• Welcome to OGBoards 10.0, keep in mind that we will be making LOTS of changes to smooth out the experience here and make it as close as possible functionally to the old software, but feel free to drop suggestions or requests in the Tech Support subforum!

MBB Game 14: VT Hokies @ LJVM - Saturday Noon - Regional Sports Networks

thatguy2016 said:
80 but they were 7-7 vs Quad 1

Rutgers played an awful, AWFUL non-con last year but also played in the B1G and beat good B1G teams, mostly at home.

I mean their non con was AWFUL

their best non-con win was Clemson at home. Other than that, they beat 5 teams rated 290 or worse in NET. They went 6-4 OOC, and Seton Hall was the only other remotely decent OOC team they played. Seton Hall blew them out.

Rutgers also lost to DePaul (103 in NET), UMASS (178), and Lafayette (319!!!)

The focus on Quad 1 above and beyond the NET rating is an absolute joke (when you do that, a road win over the #75 team in the country is exactly as good as a road win over #1, and infinitely better than a road win over #76), especially since who you played and how you did is already baked into NET.
Click to expand...
I really struggle with NET as well as the quad systems as well, but just for argument’s sake, I’ll take the positive side of their very unique resume.

While I hadn’t seen a team with metrics that low (KP 77, NET 80) make the tournament before, I also hadn’t seen a team ranked there with as many good wins as they had.

Point taken on #75, etc but none of their wins were cheapies. They beat all 8 other B1G tournament (at-large, not a random smaller conference team that snuck in) teams.
3-seed WI and 11 seed Indiana on the road.
3, 4, 5, 7, 7, 11 seeds at home, including a bunch in a row headed to Selection Sunday. The games are all supposed to count equally, but we know they don’t completely if you show an upside of being able to beat a bunch of top tier teams while improving over the course of the season.

And then the played eventual sweet-16 ND to 2OT and acquitted themselves well once they got there.

I had a bigger issue with someone like Wyoming, where the NET wins became circular in the MWC, yet none of them had beaten anyone good outside of the conference — leaving me wondering if any of those teams were particularly good.
 
I think our NET is off because it only has us with one quad 1 and one quad 2 win. There should be at least 3 combined with Wisconsin Duke and VT.
 
It has us 2-1 Q1 (Duke and Wisc) and 1-3 Q2 (VT).

UVU is at #80 so they're close to a Q2 win.
Clemson is at #75 so that should be the first of two Q1 losses. So that's a Net mistake. We should be 2-2 Q1 and 1-2 Q2 (VT; LSU and LMU). Honestly, I'm not sure if a better Q1 record and worse Q2 record looks better or worse to the selection committee.
Rutgers is going to be a Q1 loss no matter what.

Here are the remaining likely Q1 & Q2 games:
1/4 at UNC (Q1)
1/17 vs. Clemson (Q2)
1/21 vs. Virginia (Q1)
1/25 at Pitt (Q1 or Q2)
1/28 vs. State (Q2)
1/31 at Duke (Q1)
2/7 vs. UNC (Q1 or Q2)
2/18 at Miami (Q1)
2/22 at State (Q1)
3/4 at Cuse (Q2)

So that's 10 out of 17 remaining games likely to be Q1 or Q2. So let's say the remaining games are 5 Q1, 5 Q2, and 5 Q3/4. I think the following puts us firmly on the bubble.

2-3 vs. Q1
3-2 vs. Q2
7-0 vs. Q3/4

That's 22-9 (14-6). That's going to be really tough. A game better than last year in the league.
 
Last edited:
PhDeac said:
It has us 2-1 Q1 (Duke and Wisc) and 1-3 Q2 (VT).

UVU is at #80 so they're close to a Q2 win.
Clemson is at #75 so that should be the first of two Q1 losses. So that's a Net mistake. We should be 2-2 Q1 and 1-2 Q2 (VT; LSU and LMU). Honestly, I'm not sure if a better Q1 record and worse Q2 record looks better or worse to the selection committee.
Rutgers is going to be a Q1 loss no matter what.

Here are the remaining likely Q1 & Q2 games:
1/4 at UNC (Q1)
1/17 vs. Clemson (Q2)
1/21 vs. Virginia (Q1)
1/25 at Pitt (Q1 or Q2)
1/28 vs. State (Q2)
1/31 at Duke (Q1)
2/7 vs. UNC (Q1 or Q2)
2/18 at Miami (Q1)
2/22 at State (Q1)
3/4 at Cuse (Q2)

So that's 10 out of 17 remaining games likely to be Q1 or Q2. So let's say the remaining games are 5 Q1, 5 Q2, and 5 Q3/4. I think the following puts us firmly on the bubble.

2-3 vs. Q1
3-2 vs. Q2
7-0 vs. Q3/4

That's 22-9 (14-6). That's going to be really tough. A game better than last year in the league.
Click to expand...
man, something has gotten completely fucked if a 14-6 ACC team is On the bubble.
 
Haros said:
I really struggle with NET as well as the quad systems as well, but just for argument’s sake, I’ll take the positive side of their very unique resume.

While I hadn’t seen a team with metrics that low (KP 77, NET 80) make the tournament before, I also hadn’t seen a team ranked there with as many good wins as they had.

Point taken on #75, etc but none of their wins were cheapies. They beat all 8 other B1G tournament (at-large, not a random smaller conference team that snuck in) teams.
3-seed WI and 11 seed Indiana on the road.
3, 4, 5, 7, 7, 11 seeds at home, including a bunch in a row headed to Selection Sunday. The games are all supposed to count equally, but we know they don’t completely if you show an upside of being able to beat a bunch of top tier teams while improving over the course of the season.

And then the played eventual sweet-16 ND to 2OT and acquitted themselves well once they got there.

I had a bigger issue with someone like Wyoming, where the NET wins became circular in the MWC, yet none of them had beaten anyone good outside of the conference — leaving me wondering if any of those teams were particularly good.
Click to expand...

Counterpoint - maybe the Big 10 wasn’t very good, since they put 0 of their 9 teams in the elite 8 (compared to 3 of 5 teams for the ACC).
 
PhDeac said:
A 13-7 ACC team didn’t make it last year and found out they weren’t really on the bubble.
Click to expand...

And in retrospect, that was ridiculous given the ACC had the highest win percentage in the tournament for multibid leagues (and the big 10 was 6th and SEC 7th).

The ACC media talking heads should start harping on this nightly in February. I think the main talking point should be that the NET doesn’t account well for improvement as the season progresses.
 
Rafi said:
Counterpoint - maybe the Big 10 wasn’t very good, since they put 0 of their 9 teams in the elite 8 (compared to 3 of 5 teams for the ACC).
Click to expand...
This is the correct answer.
 
I think the NET has its issues but I generally like the concept as it provides a better predictive system based on adjusted efficiency when compared to RPI or the eye test. Also, the NET doesn’t itself place teams in quads, AFAIK the quad system is just a means in place to extract the NET data in a consumable manner for the committee.

There isn’t much disagreement across the board where Wake stands regardless of source right now:

81 in NET
75 in Torvik
78 in Sagarin
88 in KenPom
62 in Massey

I have a far bigger issue in the inconsistent way the committee evaluates teams’ resumes when provided with metrics on top of that. The general theme seems to be: win Q1 games and play a decent OOC schedule. Beyond that I don’t know. They don’t seem to care as much about losses (even bad losses) so playing more than a couple Q4 teams doesn’t do much for you - not always 100% way to predict entering the season who will be there but overall a good idea. Teams like Louisville who are somehow bottom 40 in the nation as an ACC team isn’t something we can control.

Last thought is that teams who are outliers in the NET right now will generally work their way out one way or the other. NET critics love to say “Florida Atlantic is 10th on January 1 what a sham system.” I believe largely this is because NET doesn’t seem to have preseason anchoring based on past performance and recruiting/transfers which makes sense if you’re just trying to get metrics for what happens on the court this year but also cuts against some predictive value since we know those factors to have some value in that area. FAU started lower in metrics with some anchoring (Torvik, KP) and if FAU isnt the 10th best team in March it will be because they didn’t continue beating teams with the same efficiency. As sample size increases, the NET only gets better. Of course since we don’t know the exact formula for the NET it’s hard to specifically criticize it’s shortcomings
 
Also to be fair, the ACC is the worst rated conference of the major six on KP. Just as many teams (5) outside the top 100 as in the top 50 right now.
 
Rafi said:
Counterpoint - maybe the Big 10 wasn’t very good, since they put 0 of their 9 teams in the elite 8 (compared to 3 of 5 teams for the ACC).
Click to expand...

Counter to this counter: a sample size of 30+ overall games and 18-20 conference games to set the tournament largely has more predictive value than one off tournament games. Obviously the only thing that matters for “resume” purposes is winning tournament games in hindsight but if the goal of the committee is to pick the best at large teams based on a combination of metrics (predictive) and resume (what you actually did on paper) I think that’s different.

I don’t think Rutgers should have made the tournament but they had a pretty weird season
 
Rafi said:
And in retrospect, that was ridiculous given the ACC had the highest win percentage in the tournament for multibid leagues (and the big 10 was 6th and SEC 7th).

The ACC media talking heads should start harping on this nightly in February. I think the main talking point should be that the NET doesn’t account well for improvement as the season progresses.
Click to expand...

I don’t know that this is really true. NET weights all games equally so your 8 point home December win over a Q4 team that a 35th or so rated at large team “should” win by 20 counts the same as a 15 point win over a Q2 team that you “should” have only won by 6

ETA: the issue for Wake last year wasn’t their NET or that it didn’t allow for improvement over the season, it was the way the committee interprets that data that hurt Wake: not many Q1 games played, not many Q1 wins, not a difficult OOC schedule, not really holding bad losses against teams
 
Wakeforest22890 said:
I don’t know that this is really true. NET weights all games equally so your 8 point home December win over a Q4 team that a 35th or so rated at large team “should” win by 20 counts the same as a 15 point win over a Q2 team that you “should” have only won by 6

ETA: the issue for Wake last year wasn’t their NET or that it didn’t allow for improvement over the season, it was the way the committee interprets that data that hurt Wake: not many Q1 games played, not many Q1 wins, not a difficult OOC schedule, not really holding bad losses against teams
Click to expand...
If the committee is going to use computer rankings, then use them correctly and don't create new metrics based off the computer rankings (like the quad system). Last year Wake was 35 in KP and 42 in NET; Rutgers was 77 in KP and 80 in NET.
 
Rafi said:
If the committee is going to use computer rankings, then use them correctly and don't create new metrics based off the computer rankings (like the quad system). Last year Wake was 35 in KP and 42 in NET; Rutgers was 77 in KP and 80 in NET.
Click to expand...

I don’t disagree but obviously people don’t want “the computers” to set the field. That’s not a NET issue though, that’s a people issue
 
Rafi said:
If the committee is going to use computer rankings, then use them correctly and don't create new metrics based off the computer rankings (like the quad system). Last year Wake was 35 in KP and 42 in NET; Rutgers was 77 in KP and 80 in NET.
Click to expand...
This 1000%

Also I demonstrated last year that garbage time in the UL game, when the bench, including walkons, was outscored something like 11-2, cost us something like 11 places in KP at the time. Probably had a similar impact on NET.
 
Next 5 games... Need to go 4-1 to get to 14-5; 6-2.

Will move up significantly in KP (and elsewhere) if we:
Lose by single digits to UNC
Win by double digits over Louisville, FSU
Just win against BC & Clemson

Sounds easy, of course... But I think FSU is improving and they'll get a piece back for our game (Baba Miller). I also think the game @BC has the potential to be overlooked given it is a pretty quick turnaround after FSU and comes ahead of a big rematch against a good Clemson team.

After the next 5, it gets tougher again... An opportunity to win some games that we aren't expected to (which will also help with the computer rankings and Q1 bullshit).
 
I believe there is some debate about how to treat garbage time. I don’t think Torvik uses garbage time in parts of his calculation. KP has said he does include it and believes it smooths over the season/is negligible overall given the number of possessions impacted. I’d prefer the time discounted personally
 
Wakeforest22890 said:
I believe there is some debate about how to treat garbage time. I don’t think Torvik uses garbage time in parts of his calculation. KP has said he does include it and believes it smooths over the season/is negligible overall given the number of possessions impacted. I’d prefer the time discounted personally
Click to expand...

I think it is really difficult to account for that. Very subjective to decide when, exactly, garbage time starts.

Tend to agree w/ KP that you just include everything and assume the impact is low. Otherwise, I think you risk inserting biases into the data.
 
You must log in or register to reply here.
Back
Top