Lamont unloads on the NET

18IsTheMan

Well-known member
Jan 19, 2022
14,146
12,144
113
Seems the gist is: we were not expected to be good and that expectation influences our NET, in spite of our results.

“I think the human element to it is, it’s six to one, half dozen the other. Someone told the world in September that we should be the last place finishing team in this conference. That’s what someone told the world. If you had never seen us play or practice, you looked on the casual fan, ‘Well, let me see what the SEC looks like. OK, Tennessee should be really good. South Carolina should be terrible. OK, they are terrible in my mind.’ The human element is affected by that. It just is. But the human element also allows you to use your eyeballs and overcome some numbers that say that Utah State, for example, is that much ahead of us in the NET. Some of them are crazy. Non-conference versus regular schedule matters until the season is over. At this point, who cares what our non-conference schedule was. We had to go to Alabama. We had to go to Tennessee. We had to play Tennessee back here. We had to go to Arkansas. We had Kentucky at home. We had Texas A&M. I can guarantee you New Mexico has none of that based on the league that they are in. They can’t be penalized for their league. I am not saying that. They can’t manufacture in any way shape or form the types of games that we have had to play on a consistent basis in this conference. Let’s stop talking about the non-conference versus the conference. At this point in the game, that is great in December. You talk about non-conference versus conference. At this point, let’s look at what did you do over the course of 31 basketball games. Who did you play and how did you fare? I think we lose that based on the way the metrics of this thing are. I think the human element has to be there so you can override because some of the numbers just don’t make sense at this point.

“If you look, not to go on a tangent, but if you look at the NET. Find any high major team that is anywhere around where we are ranked in the NET that has anywhere near the types of losses we do. We have six total losses. I don’t know this for a fact. I haven’t looked at it today. We have six total losses in a high major league. One of, if not the best. One of the best conferences in the country. Go into the NET and see who else, high major, has a similar amount of losses and I guarantee they are in the top 15 of the NET. The teams that will be around us that are high majors will have 12 losses. I don’t care who you play. At some point, you have to win the games. … You think about it. We are the 15th ranked team in the country. We have been ranked for the last six weeks. Someone beats us at their place, and it is a Quad 2. It is a Quad 2 win or loss. That doesn’t make any sense. While, Michigan State, 12, 13 losses. Whatever it is that they have at this point. If you play Michigan State at your own place, you are getting a Quad 1 win. That is wild to me. So, then everyone else feeds off of that. … Iowa gets a Quad 1 win because they beat Michigan State. Well, Michigan State is not the fourth team in the country like they were predicted to be. Then, they never really deteriorated that much based on that first initial ranking.”


p.s. before anyone whines and complains and gets their panties in a wad about someone talking about the NET rankings, this is simply a direct quote from Paris, so take it up with him.
 

Deleted11512

Well-known member
Feb 2, 2023
4,985
3,954
113
He was clearly riled up in the interview when talking about it. He has a point. At this point in the year why does non con matter? We won them all except a Q1 L to clem. Then went 13-5 in the toughest bball conference in the country. I think it's good to have some computer generated metrics. But there are clearly issues with this one that they need to address. Maybe they should weight Jan/Feb games more heavily than Nov/Dec games?
 
Last edited:

Prestonyte

Well-known member
Jun 1, 2022
5,261
5,214
113
I agree with LP and I don't understand why other coaches are not right there with him on this.
NET is clearly not performance based except at the top and the bottom of the rankings.
 

18IsTheMan

Well-known member
Jan 19, 2022
14,146
12,144
113
He was clearly riled up in the interview when talking about it. He has a point. At this point in the year why does non con matter? We won them all except a Q1 L to clem. Then went 14-4 in the toughest bball conference in the country. I think it's good to have some computer generated metrics. But there are clearly issues with this one that they need to address. Maybe they should weight Jan/Feb games more heavily than Nov/Dec games?

I could see the non-con being an issue if you got into conference play and fell on your face, but if you go 13-5 in arguably the toughest conference in the nation, then it renders the non-con schedule moot.
 

18IsTheMan

Well-known member
Jan 19, 2022
14,146
12,144
113
Case in point:

A&M, at 18-13 (9-9) is ranked 2 spots ahead of us in the NET.

Mississippi St. at 19-12 (8-10) is ranked 8 spots ahead of us.

Texas at 20-12 (9-9) is ranked 24 spots ahead of us. They have an 8-10 Quad 1/2 record. Our is 11-4.

It simply makes no sense at all.
 

Deleted11512

Well-known member
Feb 2, 2023
4,985
3,954
113
Case in point:

A&M, at 18-13 (9-9) is ranked 2 spots ahead of us in the NET.

Mississippi St. at 19-12 (8-10) is ranked 8 spots ahead of us.

Texas at 20-12 (9-9) is ranked 24 spots ahead of us. They have an 8-10 Quad 1/2 record. Our is 11-4.

It simply makes no sense at all.
We are 1 of 5 teams in the country with a 70+% win percentage against Q1/Q2 competition. The other 4 teams are vying for a #1 seed.
 

18IsTheMan

Well-known member
Jan 19, 2022
14,146
12,144
113
We are 1 of 5 teams in the country with a 70+% win percentage against Q1/Q2 competition. The other 4 teams are vying for a #1 seed.

Yep. Of course, I wouldn't argue that we should be nearly a 1 seed, but goodness we should be higher than the high 40s in the NET.
 
Last edited:
  • Like
Reactions: LonghornsGamecocks

Viennacock

Well-known member
Jan 21, 2022
2,147
2,126
113
He's doing the right thing by jockying for position. I said it earlier this year, if you are not a name team, you are going to be penalized. Win anyways!
 

Prestonyte

Well-known member
Jun 1, 2022
5,261
5,214
113
Dawn has created stature and history on the women's side and hopefully LP will do the same with time. Then we will get respect.
 

Deleted11512

Well-known member
Feb 2, 2023
4,985
3,954
113
He's doing the right thing by jockying for position. I said it earlier this year, if you are not a name team, you are going to be penalized. Win anyways!
Not necessarily. Clem has been a top 20-25 NET team all year. They dropped to 36 after last night. St. Mary's is currently 15. IA St is 9. The problem with our net is close games against Q3/Q4 teams. The net assumes good teams will blow those teams out. VMI (358), DePaul (320), ECU (186), Winthrop (180), CSU (314) were all teams we should have been able to blow out, but they were all 10 point or less Ws where we didn't get out of the 70s. Only scored 68 against ECU. So, the assumption is we're not all that good, and more lucky in our wins against better teams. Gladly there are humans in the room as well.
 
  • Like
Reactions: LonghornsGamecocks

31 Flavors

Member
Feb 9, 2024
44
34
18
"I don’t care who you play. At some point, you have to win the games."

^^^friggin exactly.

Winning is the whole point of the game. If the NET algorithm is to be legit as a tool for picking the field, it should start compounding the hurt after a team has lost a certain number of games, no matter how tough their schedule, and compounding the reward after a certain number of wins, no matter how light the schedule.

IMO, on a human level, (not that this affects the Gamecocks this year), in this hypothetical where an at-large invite is in the balance:
- The 13-loss team from the "high-power" conference has failed to prove they can succeed in the tournament.
- The 5-loss team from the "mid-power" conference has not failed to prove they can succeed.
 
  • Like
Reactions: Bubba Fett

18IsTheMan

Well-known member
Jan 19, 2022
14,146
12,144
113
Not necessarily. Clem has been a top 20-25 NET team all year. They dropped to 36 after last night. St. Mary's is currently 15. IA St is 9. The problem with our net is close games against Q3/Q4 teams. The net assumes good teams will blow those teams out. VMI (358), DePaul (320), ECU (186), Winthrop (180), CSU (314) were all teams we should have been able to blow out, but they were all 10 point or less Ws where we didn't get out of the 70s. Only scored 68 against ECU. So, the assumption is we're not all that good, and more lucky in our wins against better teams. Gladly there are humans in the room as well.

I do think there would be a much stronger positive perception of our team if we were consistently winning games 98-92 instead of 75-69.
 

LonghornsGamecocks

Active member
Feb 24, 2024
870
1,191
63
I could see the non-con being an issue if you got into conference play and fell on your face, but if you go 13-5 in arguably the toughest conference in the nation, then it renders the non-con schedule moot.
I can also see non-con being an issue if you go play a gauntlet and beat several Top 20 or Top 10 OOC teams. A rightful feather in the cap, so to speak.
 

LonghornsGamecocks

Active member
Feb 24, 2024
870
1,191
63
Seems the gist is: we were not expected to be good and that expectation influences our NET, in spite of our results.

“I think the human element to it is, it’s six to one, half dozen the other. Someone told the world in September that we should be the last place finishing team in this conference. That’s what someone told the world."
I don't know the intricacies of the NET calculation, but my understanding is that it's supposed to mitigate human bias by being completely data driven (granted, humans have to help build the algorithms and the algorithm of the algorithms).

I agree that our NET ranking seems curiously low, and others seem curiously high. From a data science best practices POV, they should be re-training and tuning the models and the models of models between every season - I have no idea if this is actually happening.
 
  • Like
Reactions: DrMickeySC

Deleted11512

Well-known member
Feb 2, 2023
4,985
3,954
113
I do think there would be a much stronger positive perception of our team if we were consistently winning games 98-92 instead of 75-69.
Agree. Those are some pathetic teams we squeaked by. Luckily we have one of the best records in the country against Q1/Q2 teams to go along with it. If we won them all the way we did against Upsate, Elon, or FAMU we wouldn’t be talking about it.
 

Deleted11512

Well-known member
Feb 2, 2023
4,985
3,954
113
I don't know the intricacies of the NET calculation, but my understanding is that it's supposed to mitigate human bias by being completely data driven (granted, humans have to help build the algorithms and the algorithm of the algorithms).

I agree that our NET ranking seems curiously low, and others seem curiously high. From a data science best practices POV, they should be re-training and tuning the models and the models of models between every season - I have no idea if this is actually happening.
The bottom line is that really good teams are supposed to blow out really bad teams. Or at least that’s what the logic is. So if everyone is beating CSU by 30, and we beat them by 4, the data suggests we’re not very good. Then when we beat a top team by 4, the logic says “that’s luck bc they barely beat CSU.” I get it to a degree. How many times have we seen an SEC drop in rankings for losing to Bama, and get jumped by an ACC team that beat Wake? Is the ACC team really better or are we penalizing the SEC team for playing elite competition? The net tries to take that volatility out.
 
  • Like
Reactions: LonghornsGamecocks

31 Flavors

Member
Feb 9, 2024
44
34
18
I do think there would be a much stronger positive perception of our team if we were consistently winning games 98-92 instead of 75-69.
Yes, that would be a bias against teams solely based on playing style. Also worth noting is that in all of those close OOC games, CMB was either out with mono or played limited minutes while recovering, which would also make for a flawed perception.
 

Tngamecock

Well-known member
Jan 22, 2022
1,703
1,809
113
Seems the gist is: we were not expected to be good and that expectation influences our NET, in spite of our results.

“I think the human element to it is, it’s six to one, half dozen the other. Someone told the world in September that we should be the last place finishing team in this conference. That’s what someone told the world. If you had never seen us play or practice, you looked on the casual fan, ‘Well, let me see what the SEC looks like. OK, Tennessee should be really good. South Carolina should be terrible. OK, they are terrible in my mind.’ The human element is affected by that. It just is. But the human element also allows you to use your eyeballs and overcome some numbers that say that Utah State, for example, is that much ahead of us in the NET. Some of them are crazy. Non-conference versus regular schedule matters until the season is over. At this point, who cares what our non-conference schedule was. We had to go to Alabama. We had to go to Tennessee. We had to play Tennessee back here. We had to go to Arkansas. We had Kentucky at home. We had Texas A&M. I can guarantee you New Mexico has none of that based on the league that they are in. They can’t be penalized for their league. I am not saying that. They can’t manufacture in any way shape or form the types of games that we have had to play on a consistent basis in this conference. Let’s stop talking about the non-conference versus the conference. At this point in the game, that is great in December. You talk about non-conference versus conference. At this point, let’s look at what did you do over the course of 31 basketball games. Who did you play and how did you fare? I think we lose that based on the way the metrics of this thing are. I think the human element has to be there so you can override because some of the numbers just don’t make sense at this point.

“If you look, not to go on a tangent, but if you look at the NET. Find any high major team that is anywhere around where we are ranked in the NET that has anywhere near the types of losses we do. We have six total losses. I don’t know this for a fact. I haven’t looked at it today. We have six total losses in a high major league. One of, if not the best. One of the best conferences in the country. Go into the NET and see who else, high major, has a similar amount of losses and I guarantee they are in the top 15 of the NET. The teams that will be around us that are high majors will have 12 losses. I don’t care who you play. At some point, you have to win the games. … You think about it. We are the 15th ranked team in the country. We have been ranked for the last six weeks. Someone beats us at their place, and it is a Quad 2. It is a Quad 2 win or loss. That doesn’t make any sense. While, Michigan State, 12, 13 losses. Whatever it is that they have at this point. If you play Michigan State at your own place, you are getting a Quad 1 win. That is wild to me. So, then everyone else feeds off of that. … Iowa gets a Quad 1 win because they beat Michigan State. Well, Michigan State is not the fourth team in the country like they were predicted to be. Then, they never really deteriorated that much based on that first initial ranking.”


p.s. before anyone whines and complains and gets their panties in a wad about someone talking about the NET rankings, this is simply a direct quote from Paris, so take it up with him.
With no paragraphs, I found I am now blind reading this on my phone. Like picking grits😄

I’m just messing wit ya……it was a good read. LP is a smart man.
 
  • Like
Reactions: 18IsTheMan

RUMMENIGGE

Active member
Jan 31, 2022
757
468
63
I do not need a Slide Scale to tell me that South Carolina is one of the best 68 teams this season.
 

Rogue Cock

Joined Sep 11, 2000
Jan 22, 2022
10,019
14,906
113
I could see the non-con being an issue if you got into conference play and fell on your face, but if you go 13-5 in arguably the toughest conference in the nation, then it renders the non-con schedule moot.
And, yet, there is Clemson who kinds fell on it's face in conference play after a good OOC performance, and are still comparitively highly ranked in the NET. It is ludicrous.
 
  • Like
Reactions: Bubba Fett