NET and KenPom Tracker 23-24 (SU = 84 3/9/24)

[wiseman3]

We dropped to 87 after the games last night, with Rutgers and Minnesota passing us.

Minnesota went up 8 places from 88 to 80 after losing to Purdue...I hate this system. This just shows that its all about the spread/winning or losing margin.

 

[rocfan44]

This is the dumbest metric ever created!

 

[pokercuse08]

This just shows that its all about the spread/winning or losing margin.

This really opens the door to a lot of perverse financial incentives for conferences and, if they succumb to the temptation, their officials. Like, the fact that making a few bad calls to help a bubble team lose by a smaller margin as a big underdog can help them a lot in a very important metric opens the door to some dubious opportunities - assuming they'd rather have more teams in than fewer, and sacrifice a bit of seeding as a result.

The line was Purdue -16.5, result Purdue by 8, and Minnesota was on the road so it's weighted more in the NET. I didn't watch a second of it, so I'm not saying this happened, but would anyone even notice/make a big deal about a couple bad calls that take the margin from 12 to 8? Doubles the strength of their result ATS though.

 

[jncuse]

I don't think beating the expected margin by 8 (even on the road) would typically move teams up by 6 spots like Minnesota did last night.. normally. It could just be a small gapping that happens at various parts of the ranking. It also probably suggests that the gap between Minny at #80 and us at #87 is probably very small right now.

Part of the problem with NET vs KP is there is "No score" associated with your ranking, so you can't see how tight certain ranges are. . I'll give you an example with KP.

#51. 13,96
#56. 13.88
Gap of .08 points between 6 teams (2 points over the course of a season)

#57. 13.69
#62. 12.58
Gap of 1.11 points between 6 teams (28 points over the course of a season)

So you basically have 6 teams that are basically all tied with each other (#51-56), so the slightest of things could jump team from #56 to #51. Could be as simple as you winning by 3 points more than expected, or an opponent doing really well.

But then team #62 could win by 20 points more than expected, and they won't even catch #57.

This isn't a defence of NET - I have already proposed the change above. But its sometimes hard to compare a jump of 5-6 spots, because there are different gaps that we just don't see.

 

[jncuse]

This really opens the door to a lot of perverse financial incentives for conferences and, if they succumb to the temptation, their officials. Like, the fact that making a few bad calls to help a bubble team lose by a smaller margin as a big underdog can help them a lot in a very important metric opens the door to some dubious opportunities - assuming they'd rather have more teams in than fewer, and sacrifice a bit of seeding as a result.

The line was Purdue -16.5, result Purdue by 8, and Minnesota was on the road so it's weighted more in the NET. I didn't watch a second of it, so I'm not saying this happened, but would anyone even notice/make a big deal about a couple bad calls that take the margin from 12 to 8? Doubles the strength of their result ATS though.

Extra 4 points doesn't typically moved much (even if they use the same 1.4 ratio for road as RPI). Spread over 26 games played its not much.

You would have to do it over and over and over, to get something even small out of it (as net has similar principles to KP) and you can really only do it with one or two elite teams.

The biggest abuse by far is still the scheduling of really bad P4 teams (sub 300) where you can control the narrative quite easily.

 

[jncuse]

We dropped to 87 after the games last night, with Rutgers and Minnesota passing us.

Minnesota went up 8 places from 88 to 80 after losing to Purdue...I hate this system. This just shows that its all about the spread/winning or losing margin.

As noted in my post above, the gap between #80 and #87 has to be very small (unfortunately its not visible) because 8 points over expectation (even on the road), especially a loss, will not normally move things this much.

My guess, is that if we win at Georgia Tech tomorrow we probably get to that 80 spot, or even high 70's, based on the fact that things have to be bunched in that range.

 

[kirbivore]

This really opens the door to a lot of perverse financial incentives for conferences and, if they succumb to the temptation, their officials. Like, the fact that making a few bad calls to help a bubble team lose by a smaller margin as a big underdog can help them a lot in a very important metric opens the door to some dubious opportunities - assuming they'd rather have more teams in than fewer, and sacrifice a bit of seeding as a result.

The line was Purdue -16.5, result Purdue by 8, and Minnesota was on the road so it's weighted more in the NET. I didn't watch a second of it, so I'm not saying this happened, but would anyone even notice/make a big deal about a couple bad calls that take the margin from 12 to 8? Doubles the strength of their result ATS though.

It was on B1G network which I don't have. I just followed the 2nd half game-cast as I had Purdue team total over 81.5. The Boilers were down 8 at the half.

 

[Buddah69]

Any info on quad wins and losses

 

[hoopsupstate]

Unfortunately, we beat both Pitt and NCS when they were quad 1 wins, but after we beat them, they fell to quad 2.

 

[Consigliere]

Unfortunately, we beat both Pitt and NCS when they were quad 1 wins, but after we beat them, they fell to quad 2.

Pitt on the road is a solid Q1 with a current NET of 47. Despite losing NC State held at 75 according to Warren Nolan. Ours was updated, so I assume NCSU is valid. And of course UNC is our third Q1.
This morning’s update has us 3-7 in Q1 wins.

 

[Consigliere]

Not too long ago RPI was the be all, end all for tournament selection. RPI is based solely on wins and losses and strength of schedule. Syracuse is currently 32 in RPI.
NET is based largely on advanced analytics including offensive and defensive efficiency ratings. Syracuse is currently 90 in NET.

Cuse is 3-7 in Quad 1 wins.
Wake is 1-6 and has a NET of 40
Pitt is 2-5 and has a NET of 47
NC State is 1-6 and has a net of 75.

And NET is now the preferred metric?

And check out some of the sheets from other conferences.
RPI led teams to schedule up as opponents W-L and SOS were paramount. NET doesn't reward that strategy. So you are better off playing and soundly beating mid majors than competing against teams from power conferences.

If RPI needed tweaking, fine. But the disparity in our ranking shows that there is no common ground in the two systems. This really needs to be addressed and made more transparent

Not saying we deserve a bid with our current resume, but the only thing truly keeping us out of the discussion is our NET. Using the old RPI system we are on the good side of the bubble.

 

[NineOneSeven]

Not too long ago RPI was the be all, end all for tournament selection. RPI is based solely on wins and losses and strength of schedule. Syracuse is currently 32 in RPI.
NET is based largely on advanced analytics including offensive and defensive efficiency ratings. Syracuse is currently 90 in NET.

Cuse is 3-7 in Quad 1 wins.
Wake is 1-6 and has a NET of 40
Pitt is 2-5 and has a NET of 47
NC State is 1-6 and has a net of 75.

And NET is now the preferred metric?

And check out some of the sheets from other conferences.
RPI led teams to schedule up as opponents W-L and SOS were paramount. NET doesn't reward that strategy. So you are better off playing and soundly beating mid majors than competing against teams from power conferences.

If RPI needed tweaking, fine. But the disparity in our ranking shows that there is no common ground in the two systems. This really needs to be addressed and made more transparent

Not saying we deserve a bid with our current resume, but the only thing truly keeping us out of the discussion is our NET. Using the old RPI system we are on the good side of the bubble.

Many good points, but there are always some outliers. Seems we may be one this year. If we had beaten GT and FSU I wonder what the NET would have been right now.

 

[EnviroSciGuy]

At the end of the day, judging a team by who it has beaten (and lost to) should be the be-all, end-all. The RPI was all-business in that regard. It seems like NET is really punishing this young team for its inconsistency and growing pains. Those advanced metrics are also based on who you play... kind of like the most important quality: winning or losing. RPI wasn't perfect but I think they've gotten way too pretty with their choice of analytics.

 

[Fly Rodder]

It seems to be a bad metric after wins, but after losses to say a 10-15 team, it's crickets in here.

The metrics aren't to blame for Syracuse not getting last four in consideration. And frankly, Syracuse shouldn't ever be striving for last four in consideration.

 

[SU94]

Many good points, but there are always some outliers. Seems we may be one this year. If we had beaten GT and FSU I wonder what the NET would have been right now.

Higher, but not so high. The point is, NET unduly measures margins of victory (and defeat) and rewards playing a mediocre schedule with few challenge games, which is bad for the sport as a whole. The B12 has manipulated the NET to advance the narrative that they are the preeminent hoops conference (last year it was true, not as much this year).

The NCAA Tournament Committee does not have to use NET, it is not some fundamental law, but it'll very likely take some time for them to change or evolve past it. Why? Because they created it and there is (unfortunately) pride in ownership. They paid for it, they will use it and they are therefore biased toward it.

The NET, explained: NCAA adopts new college basketball ranking

The NCAA has developed the NCAA Evaluation Tool (NET) a ranking system that will replace the RPI as the primary sorting tool used to evaluate teams.

www.ncaa.com

 

[orangenauburn]

At this point just have to win out regular season and then win three in ACCT.

Go 7-1 to finish season would be 24-11

Hard to keep out of tournament with that.

 

[rocfan44]

Currently we only moved up to 84 in the NET. NCST dropped to 80, so last night’s win was as expected down graded. This is such a broken system.

 

[OrangeXtreme]

 

[orange79]

Pitt on the road is a solid Q1 with a current NET of 47. Despite losing NC State held at 75 according to Warren Nolan. Ours was updated, so I assume NCSU is valid. And of course UNC is our third Q1.
This morning’s update has us 3-7 in Q1 wins.

Actually, Pitt dropped to 56 after getting crushed by Wake yesterday. That dropped them from a Q1 to a Q2.

SU currently:

Q1: 2-7
Q2: 3-1
Q3: 7-2
Q4: 4-0

 

[Cuse House]

ACC is cannibalizing itself.

 

[SUMinistry]

Didn't want to try to look through 29 pages of posts, so sorry if this was already answered, but how does the Quad 1 ranking work for neutral site games (i.e. Oregon)? How high do they need to be ranked?

 

[OptimusOtto]

Actually, Pitt dropped to 56 after getting crushed by Wake yesterday. That dropped them from a Q1 to a Q2.

SU currently:

Q1: 2-7
Q2: 3-1
Q3: 7-2
Q4: 4-0

Both Pitt games are Q1/Q2 - (road win Q1 = 1-75/ home win Q2 = 30-75). NC State dropping to 80 knocked that both those wins down a tier. Need to root for NC State to finish strong and end up in the top 75.

 

[OptimusOtto]

Didn't want to try to look through 29 pages of posts, so sorry if this was already answered, but how does the Quad 1 ranking work for neutral site games (i.e. Oregon)? How high do they need to be ranked?

Top 50 in the net

 

[OptimusOtto]

Does anyone know recent outliers in terms of making the tournament as an at-large with a high net? My assumption is high 50s/low 60s is pushing it but I’m sure there are a few exceptions.

 

[Niastri]

I don't think beating the expected margin by 8 (even on the road) would typically move teams up by 6 spots like Minnesota did last night.. normally. It could just be a small gapping that happens at various parts of the ranking. It also probably suggests that the gap between Minny at #80 and us at #87 is probably very small right now.

Part of the problem with NET vs KP is there is "No score" associated with your ranking, so you can't see how tight certain ranges are. . I'll give you an example with KP.

#51. 13,96
#56. 13.88
Gap of .08 points between 6 teams (2 points over the course of a season)

#57. 13.69
#62. 12.58
Gap of 1.11 points between 6 teams (28 points over the course of a season)

So you basically have 6 teams that are basically all tied with each other (#51-56), so the slightest of things could jump team from #56 to #51. Could be as simple as you winning by 3 points more than expected, or an opponent doing really well.

But then team #62 could win by 20 points more than expected, and they won't even catch #57.

This isn't a defence of NET - I have already proposed the change above. But its sometimes hard to compare a jump of 5-6 spots, because there are different gaps that we just don't see.

It sucks to think that a couple of meaningless threes shot at the final buzzer to drop the margin from 8 to 5 points could ultimately determine which team makes the last four into the tournament.

Also, if you don't run up the score as a coach right now, you're hurting your team. Systematic bad sportsmanship. Smh