Not Feeling the Love from NET!
MIchigan moved up to No. 7 this week in both the AP and the Coaches Poll. MSU dropped to 9 in the AP and 11 in the Coaches Poll. Yet, in the new NET rankings, MSU is 8 and Michigan is 9. Any insight into the way NET is calculated that would explain the lack of love (other than that Delany puts together the NET rankings)?
UNC also is not feeling the love from NET. They have moved up to No. 3 in both the AP and Coaches Poll, but are ranked No. 7 by NET. Personally, I think they are overrated in the polls, but can't complain about them being just ahead of Michigan in the NET, based on the way the teams have been playing over the past month or so.
Doesn’t really matter and if we beat MSU in Breslin it further won’t matter because we’ll have the record and plaplit in games
Edit:split* not plaplit. Oops
Shit, what’s a plaplit?
It's a covfefe. Check your thesaurus.
Advanced stat acronyms are getting wild.
I've looked at the keyboard to see if "plaplit" is just a result of mistakenly hitting one key to the left or right, but I got nothing here.
Might be a secret code word used by the Illuminati. They're huge March Madness fans.
I double-majored in Plap Lit and Creative Writing.
NET seems about right...
If we get a two seed it doesn’t really matter to me. Beating MSU would be beneficial.
A #2 seed while drawing Gonzaga or Kentucky as the #1 in our side of the bracket would be ideal in my book, at this stage.
By the way - gotta believe that if Michigan is a #2 seed and Houston is a #3 seed that we will be in the same bracket. The NCAA loves their year over year emotional rematches...
You know, not every single thing that doesn't go Michigan's way has to include a complaint about Jim Delany, the Big Ten, referees, etc. Especially something as miniscule as two spots in the NET rankings.
Good morning, Mr. Delaney.
I'm really curious why anyone cares about basketball rankings.
Because we are nearing the end of the season where the tourney seeds and rankings are going to probably align fairly well and it is better to have a higher tourney seed so you aren't playing universally awesome teams until later on in the tournament?
The rankings are still meaningless. The committee uses season résumés to determine seeding. Wins are what matter. Not rankings.
You are incorrect.
“What has been developed is a contemporary method of looking at teams analytically, using results-based and predictive metrics that will assist the Men’s Basketball Committee as it reviews games throughout the season,” said Dan Gavitt, senior vice president of basketball for the NCAA.
The NCAA invented NET (NCAA Evaluation Tool) to assist the selection committee when evaluating all the teams.
I don't really care personally because we're probably a 2 regardless, but saying the NET rankings don't matter is incorrect. The sole purpose of the NCAA inventing the ranking is specifically for it to matter.
The net is not meaningless, it's the new ranking that the committee uses, along with other things (quadrant 1, 2 wins) to determine seeding in Michigan's case, or who gets in for at-large teams.
The polls are irrelevant, agree there.
The season resumes are evaluated based on quadrant 1, 2 wins and the teams in quadrant one is determined by the net rankings, so pretty important. The OP has a good point in that even after UM's win over Maryland on the road and MSU's loss to Indy, MSU is at 8, still ahead of UM at 9. That doesn't mean MSU will be ranked higher by the committee, but it would have been good to see UM make some move in the net.
No, the OP does not have a good point. The OP is committing a classic fallacy by overvaluing a small set of data points. The NET is not computed by taking last week's ranking and bumping down the teams who lose, the way the polls are.
MSU has been a better team than Michigan, efficiency-wise, this season. They also have the head-to-head win, for those of you who believe in that stuff. Yes, Michigan has more wins, but MSU had a harder schedule -- other than Villanova and North Carolina, Michigan's non-conference schedule has turned out to be much weaker than we thought.
Furthermore, whereas Michigan has delivered quite a few 'meh' performances, especially in the non-conference buy games, MSU went and blew people out. Sure, we may not be impressed by a 68-point victory over Tennessee Tech in which Cassius Winston managed to play 27 minutes, but the computers are.
Now, that's not to say that Michigan can't beat them, or can't finish the season ahead of them in NET or with a better seed. But we should be cognizant of what MSU has done, even if we all hope they stop doing it ASAP. (Well, except the losing to Indiana part. They can feel free to keep doing that. :)
I didn't think UM would move up a lot, I did think they'd swap with MSU after last weekend. MSU's resume is very good agree, they were blowing out teams, some on the road, before their first conf loss. And the non-conf schedule is what's keeping UM out of the 1-seed discussion.
The problem is that they're only releasing the ordering, not the raw NET number. In KenPom, it was easy to see that Michigan wasn't going to catch MSU with one week's results. In fact, if MSU hits their KenPom number tonight against Nebraska, Michigan will probably need a double-digit victory in order to pass them with a win at Breslin. Since they don't do the same with NET, it's impossible to know how far Michigan is behind MSU.
Your comment is what the kids call "an epic fail"
#saulgoodman
So, you still have to beat all the teams. Or not. Basketball polls have always seemed superfluous to me. It's not football where finishing in that coveted top 4 means something.
Why would you compare polls to anything. Polls are worthless. SMU was ranked one year and didn't make the tourney. That's how far off the polls can be.
I remember this but what I seemed to remember is they weren't eligible for the tourney that year. I could very well be wrong on that.
While there is no way of knowing without further input, I suspect that you guys are talking about different years.
In Larry Brown's first year as coach (2013-14), SMU was somewhat controversially left out of the NCAA tournament. Technically, they didn't finish the season ranked, but they were #18 two weeks before the end of the regular season and were #24 during the penultimate week; they fell just outside the top 25 before Selection Sunday.
Two years later, SMU finished 25-5 and certainly would have been given an invitation to the tournament, but they were ineligible.
Double Post!
Triple Post!
No quadruple post? I'm upset now.
I doubledawggoblue dare you to add the quadruple post.
Feels like we're splitting hairs. All of those examples are relatively close, no real drastic differences. Seeding is still in the hands of humans so anything can happen.
...and the quality of play a team exhibits down the home stretch of a season has a big influence on how those humans make their final decisions.
Head to head?
A simple Google search could answer this question for you. But the short answer: the NET ranking does not adjust efficiency margin for competition and caps scoring margin at 10 points. Consider it a mix of KenPom and RPI.
It's not perfect, but it is a huge step up from RPI. Consider that under the old RPI system, Michigan would be rated #11.
Here is NET graphic for you:
Probably the same reason KenPom has MSU ahead of us. Their efficiency ratings are better than ours.
I think almost all the Big Ten teams are ranked slightly lower in the NET compared to KenPom. I haven't done the math, but I think the new 20-game conference schedule may be a factor. Instead of each team having two more games that would most likely be wins, there is a guarantee of losses within conference play. Because the NET (and RPI) have some weight to winning percentage, this is probably a factor. This effect would have been even more pronounced though if the RPI were still being used though as it didn't take into account efficiency at all.
I'm net gonna let it bother me.
As McGary once drew on a white board......WIN THE GAME! Michigan has a chance with MSU and the BTT to lock up a 2-seed or fumble their way into a 3-seed. Either way I like our chances.
Fun fact about our resume......
- Michigan has defeated the current first place team in the B10 (Purdue), ACC (UNC), Big East (Villanova), and MEAC (Norfolk St.).
Disappointing fact about our resume......
- It is unlikely that any other non-conference opponent makes the field this year.
On a slightly unrelated note...can we talk about how overrated the Big 12 is? Most projections have 8 Big 12 teams in the field, including a 16-14 Texas squad, Oklahoma with 10 conference losses so far, and TCU with 11 conference L's. So many quality losses to be had in the Big 12. smh
By that same math, the Big Ten should have at least 10 in the Tourney.
I'll have a problem with Texas making the tourney and Indiana missing it, assuming all else is equal to finish the season.
Texas is #26 in KenPom and 8-9 in conference play. I have much more of a problem with OU or TCU. Texas beat UNC and Purdue and has conference wins over KU, KSU, and ISU. OU beat Florida, Wofford, Creighton at home, and won at Northwestern in overtime; their only other quality win is a 2-point home win over Texas (and, if you insist, a 2-point home win over TCU).
TCU arguably has even less; they've beaten Florida, swept ISU, and beat Baylor and Texas at home.
Indiana is 3-12 since January 3. I'm not going to shed any tears for them if they miss the tournament. Hell, I'd put PSU in before Indiana, although they might have to win the Big Ten just to get to .500. :)
Because NET is a reflection of a season’s worth of work whereas polls are more a reflection of inertia.
This. The polls are an awful basis for comparison.
More like NYET amirite?!
When thinking about the NET you gotta think about Quadrants.
MSU = 11-4(Q1) 5-2(Q2) 5-0(Q3) 4-0(Q4)
Michigan = 8-5(Q1) 9-0(Q2) 2-0(Q3) 7-0(Q4)
From this perspective, you can see why MSU is ranked above Michigan given their Q1 record. Now they've got a couple of "bad" losses in Q2 unlike Michigan but then it comes down to whether you reward good wins or punish bad losses.
No, not at all.
The quadrants make use of the NET; they're not an input to it. It's much more instructive to look at KenPom, which is built using similar methods to the NET, and notice that MSU is ahead of Michigan there too, because their offense is much better than Michigan's whereas their defense isn't that much worse.
Should be 8-4 Q1. And that would mean we have the same number of Q1 losses as MSU but no Q2 losses compared to 2. 3 less Q1 wins but we can make that up over the next few games.
Or, better yet, with last night's result, Michigan and MSU are now guaranteed to be on opposite sides of the bracket. They can only meet in the final. The winner Saturday thus guarantees that they'll only have to face one or the other of the top 3 teams (e.g., if Michigan wins, they could face Purdue or MSU but not both).
Hand wringing over polls is non-productive.
Just "Win the Game" and the rest will take care of itself.
Most likely record until selection Sunday is 2-2. 1-seed seems out of reach, 3-seed seems unlikely. Not much wiggle room either way so no need to sweat the ratings.
Oh, ye of little faith. :) 2-2 may be the most likely record, but it's sure not any fun.
A 1 seed is definitely not out of reach, but may require winning out. They might also get a 1 if they win Saturday but lose to Purdue in the Big Ten Tournament final.