NET SOS components..

[SotaHawk87]

What a ****ing joke... I know the committee hardly uses Kenpom but Duke has played 3 Kenpom top 50 teams all year. Iowa has played 3 in the last 12 days and TEN so far this year.

The SOR metric the committee uses currently has Dukes SOR at 9 and Iowa’s at 92nd.. I’ll admit it’s better than the RPI but there’s still some messed up stuff going on.

 

[paladinhawk]

Any chance you've got a link to the formula used to calculate SOS? I was surprised that Rutgers SOS was top 10 when compared head to head with Iowa's. My guess is that RPI is heavily factored in, and Iowa played 4 sub 300 NET teams, while Rutger's Quad 4 wins were mostly in the 200s.

Edit I found this link from last year:

https://www.newsobserver.com/sports/college/acc/nc-state/article235014747.html

Still seems like a bogus stat to me...

 

[SotaHawk87]

Any chance you've got a link to the formula used to calculate SOS? I was surprised that Rutgers SOS was top 10 when compared head to head with Iowa's. My guess is that RPI is heavily factored in, and Iowa played 4 sub 300 NET teams, while Rutger's Quad 4 wins were mostly in the 200s.

Edit I found this link from last year:

https://www.newsobserver.com/sports/college/acc/nc-state/article235014747.html

Still seems like a bogus stat to me...

Thats part of it is the NCAA hasn’t released their formula for SOS. WarrenNolan.com has all the team sheets and information though. I’m seeing 3 300+ games whereas Duke only has 1. That being said it seems beyond stupid that a few games against 300s are having a bigger impact than several games against top 50ish NET teams.

 

[SSG T]

Not to throw gasoline on the fire...Duke is 2-1 against KenPom Top 50, Iowa is 6-3.

Duke has 4 games left against KenPom Top 50, Iowa has 11 left and is predicted to go 6-5 against those 11 (plus a win against NE).

Yeah, I get it, the Committee doesn't use KenPom, but Iowa has played and will continue to play a far tougher schedule.

 

[SDHawkDoc]

I'm ok with penalizing a team for playing the worst of D1. Playing one Kennesaw state per season is the max that a good program should schedule.

 

[SotaHawk87]

I'm ok with penalizing a team for playing the worst of D1. Playing one Kennesaw state per season is the max that a good program should schedule.

So playing 3 of them should land you 79 spots worse than a team that plays 1 when your quad 1-3s are way better?

 

[anon_gjzkhykcp9iuj]

Any metrics or statistics that value what you don't do more than what you actually do are worthless. This seems to be one of them.

 

[HawkAssassin]

Kenpom currently has Iowa's non con SOS the highest in the B1G at 67. NET has us with the third worst non con SOS.

Looking at the rest of the B1G, only 5 teams have a NET non con SOS ranking within 50 spots of their kenpom rank. 56 for Indiana in the NET is particularly laughable to me.

 

[SDHawkDoc]

So playing 3 of them should land you 79 spots worse than a team that plays 1 when your quad 1-3s are way better?

We can argue about how much of a penalty. My point is that there should be a penalty and that it should be progressively more penalizing with each one on the schedule.

 

[HeRKeYHoPeFuL]

My guess is that it's doing something similar to RPI for SOS... by playing really good B1G teams, we're getting punished because Minnesota is good, but near .500 due to their schedule (as one example).

 

[FranklinHawk]

I'm ok with penalizing a team for playing the worst of D1. Playing one Kennesaw state per season is the max that a good program should schedule.

Maybe to some extent, but is there really that much difference between #275 and #350?

And, there are always a couple games related to the early season tournaments where Iowa doesn't have a choice.

 

[paladinhawk]

I'm ok with penalizing a team for playing the worst of D1. Playing one Kennesaw state per season is the max that a good program should schedule.

Look at the number of Quad 1 games Duke and Rutgers have played compared to Iowa. You're okay with penalizing teams for playing sub 300 NET teams while not rewarding teams for scheduling tough OOC games? The optimum way to maximize SOS appears to be scheduling as many opponents with a predicted NET between 150 and 300 as possible.

Of course this debate will become moot when the Hawks raise their NET ranking to 16 or better...

 

[WorldSeriesChamps2015]

Just win, nothing else will matter.

 

[SotaHawk87]

Just win, nothing else will matter.

Haha I think that’s the point is both these teams are winning and it’s mattering very differently to each team.

 

[WorldSeriesChamps2015]

Haha I think that’s the point is both these teams are winning and it’s mattering very differently to each team.

Duck Fook!

 

[SDHawkDoc]

Look at the number of Quad 1 games Duke and Rutgers have played compared to Iowa. You're okay with penalizing teams for playing sub 300 NET teams while not rewarding teams for scheduling tough OOC games? The optimum way to maximize SOS appears to be scheduling as many opponents with a predicted NET between 150 and 300 as possible.

Of course this debate will become moot when the Hawks raise their NET ranking to 16 or better...

I don't accept your premise as correct, as teams are clearly rewarded for winning against ranked opponents. It comes down to how the wins and losses are weighted.
If I could make the rules, a win at home against a team in the bottom 10% of D1 would be near the bottom of weighting.
However it's constructed, I want to see the criteria significantly disincentivise playing the worst of the worst.

 

[Hawktagonapus]

I don't accept your premise as correct, as teams are clearly rewarded for winning against ranked opponents. It comes down to how the wins and losses are weighted.
If I could make the rules, a win at home against a team in the bottom 10% of D1 would be near the bottom of weighting.
However it's constructed, I want to see the criteria significantly disincentivise playing the worst of the worst.

Dumb. Cupcakes are cupcakes. You shouldn’t get a huge prize for scheduling the slightly better ones.

 

[Franisdaman]

What a ****ing joke... I know the committee hardly uses Kenpom but Duke has played 3 Kenpom top 50 teams all year. Iowa has played 3 in the last 12 days and TEN so far this year.

The SOR metric the committee uses currently has Dukes SOR at 9 and Iowa’s at 92nd.. I’ll admit it’s better than the RPI but there’s still some messed up stuff going on.

Iowa's SOS has to get better and better as we approach March and Duke's worse....right?

The B1G currently has 12 teams in the Top 50 NET.
The ACC has just 5.

The B1G better get rewarded when it comes to the number of Tournament bids received and the seeds B1G teams receive.

B1G NET Rankings
8 8 Michigan St. Big Ten 14-4 2-1 3-2 9-1 0-0
15 15 Maryland Big Ten 15-4 1-4 3-0 11-0 0-0
18 19 Ohio St. Big Ten 12-6 1-4 1-1 10-1 0-0
20 18 Rutgers Big Ten 13-5 1-4 0-1 12-0 1-0
21 22 Wisconsin Big Ten 12-7 3-3 0-3 9-1 0-0
22 27 Iowa Big Ten 14-5 3-2 2-2 9-1 0-0
29 37 Penn St. Big Ten 14-5 2-3 2-1 10-1 0-0
31 32 Illinois Big Ten 13-5 3-3 0-1 10-1 1-0
37 29 Michigan Big Ten 11-7 0-5 3-0 8-2 0-0
43 44 Minnesota Big Ten 10-8 0-6 1-1 9-1 0-0
46 47 Purdue Big Ten 10-9 1-5 1-2 8-2 0-0
50 51 Indiana Big Ten 14-4 1-3 2-0 11-1 0-0
...............................................
139 138 Northwestern Big Ten 6-12 1-5 1-1 4-6 0-0
169 170 Nebraska Big Ten 7-12 0-6 2-1 5-5 0-0

ACC NET Rankings
6 6 Duke ACC 16-3 4-1 3-0 9-2 0-0
11 10 Louisville ACC 16-3 4-1 1-1 11-1 0-0
14 13 Florida St. ACC 16-2 4-2 3-0 9-0 0-0
44 45 NC State ACC 14-5 3-3 0-1 11-1 0-0
45 43 Virginia Tech ACC 14-5 3-1 1-2 10-2 0-0
.................................
65 66 Virginia ACC 12-6 2-3 2-0 8-3 0-0
69 70 Syracuse ACC 12-7 4-1 0-2 8-4 0-0
74 69 Pittsburgh ACC 13-6 2-2 2-0 9-4 0-0
77 76 Clemson ACC 10-8 1-3 1-1 8-4 0-0
82 75 Notre Dame ACC 11-7 2-3 0-1 9-3 0-0
91 92 Miami (FL) ACC 10-8 3-3 2-2 5-3 0-0
94 97 Georgia Tech ACC 8-11 4-4 1-1 3-6 0-0
104 106 Wake Forest ACC 9-9 2-5 2-1 5-3 0-0
113 118 North Carolina ACC 8-10 1-4 3-1 4-5 0-0
162 171 Boston College ACC 9-10 2-5 1-0 6-5 0-0

 

[HeRKeYHoPeFuL]

Our overall SOS did jump 10 points by playing Rutgers... so it's better than when we were 102.... I guess....

 

[L. Wade Childress]

I'm ok with penalizing a team for playing the worst of D1. Playing one Kennesaw state per season is the max that a good program should schedule.

that’s mathematically impossible. Do you not expect low majors to play full schedules?