Proof the NET is bogus

[bac2therac]

 

[kcg88]

It uses RPI for SOS which is an insane thing to do as the NET is meant to be a replacement for the RPI, which everyone agreed was bad.

 

[bac2therac]

they seem to make no distinction between beating Rutgers or beating St Joes

 

[RUChoppin]

I had mentioned this earlier when comparing SOS between SHU and Rutgers - it made no sense. No idea what their formula is - but if they're really using the RPI SOS, that's dumb.

 

[Erial_Lion_rivals140457]

The SOS on the team sheets has no connection to the NET. From what I’ve been told, it’s not even as advanced as the RPI SOS. It’s simply your opponent’s winning %, not even adjusted for game locations.

 

[bac2therac]

so a 17-14 school from the Ivy League is considered the same as a 17-14 from the Big 10

 

[RUChoppin]

The SOS on the team sheets has no connection to the NET. From what I’ve been told, it’s not even as advanced as the RPI SOS. It’s simply your opponent’s winning %, not even adjusted for game locations.

Sure, because beating the 30-4 Liberty Flames or 28-4 ETSU Buccaneers should absolutely count more than beating, say, FSU, Baylor, MSU or Duke.

 

[BillyC80]

That’s an eye-opening chart.

 

[Greene Rice FIG]

I think we compared Indiana and Rutgers and they had RU with a better SOS despite the fact the NET rankings had it not even close the other way

 

[bac2therac]

by glancing at RU's sos non conference, it looks terrible but somehow it turned out okay..at around 100...needs to actually be better going forward so I hope the metrics guys that work for RU are advising Pike going forward

 

[Scarlet Blind_rivals]

Michigan 507-407 .555
Dayton 525-414 .559
Not opponent win %

 

[DHajekRC84]

I'm shocked. I must have totally misunderstood your position on this system cause I thought you liked and defended it. I'm refreshed to hear your position.

I'm no expert but was part of the list of folks chirping on the flaws all season. Its more bull crap and looks like its made up by pocket protector guys who gave no common sense. Like the folks at corp. Who shoved SAP down our throats when all the foljs were saying hold on , only to see a 3 year nightmare of false assumptions and pieces that didn't fit or make sense.

 

[bac2therac]

Its a sorting tool they use. I defend it because its what used not other metrics. That said its deeply flawed even as it probably benefits Rutgers sos vs other metrics

 

[ecojew]

It's very odd that the SOS lower for Michigan while the NET overall rates the B1G team quite highly, with something like 10 in the top 40. Meanwhile the RPI seems to slight the B1G in its rankings this year while favoring the BE.

Right now, though, with RU's NET ranking quite high, I'm hoping the Committee will focus on that when comparing us to teams from other conferences.

 

[DHajekRC84]

Its a sorting tool they use. I defend it because its what used not other metrics. That said its deeply flawed even as it probably benefits Rutgers sos vs other metrics

Yeah its a sorting tool all right. Kinda like picking out the hottest lingerie for your wife or girlfriend to wear by putting colors in 1 pile and whites in another. Who cares about material and size

 

[zappaa]

We have a better NET rank at 32 than PSU, IOWA or Illinois.
All behind us.
Am I completely off the topic? Like I said earlier, I’m clueless with all these metrics

 

[Scangg]

How closely did seeding reflect NET last year?

 

[Scarlet Blind_rivals]

The quick answer

32 Rutgers Q1 4-9 Q2 5-1 9-10
34 Iowa Q1 7-8 Q2 6-1 13-9
35 PSU Q1 7-7 Q2 6-2 13-9
38 Illinois Q1 5-8 Q2 5-1 10-9

 

[zappaa]

The quick answer

32 Rutgers Q1 4-9 Q2 5-1 9-10
34 Iowa Q1 7-8 Q2 6-1 13-9
35 PSU Q1 7-7 Q2 6-2 13-9
38 Illinois Q1 5-8 Q2 5-1 10-9

Only I don’t understand the answer

 

[bac2therac]

The quick answer

32 Rutgers Q1 4-9 Q2 5-1 9-10
34 Iowa Q1 7-8 Q2 6-1 13-9
35 PSU Q1 7-7 Q2 6-2 13-9
38 Illinois Q1 5-8 Q2 5-1 10-9

they all have better resumes...but its very likely that because RU did not get blown out this year especially on the road, it has really helped our overall NET. Some of these other teams have one if not two blowout losses. Imagine if RU had won 2 more road games their NET really would be through the roof

 

[RUChoppin]

Here's a breakdown of NET vs. seeding from last year that's worth a read:

https://watchstadium.com/what-would...t-look-like-based-on-net-rankings-03-18-2019/

 

[bac2therac]

USC rose 2 spots in the NET despite not playing......is it just because St Marys beat BYU....note that St Marys has a couple wins over Pepperdine and USC beat Pepperdine...is that truly the reason.

 

[fluoxetine]

This isn't proof the NET is bogus, because I'm pretty sure this SOS number is not a component of the NET. However, this SOS is demonstrably ridiculous and the NET is flawed at best.

 

[Big boy stan]

How closely did seeding reflect NET last year?

Its better than you think. According to the article RUChoppin mentioned...

After sorting the tournament teams from top to bottom and assigning each team a seed 1-16 based on NET, I calculated the seed difference for each team based on the number it received from the committee. When averaging the seed difference across all 68 teams, the average seed difference comes out to 0.735 either way. This means the committee sorted the field it selected very similarly to how it would’ve been sorted by NET.

 

[bac2therac]

for the schools seeded 13-16 its very likely they go by order.

 

[PSAL_Hoops]

What exactly is strength of record? Apparently Archie Miller went on a public call out session of Lunardi over the weekend saying Indiana should be in as a lock because of their good strength of record. He called Lunardi’s bracketology the sesame street show or something like that. How that calculated?

 

[bac2therac]

here are some answers here

https://www.espn.com/mens-college-basketball/bpi

 

[PSAL_Hoops]

Thanks - I see the little blurb at the bottom about probability of a 25th ranked team matching your record but how are they determining that chance?