IUs NET ranking?
- Thread starter IUNorth
- Start date
They never released the algorithm on how they do it. Unless I missed it somehow. So no way of knowing.
was told earlier in the season that some teams' NET was a result of carry over from last year, but that never seemed to self correct as this season progressed, so who knows?
Has Rutgers won a single game away from home this year?
Has Rutgers won a single game away from home this year?
there’s gotta be some sort of accepted formula for it. Can’t imagine I’m the first person to question it? If it’s truly just a behind the scenes calculation that no one can actually explain, the pundits and coaches would be going crazy over it...I would think.
I’ve not really heard anyone question it before. Not that no one has...just that it’s not widely questioned like RPI was when it died.
The Rutgers comp is actually starting to piss me off. Someone figure it out and explain it!
I’ve not really heard anyone question it before. Not that no one has...just that it’s not widely questioned like RPI was when it died.
The Rutgers comp is actually starting to piss me off. Someone figure it out and explain it!
Lunardi just updated...and he has IU as one of the last 4 teams in. They went down in his eyes after beating Minnesota.
Kind of what’s driving my question on the NET rankings. Any other metric or discussion and we’re safely in at this point.
So if the NET ends up keeping us out, if we say, lose to Wisconsin...beat Nebraska and lose 2nd round of BTT... that scenario, if Lunardis right in his assessment (big If I know)... but if he’s right...we’d be very vulnerable to being bumped. And the NET would have to be one of main reasons.
Kind of what’s driving my question on the NET rankings. Any other metric or discussion and we’re safely in at this point.
So if the NET ends up keeping us out, if we say, lose to Wisconsin...beat Nebraska and lose 2nd round of BTT... that scenario, if Lunardis right in his assessment (big If I know)... but if he’s right...we’d be very vulnerable to being bumped. And the NET would have to be one of main reasons.
Yeah he certainly doesn't think much of us or purdue: "All that Indiana and Purdue have really proven this year is an ability to lose regularly to good teams" "Indiana is alive by a fraction, and I'm not sure that fraction holds for either team without a win in the Big Ten Tournament."
He’s usually pretty accurate when it all shakes out.
I’d say that generally Purdue and IU have lost to more good teams than we’ve beaten. But the generalized comment is dumb. Florida State, Virginia, Michigan State are “good teams”. Penn State is a good team. Iowa is a good team.
I’d say that generally Purdue and IU have lost to more good teams than we’ve beaten. But the generalized comment is dumb. Florida State, Virginia, Michigan State are “good teams”. Penn State is a good team. Iowa is a good team.
Need to beat Wisconsin. I can't see us being left out with a 500 record in a highly regarded conference - regardless of what we do in the btt.
Lunardi is a dipshît. What other bubble team has as many good wins as us with no Q3/4 losses? And we have a top 15 schedule strength. He’s a moron.
But, morons make these decisions. A win vs Wisconsin and we are in easily. A loss and we better beat someone not named NEB/NW or we are probably out.
Same could be said for Texas Tech, Oklahoma, WVU, Florida, LSU, all of whom are safely in by Lunardi standards. Plus, those 3 B12 schools are all staring down .500 conference records or worse, just like IU
Just looked through Dayton and SDSUs schedules. Being conservative I’d put this IU team at 25-5 or so with the same schedules. Dayton is 2-2 against their only power 5 opponents. 0-2 against power 5 teams that’ll be in the tournament.
Even more fuel to my desire to figure out what this dang NET ranking is being derived from.
Even more fuel to my desire to figure out what this dang NET ranking is being derived from.
I'm pretty sure the NET rankings include a large dose of KenPom type efficiency numbers. In the old days a loss was a loss, but now getting hammered (like IU did) on the reg can really hurt the resume.
Maybe...but there’s suppose to be a 10 point cap on margin of victory and loss. And they’ve only had a couple of those thumpings as it is.
I think his general premise is based on a resistance to put 10 B1G teams in, especially since the ACC and SEC are going to get 4 each. With that preconceived notion in mind, Lunardi is trying to justify any reason to keep a B1G team out.
It's not so much the margin as it is the poor efficiency numbers that result from the margin. Anyway, I don't even know for sure what makes up the NET either.
He’s not really a dipshit though. He’s usually very accurate at the end. Just wouldn’t mind knowing what’s driving the NET and his thoughts. Because they don’t seem sensical to me at this point.
Go to www.bracketmatrix.com for a ranking of all the bracket predictors. Lunardi is ranked 55. You can also click on links to see where each currently has us seeded.
Syracuse got an 8 seed last year with the same efficiency rankings as IU (60th offense, 30th defense) and home losses to Old Dominion (#112) and Georgia Tech (#115). The Orange had wins over #11 seed Ohio St., #7 seed Louisville and #1 seed Duke.
IU's resume looks a lot better with the same efficiency numbers, so now I need to see if I can find the NET ranking of Syracuse last year.
Edit** Syracuse was #42 in NET heading into conference tournament week. They were 3-9 in Quad 1 (IU 5-9), they were 3-2 in Quad 2 (IU 4-2), they were 10-2 in Quad 3 (IU 4-0), they were 4-0 in Quad 4 (IU 6-0)
IU looks at least a seed line better imo, but last year maybe wasn't as deep with good teams as this year.
Last edited:
Just Google NET basketball rankings and search for the NCAA site. It details the metric. It says that early games have the same weight as later games. But then there is a contradictory statement that weights opponents. And the later obviously changes over time. So one might speculate that it is the opponent's strength at the time of the game.
Then there is the Offensive Efficiency minus the Defensive Efficiency which is explained. Of the other considerations, it ranks the amount of the winning scores up to 10 points difference. It considers home, away and neutral site games.
But look it up. There are a multitude of sites that give examples.
Nobody other than the creators do....as the formula isn't public.
That said.... it's just one metric the committee uses. They also use Kenpom (36th) and Sagarin (29th) along with traditional non analytics.
I thought they had released an explanation of the NET that I had read before, but didn't remember much other than the addition of kenpom etc
The primary importance of NET rankings is that they are used to define Quad wins/losses (IU is solid here).
After that, NET becomes one of several metrics that the committee reviews. Now, its hard to say how much weight is given to each individual metric/computer model (all viewed equally or some given more consideration than others), which is something likely to even vary among committee members. However, I would imagine that since on aggregate, most metrics have IU rated between 25-35, the NET rankings will likely be viewed as an outlier for IU.
I haven't really used the NET or really looked at it that closely. I had thought that the pundits were really talking up the use of the efficiency ratings when it was first introduced. Are they not really using them after all, or is it just not as big a part as I understood it to be.
Since it's not the efficiency ratings holding IU in the 50's, what do you see as causing the big separation between IU and PU/Rutgers in the NET?
I guess I'm going to need to do a little reading about the NET
It's basically an RPI with tweaks (including a half assed efficiency rating) and probably as stupid as the RPI was. So, it's winning percentage, opponents winning percentage, opponents opponents winning percentage with a .6 weight for home wins, 1.4 for road wins, and other things thrown in at degrees we don't know.I haven't really used the NET or really looked at it that closely. I had thought that the pundits were really talking up the use of the efficiency ratings when it was first introduced. Are they not really using them after all, or is it just not as big a part as I understood it to be.
Since it's not the efficiency ratings holding IU in the 50's, what do you see as causing the big separation between IU and PU/Rutgers in the NET?
I guess I'm going to need to do a little reading about the NET
I don't put much credence into gorilla math and that's what it is.
Kenpom has 8 B10 teams in his top 25, net has 4. Kenpom has OSU 8, Michigan 11, and PU 22. Those all seem way too high.
Kenpom seems to overvalue the B10, which would prop up IUs value. While NET seems like IU might be undervalued, the rest of the league looks much more reasonable.
Kenpom seems to overvalue the B10, which would prop up IUs value. While NET seems like IU might be undervalued, the rest of the league looks much more reasonable.
Everyone should probably not read it as a ranking ... that's not what it was intended to do.
If you like rankings based on possession efficiency, Kenpom. If you like rankings that incorporate SOS and margin of victory, Sagarin. If you like arbitrary opinion based ranking, either poll will do. Plus there's BPI, Torvik and many many others that are ranking systems. Net like RPI, is basically useless as a ranking system.
TeamRankings.com has the old RPI just for comparison. I made a little Big Ten chart comparing the RPI with KenPom and Sagarin. I think that it appears that the NET is heavily influenced by the predictive models. Here's the chart... what do you think? It looks like the old RPI doesn't think much of the Big 10.
TeamRankings.com has the old RPI just for comparison. I made a little Big Ten chart comparing the RPI with KenPom and Sagarin. I think that it appears that the NET is heavily influenced by the predictive models. Here's the chart... what do you think? It looks like the old RPI doesn't think much of the Big 10.
interesting that both net and Kp have IU as the worst BT team outside of nu/neb .,.. even having Minnesota (who we swept) above us.
I don’t get the love for the BT by Kp. The old RPI may be a tad bearish, net looks about right to me. I don’t get the osu love in any of the 3 rankings on the right.
That’s good, right? We want to play Kentucky. I know the board goes on about wanting to play Kentucky.
The RPI doesn't consider home court or location, or margin of victory or efficiency, while the other three do. Considering that ... NET should be closer to the other systems.TeamRankings.com has the old RPI just for comparison. I made a little Big Ten chart comparing the RPI with KenPom and Sagarin. I think that it appears that the NET is heavily influenced by the predictive models. Here's the chart... what do you think? It looks like the old RPI doesn't think much of the Big 10.
It's the very reason they added those half assed measures, so it would be. They added margin like Sagarin. They added Eff like Kenpom. Both systems have home court built in so they added home court weights.
I think you found that water is in fact wet?
"NET" rating. Like a net rating of a margin based system, eff based systems, location considered and the old RPI formula as a base.
Last edited:
It's very simple. The efficiency numbers vs average when compared to teams of other conferences vs average, are higher. His intent is not to rate the "best team", but to rate the highest net efficiency per 100 possessions. We rank so high because the conference had the best efficiency in non con play and by a considerable amount.
Last edited:
That makes sense. I get that it's still a flawed system, but, since it's much more correlated with KenPom/Sagarin now than it is to RPI, it's a big improvement over RPI alone, right? Or do you think that it's actually worse now because the committee didn't really pay that much attention to RPI anyway?
TeamRankings.com has the old RPI just for comparison. I made a little Big Ten chart comparing the RPI with KenPom and Sagarin. I think that it appears that the NET is heavily influenced by the predictive models. Here's the chart... what do you think? It looks like the old RPI doesn't think much of the Big 10.
Whats “interesting” there is that every other BT team on average is about 20 spots higher on the NET vs the old RPI.
Except us. That’s really strange.
Latest posts
-
Report: Arizona transfer big man Oumar Ballo to visit Indiana- Latest: Aloha Hoosier
-
-