2023-2024 B1G Basketball Thread

[MaximumSam]

does anyone understand the NET rankings?

Not really. It is some sort of algorithm. Ones like KenPom use some starting data that slowly fades out as the season progresses. I'm not sure the NET rankings do, which is why the early rankings look wild.

[medinabuckeye1]

does anyone understand the NET rankings?

They are confusing to say the least and they tend to contain oddities early in the season but once there is enough data they tend to roughly match other computer rankings.

Then there is the Quad thing which divides games into four "quads" as follows:
Quad 1:

• Home games against teams ranked 1-30
• Neutral site games against teams ranked 1-50
• Road games against teams ranked 1-75

Quad 2:

• Home games against teams ranked 31-75
• Neutral site games against teams ranked 51-100
• Road games against teams ranked 76-135

Quad 3:

• Home games against teams ranked 76-160
• Neutral site games against teams ranked 101-200
• Road games against teams ranked 136-240

Quad 4:

• Home games against teams ranked 161-353
• Neutral site games against teams ranked 201-353
• Road games against teams ranked 241-353

Current NET rankings:

• 3 Purdue, 5-0 in Q1, loss was Q2
• 14 Illinois, 1-2 in Q1
• 18 Wisconsin, 3-3 in Q1
• 32 Ohio State, 1-1 in Q1
• 43 Michigan State, 1-5 in Q1
• 51 Nebraska, 0-1 in Q1
• 59 Iowa, 0-4
• 70 Michigan, 2-1
• 77 Northwestern, 2-1
• 79 Rutgers, 0-2
• 97 Minnesota 0-1
• 113 Indiana, 1-3
• 136 Penn State, 0-1
• 142 Maryland, 0-1

So, for an example, #32 Ohio State lost on the road to #136 Penn State:

• For Ohio State this is a Q3 loss but right on the border of a Q2 loss.
• For Penn State this is a Q2 win but right on the border of a Q1 win.

[bayareabadger]

Not really. It is some sort of algorithm. Ones like KenPom use some starting data that slowly fades out as the season progresses. I'm not sure the NET rankings do, which is why the early rankings look wild.

It’s kind of like a bastardized version of those sorts of rankings. It uses scoring caps in some situations, but not others. It weighs certain things weirdly.

I think you are correct about not using backwards looking data, which is fine, as long as someone doesn’t take today’s rankings to seriously.

As long as it’s used just as a résumé tool, it’s perfectly fine. But people also can’t help but get out of sorts about rankings.

[medinabuckeye1]

Purdue is back to #1 in the AP and Coaches polls.

[ELA]

It’s kind of like a bastardized version of those sorts of rankings. It uses scoring caps in some situations, but not others. It weighs certain things weirdly.

I think you are correct about not using backwards looking data, which is fine, as long as someone doesn’t take today’s rankings to seriously.

As long as it’s used just as a résumé tool, it’s perfectly fine. But people also can’t help but get out of sorts about rankings.

They seem to primarily use it as a sorting tool, as to say Q1, Q2, Q3, Q4 W/L.  So not to take the rankings themselves totally seriously.  But, I still have 2 problems with that.  #1, I think the Q1 is WAY too generous, particularly with road games.  Granted I'm guessing that's to throw a bone to the P5 leagues, so that every single conference road game is Q1.  Second, the hard line for the boxes.  Beating the #75 team on the road is the same as beating #1.  But beating #76 is the same as beating #135.

[medinabuckeye1]

They seem to primarily use it as a sorting tool, as to say Q1, Q2, Q3, Q4 W/L.  So not to take the rankings themselves totally seriously.  But, I still have 2 problems with that.  #1, I think the Q1 is WAY too generous, particularly with road games.  Granted I'm guessing that's to throw a bone to the P5 leagues, so that every single conference road game is Q1.  Second, the hard line for the boxes.  Beating the #75 team on the road is the same as beating #1.  But beating #76 is the same as beating #135.

True but not much different than talking about ranked opponents in football. There is a big difference between playing #1, #2, and #3 on the road and playing #23, #24, and #25 at home but both are "three ranked opponents".

[ELA]

I think it's an overreaction to how the RPI treated every gap as equal.

[boilerbanger]

They seem to primarily use it as a sorting tool, as to say Q1, Q2, Q3, Q4 W/L.  So not to take the rankings themselves totally seriously.  But, I still have 2 problems with that.  #1, I think the Q1 is WAY too generous, particularly with road games.  Granted I'm guessing that's to throw a bone to the P5 leagues, so that every single conference road game is Q1.  Second, the hard line for the boxes.  Beating the #75 team on the road is the same as beating #1.  But beating #76 is the same as beating #135.

I hadn't thought about this before in how they are weighted within the 4 quadrants, you make a good point not all wins within the Quad are equal.

[ELA]

Indiana down. 11 at home to Morehead State

[ELA]

Indiana down. 11 at home to Morehead State

20-4 run to close for Indiana, after falling behind by 15.  IU wins by 1

[medinabuckeye1]

20-4 run to close for Indiana, after falling behind by 15.  IU wins by 1

Not a banner evening for the B1G. Indiana and Maryland both more-or-less escaped in home games that should have been blowout wins.

Michigan almost made up for it. They were a slight underdog and pushed Florida to 2OT before coming up short in Charlotte. 

[medinabuckeye1]

Relating to @boilerbanger 's question and our discussion of ranking systems, here are the latest NET and KenPom rankings for all B1G teams. The format here is league rank (based on NET), Team, NET ranking, KenPom ranking:

• Purdue, 4, 2
• Wisconsin, 18, 17
• Illinois, 20, 14
• Ohio State, 34, 27
• Michigan State, 41, 23
• Nebraska, 49, 45
• Iowa, 62, 53
• Michigan, 72, 47
• Rutgers, 79, 75
• Northwestern, 82, 66
• Minnesota, 96, 89
• Indiana, 110, 87
• Penn State, 136, 111
• Maryland, 147, 79

Gaps of 15+:

• Northwestern is 16 spots higher in KenPom 
• MSU is 18 spots higher in KenPom 
• Indiana is 22 spots higher in KenPom 
• Michigan is 25 spots higher in KenPom 
• PSU is 25 spots higher in KenPom 
• Maryland is 66 spots higher in KenPom 

[ELA]

Nice bounce back for Northwestern 

[boilerbanger]

@medinabuckeye1  what data in the models drives such a rankings difference for those teams?  Or is there not much difference a NET of 136 and 111 that it is small differences that just look larger because the teams in that range are all clumped so closely together?