NET Nonsense (Part-20)

ib4ky_rivals111873

All-Conference
Dec 18, 2005
1,323
2,302
113
Sorry to start another NET bashing thread, BUT here goes:
(Wednesday games/ Thursday NET changes)

Some more crazy, unjustified, inexcusable movements within the "infallible" NET system!!!!
  • Syracuse moves from #67 up to #59 after winning AT #99 Pitt who drops to #108
  • Virginia moves from #52 up to #51 after winning AT #85 V.Tech who drops to #84
  • SFA (24-3/ 15-1) moves from #88 up to #82 after winning AT HOME versus #314 Northwestern St. (12-13/ 9-7) who drops to #315
  • Marquette moves from #26 up to #25 after winning AT HOME versus #59 Georgetown who moves up to #58
  • Arkansas moves from #45 up to #40 after winning AT HOME versus #65 Tennessee who drops to #70
  • Missouri moves from #89 up to #84 after winning AT #144 Vandy who drops to #153
  • Florida moves from #33 up to #31 after winning AT HOME versus #29 LSU who drops to #33
  • Notre Dame stays at #56 after winning AT #150 Boston Coll. who MOVES UP to #144
 
Feb 4, 2004
2,604
147
0
I am certainly not justifying because I thinks it is garbage.

But you need to put context around teams that were ahead of teams that moved up... For instance, for Syracuse, what happened to teams 59 - 67 ?

I keep seeing a lot of people talking about a certain team moving but no context about teams they moved in front of. If you win and all 6 teams ahead of you lose shouldn't you move up ?

I know it takes a lot of effort to figure it all out but if you are going to trash it be honest about all of it and let it implode on itself if it will.
 

struggler

All-Conference
Jan 23, 2013
1,358
3,283
0
The Net was composed to be a way for the NCAA to make the ACC look like the best conf. All it does is move the favorite's to a higher place so they can place them in better spot's when the terminate starts.. JMOO... Sorry brain fart, lol.. tournament...
Built into the algorithm... easily done and easily seen. Prime example is Duke.
 
Last edited:
May 27, 2007
31,941
25,080
113
I am certainly not justifying because I thinks it is garbage.

But you need to put context around teams that were ahead of teams that moved up... For instance, for Syracuse, what happened to teams 59 - 67 ?

I keep seeing a lot of people talking about a certain team moving but no context about teams they moved in front of. If you win and all 6 teams ahead of you lose shouldn't you move up ?

I know it takes a lot of effort to figure it all out but if you are going to trash it be honest about all of it and let it implode on itself if it will.

Right and not only that but the system has an efficiency component. The OP simply posted who won and who lost. What were the scores of those games?

It's not perfect by any means but there needs to be more context before you bash it.
 
Mar 13, 2004
14,745
12,925
0
I'm honestly not seeing, among this group of results, what's so unbelievable. The SFA one is the only one that looks a little odd to me, but they did win by 31, and at this point the things the teams you've played are doing can move you around, along with what the teams ranked near you are doing. ND beat BC by 1, so BC did better than the model expected them to do, hence moving up.

And, once again, I don't think the NET is infallible (lol), and I have never even once argued that the NET is in any way a good predictive model, or has any particular logic behind what criteria it uses. But it's pretty simple to understand why computer models produce the outputs they do.
 

Big John Stud

All-American
Jan 14, 2003
23,281
8,876
0
The Net was composed to be a way for the NCAA to make the ACC look like the best conf. All it does is move the favorite's to a higher place so they can place them in better spot's when the terminate starts.. JMOO... Sorry brain fart, lol.. tournament...
The ACC is the 6th rated conference in the NET but it was made to make the ACC look like the best conference? I guess the NET does suck.
 

struggler

All-Conference
Jan 23, 2013
1,358
3,283
0
Right and not only that but the system has an efficiency component. The OP simply posted who won and who lost. What were the scores of those games?

It's not perfect by any means but there needs to be more context before you bash it.
How about this for context - Duke loses 2 of it's last 3 games - to NET ranked 57 and NET ranked 103 and does not move...
 

Big_Blue79

All-Conference
Apr 2, 2004
52,487
2,147
0
Sorry to start another NET bashing thread, BUT here goes:
(Wednesday games/ Thursday NET changes)

Some more crazy, unjustified, inexcusable movements within the "infallible" NET system!!!!
  • Syracuse moves from #67 up to #59 after winning AT #99 Pitt who drops to #108
  • Virginia moves from #52 up to #51 after winning AT #85 V.Tech who drops to #84
  • SFA (24-3/ 15-1) moves from #88 up to #82 after winning AT HOME versus #314 Northwestern St. (12-13/ 9-7) who drops to #315
  • Marquette moves from #26 up to #25 after winning AT HOME versus #59 Georgetown who moves up to #58
  • Arkansas moves from #45 up to #40 after winning AT HOME versus #65 Tennessee who drops to #70
  • Missouri moves from #89 up to #84 after winning AT #144 Vandy who drops to #153
  • Florida moves from #33 up to #31 after winning AT HOME versus #29 LSU who drops to #33
  • Notre Dame stays at #56 after winning AT #150 Boston Coll. who MOVES UP to #144

  1. Who is saying the NET system is "infallible?" If no one is, why did you quote it?
  2. Movements in rankings hardly mean anything in any system like NET (or KenPom) because the teams are largely clustered such that a small change in RATING leads to a seemingly large change in RANKING.
  3. IIRC, NET has adjustments for SOS, meaning that a team's rating (and ranking) can move based on the results of games that the team did not play in. In other words, you appear to be missing many pieces of the puzzle and drawing conclusions.
  4. You're also not taking into consideration the score, which factors into the efficiency ratings and therefore the ratings. In other words, you appear to be missing many pieces of the puzzle and drawing conclusions.
That said, NET is far from perfect (let alone "infallible").
 
  • Like
Reactions: KyCat

ib4ky_rivals111873

All-Conference
Dec 18, 2005
1,323
2,302
113
I am certainly not justifying because I thinks it is garbage.

But you need to put context around teams that were ahead of teams that moved up... For instance, for Syracuse, what happened to teams 59 - 67 ?

I keep seeing a lot of people talking about a certain team moving but no context about teams they moved in front of. If you win and all 6 teams ahead of you lose shouldn't you move up ?

I know it takes a lot of effort to figure it all out but if you are going to trash it be honest about all of it and let it implode on itself if it will.

Previous Net Rank added to Wednesday 2/26 games impacting the current 2/27 NET Rankings:
  • Previous #58 lost on road, dropped to #61
  • Previous #59 lost on road by 21-POINTS, moved up to #58
  • Previous #60, DNP, still #60
  • Previous #61, lost at home, dropped to #74
  • Previous #62, DNP, still #62
  • Previous #63, won at home, dropped to #64
  • Previous #64, DNP, moved up to #63
  • Previous #65, lost on road, dropped to #70
  • Previous #66, DNP, moved up to #65
  • Previous #67, won on road, moved up to #59
  • Previous #68, DNP, still #68
As the season goes on and more "data" info is added into the NET metrics, the LARGE swings in movement should be reducing and ONLY happen when a lower rated team wins or wins big on the road at a higher rated team.

There is ZERO reason for Georgetown to move UP 1-spot after losing on road by 21-POINTS!!
There is ZERO reason for SFA to move up 6-spots passing one team that won and two teams who did not play Wednesday by beating #314 at home!! SFA was five spots BEHIND #83 Nevada and Nevada beat #87 and still stays at #83 while SFA gets rewarded for pounding on #314?!?!?!?!?!?!?!?!?!?!?!
  • Previous #66,
 

UKGrad93

Heisman
Jun 20, 2007
17,437
22,789
0
Sorry to start another NET bashing thread, BUT here goes:
(Wednesday games/ Thursday NET changes)

Some more crazy, unjustified, inexcusable movements within the "infallible" NET system!!!!
  • Syracuse moves from #67 up to #59 after winning AT #99 Pitt who drops to #108
  • Virginia moves from #52 up to #51 after winning AT #85 V.Tech who drops to #84
  • SFA (24-3/ 15-1) moves from #88 up to #82 after winning AT HOME versus #314 Northwestern St. (12-13/ 9-7) who drops to #315
  • Marquette moves from #26 up to #25 after winning AT HOME versus #59 Georgetown who moves up to #58
  • Arkansas moves from #45 up to #40 after winning AT HOME versus #65 Tennessee who drops to #70
  • Missouri moves from #89 up to #84 after winning AT #144 Vandy who drops to #153
  • Florida moves from #33 up to #31 after winning AT HOME versus #29 LSU who drops to #33
  • Notre Dame stays at #56 after winning AT #150 Boston Coll. who MOVES UP to #144
In all but one case the team that won went up in the NET and the team that lost went down. But a lot would depend on teams that these teams had played previously as well. My guess is that the calculation is similar to the RPI calculation: your win-loss record, your opponents win-loss record, their opponents win-loss record. So even if a team doesn't play, maybe their former opponent or opponent's opponent played.

Just look at at UK's NET, we are 6-3 in Quad 1 wins. UF was a Quad 2 win, but is UF wins enough games, they may move up to be a Quad 1 win.

I think it is more useful to look at the NET in terms of asking was it a good win or a bad loss. I think that is really what the NET is trying to answer.

Just scrolling thru the team sheets for the NET, I notice that UK is the highest NET of any team with a Quad 4 loss. The next is ETSU NET 38, then Bama NET 39. No team above us has a Quad 4 loss.
 

ib4ky_rivals111873

All-Conference
Dec 18, 2005
1,323
2,302
113
  1. Who is saying the NET system is "infallible?" If no one is, why did you quote it?
  2. Movements in rankings hardly mean anything in any system like NET (or KenPom) because the teams are largely clustered such that a small change in RATING leads to a seemingly large change in RANKING.
  3. IIRC, NET has adjustments for SOS, meaning that a team's rating (and ranking) can move based on the results of games that the team did not play in. In other words, you appear to be missing many pieces of the puzzle and drawing conclusions.
  4. You're also not taking into consideration the score, which factors into the efficiency ratings and therefore the ratings. In other words, you appear to be missing many pieces of the puzzle and drawing conclusions.
That said, NET is far from perfect (let alone "infallible").

Sorry, I forgot to put in my SARCASM emoji!
 

Cat_Incognito

Heisman
Dec 17, 2017
4,543
21,330
0
Yeah, some weirdo will probably add context and ruin the outrage fest.
Well, when someone finally does add some context, maybe they can also explain the formula that is being used to calculate the NET. So far none of us simpletons have seen it. Can you help us out?
 

Ineverplayedthegame

All-Conference
Aug 12, 2005
5,139
4,960
0
As I post in every one of these threads:
The NCAA wants a metric that no one understands so the selection committee can place whoever they want wherever they want and at whatever seed they want and use said metric to explain anything that really can’t be explained.
When people figure out this metric, they will switch to a different one.
 

ib4ky_rivals111873

All-Conference
Dec 18, 2005
1,323
2,302
113
So Duke DOES NOT get negatively impacted for EFFICIENT LOSSES but UK gets negatively impacted by INEFFICIENT WINS?????

Well, now it all makes sense!!
 

fatguy87

All-American
Oct 8, 2004
13,764
9,093
0
So Duke DOES NOT get negatively impacted for EFFICIENT LOSSES but UK gets negatively impacted by INEFFICIENT WINS?????

Well, now it all makes sense!!

You cannot make that claim because we don't know their actual rating. A team can suffer a substantial decrease in rating without dropping in ranking. UK in 2015, for example, lost to Wisconsin in the Final Four. Their rating dropped in Kenpom, but they didn't drop in ranking.
 

Big John Stud

All-American
Jan 14, 2003
23,281
8,876
0
Well, when someone finally does add some context, maybe they can also explain the formula that is being used to calculate the NET. So far none of us simpletons have seen it. Can you help us out?
Like KenPom no one knows the exact formula, you just have to figure it out as best you can from the ingredients the formula uses which are known. Also, if you're a simpleton, knowing the formula wouldn't help anyway.
 
  • Like
Reactions: KyCat

Sparkaces

All-American
Sep 19, 2012
6,157
6,289
0
Sorry to start another NET bashing thread, BUT here goes:
(Wednesday games/ Thursday NET changes)

Some more crazy, unjustified, inexcusable movements within the "infallible" NET system!!!!
  • Syracuse moves from #67 up to #59 after winning AT #99 Pitt who drops to #108
  • Virginia moves from #52 up to #51 after winning AT #85 V.Tech who drops to #84
  • SFA (24-3/ 15-1) moves from #88 up to #82 after winning AT HOME versus #314 Northwestern St. (12-13/ 9-7) who drops to #315
  • Marquette moves from #26 up to #25 after winning AT HOME versus #59 Georgetown who moves up to #58
  • Arkansas moves from #45 up to #40 after winning AT HOME versus #65 Tennessee who drops to #70
  • Missouri moves from #89 up to #84 after winning AT #144 Vandy who drops to #153
  • Florida moves from #33 up to #31 after winning AT HOME versus #29 LSU who drops to #33
  • Notre Dame stays at #56 after winning AT #150 Boston Coll. who MOVES UP to #144
That Florida number is very big, need them to be top 30!
 

ib4ky_rivals111873

All-Conference
Dec 18, 2005
1,323
2,302
113
The NET scoring margin is CAPPED at 10-points, no matter if the score differential is one point or fifty points. So, to me an area of the NET that seems to be weighted in the extreme is the magic O/D EFFICIENCIES.

So, if Team-A beats Team-B by 60-points for example, the point difference is capped at ten points (sportsmanship), but Team-A's O/D efficiencies are going to be great, hence maybe part of the reason it's hard to figure out the NET ranking bumps of some teams.
#88 wins by 31-points against #314 and moves up 6-spots to #82.
#67 wins by 23-points against #99 and moves up 8-spots to #59.
#206 wins by 20-points against #214 and moves up 7-spots to #199.
UK info affecting their NET since 12/16/2019 (NET debut):
Two best NET wins, both OT wins versus UL and Texas Tech only count as 1-point wins because of NET OT score cap.
Duke's 12-point OT loss at Wake only counts as 1-point NET loss.​
So UK has played 19-games since NET roll-out on 12/16 and 16-games have been played between 1-point (OT cap) and 10-points. Only 3-games have been played (won) by more than 10-points.
 
May 27, 2007
31,941
25,080
113
We've been through this like 100 times already.
There IS a difference between winning by 10 or 50.

The scoring margin component is capped.........at 10.
The efficiency component is uncapped.

Winning by 50 would count more than winning by 10.

We just don't know how much given the complete lack of transparency by the NCAA.
 
  • Like
Reactions: MoneyMuntz