19-10 (10-8) against the 20th hardest schedule

[GoSU96]

If you are citing KenPom's "luck" data point, you have a fundamental misunderstanding of how that is used.

Luck in KenPom isn't factored into any calculation of his ranking. It just ranks a team exceeding or not living up to his ranking's expectations.

Well that is both arrogant and stupid, a historicaly bad combination

 

[tptcuse]

If you took away the GT loss, I'd be much more vocal about complaining.

With it, I really feel like we need to win the last 2, then if we go 1-1 in the ACC tourney (or lose to Clemson and go 2-1 in the ACCT), I'll join in with the complaining

This is where I am, but I have to admit it is nice to be in the conversation regardless.

 

[jncuse]

Its one of the inherent flaws in the NET because it doesn't take game winning strategy into consideration. What i mean by that is during end of game situations come coaches know the analytics of time vs points scored (JB was very good at this). If you're up 10 pts with 3:30 seconds left you eat the clock on offense and, subsequently, take lower % shots at the end of the shot clock. The opposing team may/may not come down and score, but they are usually taking higher % shots (and the defense doesn't want to give up 3s or and-1s so the defense gives up better 2pt looks) either way you do the same thing next possession. Often times the team thats down will cut into the lead but the goal ultimately is to WIN. Weather thats by 4 pts or 12pts doesn't matter to the coach. NET on the other hand does care and that is an issue. Now also factor in end of game situations when the game is out reach (like last night). NET wants us to score on those last possessions and run up the score, this is bad basketball etiquette and coaches hate that. It also results in higher risks of fights and other issues that are bad for the sport. JMHO

It's a flaw of NET, and could lead to poor sportsmanship for sure. But its not at all why we are getting blasted by NET.

A lot of those minor swings of 2 to 4 points will offset and won't accumulate to much over the course of a season.

What does accumulate is when you have 10+ games that you underperform by 10 points or more (either in wins or losses)... and there is very few positive overperformances to offset the bad. Those things add up to something much more signifiant.

Not that this is great either, but selections are ultimately made on W/L's, not individual NET. So just win the games is a fine strategy to pursue in closer games.

 

[Janner]

It's a flaw of NET, and could lead to poor sportsmanship for sure. But its not at all why we are getting blasted by NET.

A lot of those minor swings of 2 to 4 points will offset and won't accumulate to much over the course of a season.

What does accumulate is when you have 10+ games that you underperform by 10 points or more (either in wins or losses)... and there is very few positive overperformances to offset the bad. Those things add up to something much more signifiant.

Not that this is great either, but selections are ultimately made on W/L's, not individual NET. So just win the games is a fine strategy to pursue in closer games.

I am not by any means saying the NET is a total garbage system of measurement, I actually think it has its merits. But there are clear issues, as I stated in my previous post, that run contrary to both an accurate analysis and spirit of the game. If we keep the NET as it is i will guarantee at some point in the future there will be an ugly incident between teams that is the result of a coach telling his players to score more points because they need it to help their chances of making the NCAA tournament. That will be a problem and a bad look. I for one really like your idea of using an average of the RPI and the NET as the metric for the reasons you stated in the other thread because they help to offset each others weakness and reduces the reliance on margin of victory. And it would be very easy to implement. Both systems have been refined to a degree and the data/formula is being generated currently The work is already done, so to speak. Will it be perfect? No. But it gets us closer, which is what is important.

 

[Millhouse]

SRS Calculation Details | Sports-Reference.com

SRS Calculation Details

www.sports-reference.com

I think this applies to net or any other other points
margin based calc

• It is a predictive system rather than a retrodictive system - this is a very important distinction. You can use these ratings to answer the question: which team is stronger? I.e. which team is more likely to win a game tomorrow? Or you can use them to answer the question: which of these teams accomplished more in the past? Some systems answer the first questions more accurately; they are called predictive systems. Others answer the latter question more accurately; they are called retrodictive systems. As it turns out, this is a pretty good predictive system. For the reasons described below, it is not a good retrodictive system.

 

[Bulldog04]

JB stated (I believe on Dan Patrick show) that the committee used to always look at who got better as the season went on for a big determining factor on who deserves to be in? This team has made leaps and bounds to finish the season.

Also SU has 12 double digit margin of victory wins out of their 19 wins. That’s a lot!

Win the next 2 and eliminate the debate fellas

 

[kcsu]

The NET is beyond flawed. It needs to be modified.

 

[jncuse]

JB stated (I believe on Dan Patrick show) that the committee used to always look at who got better as the season went on for a big determining factor on who deserves to be in? This team has made leaps and bounds to finish the season.

Also SU has 12 double digit margin of victory wins out of their 19 wins. That’s a lot!

Win the next 2 and eliminate the debate fellas

I would have to disagree with JB's opinion here.

A core principle of the selection game is "Fully Body of Work", the underlying principle being that all games are valued the same no matter when they are played... that is a good win (or bad loss) is valued the same whether it happened in November, January, or late February.

Not only was the last 10 games metric dropped about 10-12 years ago, the committee goes out of its way year after year in their interviews to adamantly state that selection is not based on this principle, and that its not acceptable for them to select teams based on last 5 or last 10 games..

And its not just word salad from them. We have constantly seen teams with good ends to the season (like Texas A&M) not get in, and teams like Oklahoma recently get in because of huge wins in November. Remember the disaster we had to close the 2016 season? I think we lost 5 of 6.

Not saying going that way is the right or wrong way of doing things, but it is the way they do things and their historical selections back it up.

 

[RF2044]

How do you leave a team out based on average point margin?

This is crazy.

They give auto berths to terrible teams just because they happen to win a couple lucky games

And team that wins more than the math predicts they ought to is ignored.

Predictions are great, these ratings are good for it. But tournament spots aren't predictions, if some great player is hurt all year and comes back at the end, they'd be predicted to be good but the fact is they weren't.

They play a very hard schedule and they won enough.

Simple -- because mathematic inputs establish the weighting of evaluative criteria.

And when you emphasize certain inputs over others, that influences the calculation / score / ranking.

 

[Dinkyspond]

I hate to say this, and please roast me later if I'm wrong...but I have a good feeling we are beating both lville and Clemson. Despite the NCst and ND second halfs, this team has looked different recently. Better ball movement, shooting, a pep in their step if you will. Hell, even JT has been playing his best all year and getting involved. I think we'll win out and win at least 1 in the acct.

Hall, Shefflein and Godfrey make them so big and we counter with a game but undersized Maliq and we never play well at Littlejohn. But I like your thinking, the way we've been playing gives us a better than puncher's chance.

 

[Bulldog04]

I just saw Andy Katz bracket. I cannot believe some of the teams he has in and where he has them. Has Mich state as 8 seed. I can’t believe what I just saw

 

[bnoro]

Calling that "luck" is so messed up. "Could my rankings be wrong, nope, they're just lucky."

He could call it standard deviations and you'd still complain about it.

 

[Millhouse]

Simple -- because mathematic inputs establish the weighting of evaluative criteria.

And when you emphasize certain inputs over others, that influences the calculation / score / ranking.

I'm trying to figure out the quad system

It makes no sense to bundle 363 teams into 4 groups and treat teams differently who beat the 75th and 76th ranked teams on the road

 

[upperdeck]

no different than saying 1 team beat 25th ranked opp 4 times and another team beats #26 7 times and gets nothing for it.

Also nothing adds in how hard it is to beat teams on longer rd trips or back to back rd trips or rivalry rd trips or lots of variables that make a game harder.

 

[Orangezoo]

I just saw Andy Katz bracket. I cannot believe some of the teams he has in and where he has them. Has Mich state as 8 seed. I can’t believe what I just saw

All back to MOV- MSU with a total net MOV of 235 which I broke down others ahead of us with worse or similar records. They have had some big wins over weaker opponents in non conf and in conference driving that figure. They had one big win over Baylor and then at home vs Illinois. Decent Resume for a bubble team but only an 8 seed if you believe the very questionable Net ranking.

 

[RF2044]

I'm trying to figure out the quad system

It makes no sense to bundle 363 teams into 4 groups and treat teams differently who beat the 75th and 76th ranked teams on the road

Right -- the mathematical model is goofy.

But that's all it is -- a model. The output isn't the problem, the criteria used to calculate it is what's strange.

 

[All4SU]

Complete resume should certainly carry major weight, but with some acknowledgement that in this era of the transfer portal, teams evolve as the seasons plays itself out.

• I also believe that the college football playoff has skewed things a bit with their emphasis/directive to identify the best teams as opposed to the most deserving teams. And that has bled into college hoops.
• If this was solely about the "best teams," then the auto bids from the single-bid conferences should go away (I don't want them to, but that a definite inconsistency).

This should be about MOST DESERVING.

I don’t think so. It was never about the 64 (68) BEST teams in the country. It’s about the 40 or so best teams in the country, and enough other spots for every league champ. It’s a great system.

 

[MikeSU02]

I'm trying to figure out the quad system

It makes no sense to bundle 363 teams into 4 groups and treat teams differently who beat the 75th and 76th ranked teams on the road

This is one of my qualms as well. Why are there quadrants at all? Every team has a rank and can be assigned a value. Why does, to your point, 75 v 76 have a large chasm?

 

[Millhouse]

Right -- the mathematical model is goofy.

But that's all it is -- a model. The output isn't the problem, the criteria used to calculate it is what's strange.

Simple and transparent usually perform just as well and you can understand where they depart from reality. Complex and opaque rarely perform better and it's hard to understand what went wrong

You could easily rate strength of schedule with the weighted point margin and apply it to winning percentage. You don't get dinged for playing tough unlucky teams and you don't get dinged for consistently winning more than the Pythagorean margins would predict

 

[Scooch]

I would have to disagree with JB's opinion here.

A core principle of the selection game is "Fully Body of Work", the underlying principle being that all games are valued the same no matter when they are played... that is a good win (or bad loss) is valued the same whether it happened in November, January, or late February.

Not only was the last 10 games metric dropped about 10-12 years ago, the committee goes out of its way year after year in their interviews to adamantly state that selection is not based on this principle, and that its not acceptable for them to select teams based on last 5 or last 10 games..

And its not just word salad from them. We have constantly seen teams with good ends to the season (like Texas A&M) not get in, and teams like Oklahoma recently get in because of huge wins in November. Remember the disaster we had to close the 2016 season? I think we lost 5 of 6.

Not saying going that way is the right or wrong way of doing things, but it is the way they do things and their historical selections back it up.

Plus JB is kinda wrong. Even if there was a "last 10" evaluation criteria, we're 6-4 in the past 10. Which is... fine.

It's great that we've won the last 3, and 4 of 5, but it's not like we've been a juggernaut since mid-January.

This is precisely why there needs to be some kind of statistical element to selection, even if NET is inscrutable. Otherwise it's all about narrative, which is clearly more unreliable.

It's like when we make the Sweet 16 (and lose) and people proclaim it was a "great run". I mean, we won two games. That's not a "run" by any definition. But it feels like a run given the stakes. Narrative.

 

[Fly Rodder]

Why does, to your point, 75 v 76 have a large chasm?

It shouldn't because to me there's essentially no meaningful difference between a 72nd ranked team and a 82nd ranked team, like there is between a top-20 team and 60th ranked team. They can do percentiles using an excel spreadsheet. There's no need for it to be only 4 tiers unless they've already done the math and figured out there's no significant difference, and there would be cutoffs even with 10 percentiles divisions anyway.

I just don't think there's that big of a difference except when getting down to brass tacks trying to choose between 6-8 teams for the last couple of spots.

 

[Fly Rodder]

It's great that we've won the last 3, and 4 of 5, but it's not like we've been a juggernaut since mid-January.

We've been improving overall, and the offense efficiency has been lights out lately, but the defense has taken a header. We've been outscoring teams by brute force and I just don't know how sustainable 55%+ eFG is.

It's taken a lot to improve from a middling D1 offense (all of D1, not just P6), but it's been improvement.
From Bart Torvik:

 

[tbonezone]

 

[MikeSU02]

We've been improving overall, and the offense efficiency has been lights out lately, but the defense has taken a header. We've been outscoring teams by brute force and I just don't know how sustainable 55%+ eFG is.

It's taken a lot to improve from a middling D1 offense (all of D1, not just P6), but it's been improvement.
From Bart Torvik:

View attachment 238268

View attachment 238269

Shorter, we have imperfect parts and lots of one-dimensional players, so it’s pick your poison when Red is deciding on lineups.

It’s also why Maliq is our MVP.

 

[Cheriehoop]

I would have to disagree with JB's opinion here.

A core principle of the selection game is "Fully Body of Work", the underlying principle being that all games are valued the same no matter when they are played... that is a good win (or bad loss) is valued the same whether it happened in November, January, or late February.

Not only was the last 10 games metric dropped about 10-12 years ago, the committee goes out of its way year after year in their interviews to adamantly state that selection is not based on this principle, and that its not acceptable for them to select teams based on last 5 or last 10 games..

And its not just word salad from them. We have constantly seen teams with good ends to the season (like Texas A&M) not get in, and teams like Oklahoma recently get in because of huge wins in November. Remember the disaster we had to close the 2016 season? I think we lost 5 of 6.

Not saying going that way is the right or wrong way of doing things, but it is the way they do things and their historical selections back it up.

In 2016, players had to sit out a year to get acclimated, in 2016 there was no NIL luring players to leave their current teams to play immediately for a new team. In 2016, there were more experienced coaches with decades of experience vs the current college scene where new coaches are filling the ranks for coaches who have aged out and/or didn’t want to deal with rerecruiting their own players, working to integrate transfers immediately to meet both team expectations and academic ones, finding NIL donors to keep who they have or immediately fill gaps they don’t have. It’s a new college basketball world out there.