GW Hoops / NET Ratings

**Posted by GW0509** 12/04/2023 9:28 am

Initial NET ratings have been released. Here's the A10. GW is safely within the top 100 NET.

As of now the SC loss is a Quad 1 and the UIC loss is a Quad 2.

Last edited by GW0509 (12/04/2023 9:29 am)

**Posted by Free Quebec** 12/04/2023 10:41 am

That’s NIT territory!

(I kid - unlikely even going 4-0 the next 4 that we stay in the top 100).

Edit: Hofstra is 67 so that’s a Quad 2 win.

Last edited by Free Quebec (12/04/2023 10:43 am)

**Posted by BM** 12/04/2023 11:55 am

South Carolina at 29, Delaware 98.

**Posted by Gwmayhem** 12/04/2023 12:24 pm

Curious as to what people think:

South Carolina is a Q1 loss but theoretically speaking, they could have a disastrous season from here on in and end up being a Q2 or perhaps even a Q3 loss. Yet at the time we played them, they were undefeated.

Do you think NET should be based on where a team was ranked at the time you played them, or based on where the opponents end up by season's end?

**Posted by GW0509** 12/04/2023 7:53 pm

Gwmayhem wrote:
Curious as to what people think:

South Carolina is a Q1 loss but theoretically speaking, they could have a disastrous season from here on in and end up being a Q2 or perhaps even a Q3 loss. Yet at the time we played them, they were undefeated.

Do you think NET should be based on where a team was ranked at the time you played them, or based on where the opponents end up by season's end?

I think it has to be the former. I understand the logic of teams getting better over time so a win/loss in November might look better in March. But GW beat November's version of Hofstra, not March's. Hofstra may get better or worse as the season goes on but it'll be for reasons unrelated to GW. Not sure why GW should be rewarded/punished for that.

**Posted by The Dude** 12/04/2023 10:11 pm

Very cool, thx for posting!

6 of our 8 games have been vs teams in the NET top 122. We're 4-2 in those games

2-1 in Quad 2,

**Posted by Free Quebec** 12/05/2023 12:21 am

GW0509 wrote:
Gwmayhem wrote:
Curious as to what people think:

South Carolina is a Q1 loss but theoretically speaking, they could have a disastrous season from here on in and end up being a Q2 or perhaps even a Q3 loss. Yet at the time we played them, they were undefeated.

Do you think NET should be based on where a team was ranked at the time you played them, or based on where the opponents end up by season's end?

I think it has to be the former. I understand the logic of teams getting better over time so a win/loss in November might look better in March. But GW beat November's version of Hofstra, not March's. Hofstra may get better or worse as the season goes on but it'll be for reasons unrelated to GW. Not sure why GW should be rewarded/punished for that.

I don’t think the former is workable for early season games. Like our first game against Stonehill, they didn’t have a rating so how would you score that?

It would encourage teams not to schedule hard games early because what’s the point of, say, James Madison knocking off Michigan St if they get no credit for it because MSU was unrated?

I understand the challenges of teams getting better or worse (especially with injuries affecting teams), but the formula isn’t valid until enough games have been played. That’s why it can’t be gospel, but just one tool.

**Posted by Free Quebec** 12/05/2023 12:25 am

Also, Siena is dead last at 362.

Georgetown is 221.

**Posted by Gwmayhem** 12/05/2023 9:45 am

Our next 5 are against (in the NET rankings) 295, 358, Division 2 opponent, 324 and 336.

As Bruce Springsteen once sang, "I'm going down, down, down, down...."

**Posted by GW0509** 12/05/2023 9:55 am

Gwmayhem wrote:
Our next 5 are against (in the NET rankings) 295, 358, Division 2 opponent, 324 and 336.

As Bruce Springsteen once sang, "I'm going down, down, down, down...."

Depends on if we can blow them out of the water. It's probably not realistic, but you never know.

For example, Iowa State is currently #13 in the NET. They've played 1 Q1 game (loss) 1 Q2 game (loss) 1 Q3 game (win) and then a bunch of Q4 games where they've absolutely annihilated sub-200 teams.

Last edited by GW0509 (12/05/2023 9:56 am)

**Posted by DC Native** 12/05/2023 1:06 pm

A10 is dominating the DMV right now:

62 George Mason
86 George Washington
190 Howard
210 Maryland
224 Georgetown
327 American

**Posted by BGF** 12/06/2023 12:56 pm

I always found the RPI to be simple, which may have been its problem. The movement on the NET rankings is much more difficult to understand, with the margin of victory/defeat having value..

In any case, after last night's game, GW moved from 86 to 92. Navy moved from 295 to 282. We sit 7th in the A-10 now, with Duquesne and UMass ahead of us.

**Posted by PKGW** 12/07/2023 4:33 pm

Not up to date on the analytics - what is the main difference between kenpom having us in the 150+ and NET having us under 100? Just curious

**Posted by Free Quebec** 12/07/2023 5:15 pm

PKGW wrote:
Not up to date on the analytics - what is the main difference between kenpom having us in the 150+ and NET having us under 100? Just curious

NEt is like a cross between RPI and KenPom so it weights wins more heavily. KenPom doesn’t care at all if you win.

**Posted by Alum1** 12/07/2023 5:22 pm

PKGW wrote:
Not up to date on the analytics - what is the main difference between kenpom having us in the 150+ and NET having us under 100? Just curious

Here is a handy explanation:

While both methods use similar data, the biggest differences are

1) NET uses game location as a factor in the ranking system, while KenPom does not factor where the game was played in rankings.

2) Ken Pom uses adjusted efficiency based on how many possessions a team has per game while NET uses points per 100 possessions regardless of possession per game.

3) Scoring margin does not matter to the NET, but it does to KenPom.

4) KenPom does not factor wins and losses, it is simply a efficiency number.

5) KenPom uses strict formulas (predictive) to create ranking while NET uses a learning algorithm comparing what is expected to what happens

https://www.mwcconnection.com/2023/1/12/23552090/stats-corner-net-vs-kenpom-rankings

Last edited by Alum1 (12/07/2023 5:22 pm)

**Posted by jf** 12/07/2023 5:23 pm

Still think it's unsportsmanlike and somewhat immaterial to consider running up the score in the rankings..
A win is a win in general, outside of coaches and fans analyzing what happened.
And rewarding running up the margin is against everything college sports should be. Doesn't help getting players who practice every day in the game for even a little bit.
College basketball falls way short already in the notion of intercollegiate sports, much less scholar athletes. Don't think this helps.

**Posted by Free Quebec** 12/18/2023 12:41 pm

The A10 is playing the NET game really well this year.

Currently 2 teams in the top 50.
5 in the top 75
8 in the top 100.

We are 10th at 107.

If the league can maintain these ratings, there are 5 teams that offer quad 1 opportunities (a road game vs top 75 is quad 1) and 5 more teams that offer quad 2 opportunities (76-135 in the road is quad 2).

With Dayton at 30 and St Joe at 48 with some high profile good results, the A10 has a strong shot at 2 or even 3 bids (if both get to at large and someone else wins the auto bid). Of course, we are all going to beat each other up, so will be interesting to see where our net ratings sit later in the year.

**Posted by GW0509** 12/18/2023 1:31 pm

Free Quebec wrote:
The A10 is playing the NET game really well this year.

Currently 2 teams in the top 50.
5 in the top 75
8 in the top 100.

We are 10th at 107.

If the league can maintain these ratings, there are 5 teams that offer quad 1 opportunities (a road game vs top 75 is quad 1) and 5 more teams that offer quad 2 opportunities (76-135 in the road is quad 2).

With Dayton at 30 and St Joe at 48 with some high profile good results, the A10 has a strong shot at 2 or even 3 bids (if both get to at large and someone else wins the auto bid). Of course, we are all going to beat each other up, so will be interesting to see where our net ratings sit later in the year.

Equally important is that currently we only have 4 Q4 teams (Loy-Chi, SLU, Fordham, URI) and those are only Q4 if you play them at home. On the road they're Q3.

As much as it stinks playing a cupcake schedule, this is what CC was talking about at the beginning of the season. If you're not expecting to be a top-4 team in the A10, your responsibility to the conference is just to win games and raise the floor for the bottom teams so nothing is truly a bad loss.

Last edited by GW0509 (12/18/2023 2:54 pm)

**Posted by BM** 12/18/2023 2:50 pm

The A-10 teams are favored by KenPom in all but three of the remaining 29 OOC games moving forward. Underdogs in George Mason at Tulane and SLU at NC State, and toss-up in Georgia Tech v. UMass (The Gapare Bowl).

The winning percentages are likely to stay favorable.

**Posted by GW0509** 12/19/2023 11:31 am

@GoodmanHoops 1h
Iowa State moved up to No. 6 in the NET after its 38-point win over FAMU. The Cyclones don't have a win over an NCAA tourney team, and seven of their nine victories have come against sub-200 teams.

Last edited by GW0509 (12/19/2023 11:32 am)