computer formulas all you like, color me skeptical that a ranked team from the Big Ten will be seeded below a 6 or 7 at worst for the simple reason that it would SCREW some 23-win team from a weaker conference to match them up with us. Human decision making will simply be way more important than most are counting on, IMO.
Color me skeptical, and I hope I am wrong, but I do not think we will be ranked on Monday.
 
       
You can't reconcile because you're only looking at the stats and games that support your belief. Don't ignore the stats that make Purdue look better than us like...

-They beat MSU by 30, we lost to MSU by 1
-They beat Wisconsin by 20, we beat Wisconsin by 1
-They beat NW by 14, we beat NW by 4

The biggest takeaway is margin of victory does matter a lot in these rankings. For example, if we would have lost to Wisconsin by 1, lost to PSU by 1, and lost to Michigan by 1 our record would look very similar to Purdue's. But our NET ranking would be almost the same as it is now, despite our record looking much worse.
Mmmm... supposedly they don't look at any margin greater than 10. They lost at home again today and were never really in the game vs. Michigan. All this computer mumbo jumbo sometimes over looks what we can all see. We beat all three of those teams on the road. Purdue has barely beat NW on the road and one good win vs Indiana.
 
       
  • Like
Reactions: Tacomallini and Fighter of the Nightman
Based on yesterday’s Net Rankings Purdue is #33, and we’re #32..
 
       
Cincinnati, OH
Those two Michigan wins are looking better by the week.
 
       
  • Like
Reactions: MoCoMdIllini, Fighter of the Nightman and BBlock24
Color me skeptical, and I hope I am wrong, but I do not think we will be ranked on Monday.
Well, #17, 18, 19 and 22 have already lost today and #23 and #24 are playing higher ranked teams tonight. With the road win over #9 Penn State this week, I'd be shocked if we're not ranked.
 
       
  • Like
Reactions: CHEEF and IndyILL
As best I can tell, NET does not consider head to head. It’s all numbers and positioning (as in rank of opponents). Not surprising and not sure why it is hard for people to understand. As such it is a tool for the selection committee. And I would think they’ll lean on it fairly hard in selecting at large teams. I would hope when it comes to bubble teams and seeding that head to head and common opponents receives the weight it should.
 
       
  • Like
Reactions: bert1991
That garbage SDSU team lost, I wonder if the committee would take them off the #1 line considering they beat Iowa and no one else.

Regardless, we'll be an 8/9 in Spokane against Gonzaga because Illini.
 
       
That garbage SDSU team lost, I wonder if the committee would take them off the #1 line considering they beat Iowa and no one else.

Regardless, we'll be an 8/9 in Spokane against Gonzaga because Illini.
While i agree with you that SDSU is overrated they are far from garbage, they have a legit all american in there back court
 
       
  • Like
Reactions: FiveStar
Waukegan
Some posters act like the committee is just going to go down the latest NET list on Selection Sunday ... I was under the impression it was a new tool they were using to help them a bit more, nothing more and nothing less.

If we finish 4-1 (difficult, but that’s my prediction), we’re about a 7, possibly even a 6. BTT could get us in some real nice territory.
I was going to make a comment or question along the same lines. Isn't this just the second year the committee is using NET? Hopefully, the powers that be have made some modifications from last year and will continue to make adjustments along the way. And more importantly hopefully they have been reading this thread on IL to give them guidance.
 
       
  • Like
Reactions: Fighter of the Nightman
I don’t see 4 of the 5 factors changing much as their pretty straight forward calculations, but having the Team Value Index component undefined they have a lot of flexibility with that aspect. NET was developed in collaboration with Google Cloud so I’m fairly certain the TVI is the output of one of Google‘s machine learning products. They likely have a model that’s consumed every box score past and present, and ran an analysis against historical tourney results, then added exception rules for if the team name is Duke, Kansas, etc (haha).

I’m not defending it as great but I like it much much more than RPI.

Also, Purdue has got to have bubble bursted now, right? And I think Indiana same spot as us now - just don’t fully collapse and in.
 
       
I don’t see 4 of the 5 factors changing much as their pretty straight forward calculations, but having the Team Value Index component undefined they have a lot of flexibility with that aspect. NET was developed in collaboration with Google Cloud so I’m fairly certain the TVI is the output of one of Google‘s machine learning products. They likely have a model that’s consumed every box score past and present, and ran an analysis against historical tourney results, then added exception rules for if the team name is Duke, Kansas, etc (haha).

I’m not defending it as great but I like it much much more than RPI.

Also, Purdue has got to have bubble bursted now, right? And I think Indiana same spot as us now - just don’t fully collapse and in.

Purdue and Minnesota both still have a chance. They would both be out right now, but both have good wins left on their schedule. 3 wins probably gets Purdue right on the bubble and 4 wins gets them in. 3 wins doesn't do it for Minnesota, but 4 wins probably gets them in. That being said, both teams probably aren't good enough to do that. Indiana probably only needs 1 more win, but 2 would get them in comfortably. Their schedule is tricky though so that's no guarantee. They don't have Nebraska or NW left like we do.
 
       
Right now I think we will be ranked. Of the four public brackets I saw it was NR, NR, 19, 13. We can half be NR if people are ranking us in the low 20’s. Either way it will be between 24-26 IMO.
 
       
Right now I think we will be ranked. Of the four public brackets I saw it was NR, NR, 19, 13. We can half be NR if people are ranking us in the low 20’s. Either way it will be between 24-26 IMO.
It's very unlikely that we'll be ranked. A win over Penn State doesn't just erase the four straight losses prior. We only received 12 votes last week, while teams like Michigan and Arizona State are on fire.

Who knows. Maybe I'm way off base.

More importantly, we just need to keep winning.
 
       
Waukegan
It's very unlikely that we'll be ranked. A win over Penn State doesn't just erase the four straight losses prior. We only received 12 votes last week, while teams like Michigan and Arizona State are on fire.

Who knows. Maybe I'm way off base.

More importantly, we just need to keep winning.
That's pretty much my thought as well, the PSU win is not going to negate the four straight losses, two at home. But don't the rankings come out Sunday night or Monday morning. I just checked and the previous week is still up (I think)?
 
       
That's pretty much my thought as well, the PSU win is not going to negate the four straight losses, two at home. But don't the rankings come out Sunday night or Monday morning. I just checked and the previous week is still up (I think)?
The new AP Poll should be released around 11 AM Central. Coaches Poll is released around 1 PM or so.
 
       
  • Like
Reactions: jmilt7
It's very unlikely that we'll be ranked. A win over Penn State doesn't just erase the four straight losses prior. We only received 12 votes last week, while teams like Michigan and Arizona State are on fire.

Who knows. Maybe I'm way off base.

More importantly, we just need to keep winning.
Not in AP top 25 with ASU also unranked but receiving more votes
 
       
I don't care where the Illini are ranked today. Let's see where there are in three weeks.
 
       
  • Like
Reactions: BlueLeader and pvaughn8
danville il
Curious to see what kind of respect the winner of the BTT gets, especially if its not maryland.
 
       
I just want to know how any other these metrics work - ap poll, coaches, kenpom, NET, Momma and Friends, etc. It seems like when we get a great win (on the road included) there is not much value, but let us loose to a ranked opponent on the road or a top 10 team and we have to prove ourselves for multiple weeks to get any kind of recognition.
 
       
I just want to know how any other these metrics work - ap poll, coaches, kenpom, NET, Momma and Friends, etc. It seems like when we get a great win (on the road included) there is not much value, but let us loose to a ranked opponent on the road or a top 10 team and we have to prove ourselves for multiple weeks to get any kind of recognition.
They all have different added flavors. Kenpom is focused on efficiency, rather than win/loss. AP/Coaches polls are just who writers and coaches think are good. NET is supposed to be the perfect love child of win percentage/margin of victory/efficiency.

This season has emphasized parity, but (and I say this without actual data, just feel) I don't think it's that unique of a season, the B1G not included. Yes, the top 3ish is normally more stable, but the usual suspects are represented well and often in the top 15. There's always flux between the 20-40 range teams.

That being said, the B1G is basically two tiers NW and Neb, then everyone else. Sure, Purdue is probably going to lose more games than MD, but any given night it isn't a complete shock when a Purdue routs an MSU, or we beat a PSU on the road. I think the models are having a tough time differentiating the B1G teams, probably over inflating their overall value. At least that's the glass half empty explination of how theres about 12 B1G times in the top 50 of NET. Half full would be saying there really is that much talent, which is possible.

At the end of the day, it does look great for an unranked illini team to beat PSU away, but it isn't something that's unheard of this year, and we didn't particularly demolish them. A single digit win in the fashion we did it, against a team that was probably overranked at 9th in the AP moves the needle a few spots, but it doesn't mean we flip flop rankings in all the ones you mentioned. Also important to note that even if we did beat the 9th best team in the nation, we're probably the 28th-ish best team in the nation. Not a rediculous win by any means.


Something I still don't understand is how all these rankings manage to place teams like Wisconsin and Purdue consistently above us (not to mention Michigan or OSU). My assumption those losses to Miami and Mizzou look bad, and we haven't really crushed any A+ opponent. We've come out ahead at the end of a lot of B/B+/A- wins, but we've been comfortable in like 2? Ayo's won us at least 3 himself. 0-4 against MSU and MD hurts a lot too, considering they seem to have a stronghold on the top of the conference. I think if we beat an OSU/Indiana/Iowa handedly, we start moving up to the teens in these rankings.
 
       
  • Like
Reactions: JLAandLKR
They all have different added flavors. Kenpom is focused on efficiency, rather than win/loss. AP/Coaches polls are just who writers and coaches think are good. NET is supposed to be the perfect love child of win percentage/margin of victory/efficiency.

This season has emphasized parity, but (and I say this without actual data, just feel) I don't think it's that unique of a season, the B1G not included. Yes, the top 3ish is normally more stable, but the usual suspects are represented well and often in the top 15. There's always flux between the 20-40 range teams.

That being said, the B1G is basically two tiers NW and Neb, then everyone else. Sure, Purdue is probably going to lose more games than MD, but any given night it isn't a complete shock when a Purdue routs an MSU, or we beat a PSU on the road. I think the models are having a tough time differentiating the B1G teams, probably over inflating their overall value. At least that's the glass half empty explination of how theres about 12 B1G times in the top 50 of NET. Half full would be saying there really is that much talent, which is possible.

At the end of the day, it does look great for an unranked illini team to beat PSU away, but it isn't something that's unheard of this year, and we didn't particularly demolish them. A single digit win in the fashion we did it, against a team that was probably overranked at 9th in the AP moves the needle a few spots, but it doesn't mean we flip flop rankings in all the ones you mentioned. Also important to note that even if we did beat the 9th best team in the nation, we're probably the 28th-ish best team in the nation. Not a rediculous win by any means.


Something I still don't understand is how all these rankings manage to place teams like Wisconsin and Purdue consistently above us (not to mention Michigan or OSU). My assumption those losses to Miami and Mizzou look bad, and we haven't really crushed any A+ opponent. We've come out ahead at the end of a lot of B/B+/A- wins, but we've been comfortable in like 2? Ayo's won us at least 3 himself. 0-4 against MSU and MD hurts a lot too, considering they seem to have a stronghold on the top of the conference. I think if we beat an OSU/Indiana/Iowa handedly, we start moving up to the teens in these rankings.

That's exactly my point. We literally moved 0 spots in the coaches poll for beating Penn State on the road. I didn't think we would jolt 7 spots but 3 or 4 would seem reasonable.
 
       
Status
Not open for further replies.