Welcome!

By registering with us, you'll be able to discuss, share and private message with other members of our community.

SignUp Now!

Bracketology

If you believe that the Committee will stick to its principles and you are hoping to avoid the East region with Villanova (where Duke almost certainly sits now), there are a few paths to that happening:

1. Most obviously, earn a #1 seed. Assuming UVA and Villanova are locks, that means finishing within the top 2 of this group of 8: Xavier, Kansas, Duke, Michigan State, Purdue, Auburn, Texas Tech, UNC. I think Duke is probably sitting at #3 right now in that group.

2. Earn the best #2 seed but have Villanova and UVA switch places such that Villanova is the overall #1. In this case, Duke would not be able to be placed with the top overall seed and would avoid Virginia in all likelihood, given they're in the same conference. This would probably result in Duke going to Omaha, ideally with Xavier or KU as the #1.

3. Earn the last #2 seed and likely get shipped out West to the least desirable region in the Committee's eyes. Never mind that it will have the worst team on every seed line other than possibly where Arizona and Gonzaga are slotted.
 
Not that I want Duke as the #1 out West, but I don’t get why the overwhelming majority of brackets currently have KU over Duke for that last #1. Duke has a better record (barely), and Duke is ranked ahead of KU in RPI, BPI, Kenpom, and Sagarin which are the rankings the committee is looking at this year, right? And Duke is ahead of KU in the human polls.
 
Last edited:
@physicsfactor – It remains to be seen, I guess, exactly how much this RPI "quadrant" stuff factors into the final seeding, but people seem to be talking a lot about it and that's the biggest separator between Duke and KU right now.

At this moment, Duke is very weak against Q1 with a 3-4 record. Kansas is 9-4 against Q1 opponents. (UNC is 9-5).

As far as I can tell, Duke has the worst record against Q1 opponents as any of the teams that have reasonable shots at No. 1 seeds. Michigan State is 3-3.

https://extra.ncaa.org/solutions/rpi/Stats Library/Feb. 18, 2018 Team Sheets.pdf
 
The main takeaway from this quadrant stuff is that Duke really fucked itself by losing those Virginia and St. John's games.
 
The RPI quadrant stuff is retarded to me. Duke is ranked ahead of KU in the actual RPI. Why would records vs a certain set of RPI matter more than the actual RPI ranking if they’re going to use the RPI at all?
 
physicsfactor said:
The RPI quadrant stuff is retarded to me. Duke is ranked ahead of KU in the actual RPI. Why would records vs a certain set of RPI matter more than the actual RPI ranking if they’re going to use the RPI at all?
Click to expand...

I think it's the NCAA's way of finally admitting that the RPI is deeply flawed, but without going all the way and flat-out scrapping it. So they are trying to use it more "effectively" by essentially weighting wins.

But yeah, if the quadrant concept ends up factoring heavily into the final seeding, then it raises the question of whether the overall RPI rating matters much at all, except in terms of separating teams into quadrants.
 
I'm going to be as biased in favor of Kansas as possible in the sample sizes I use for this.

Since 1988 (and including 1988), when Kansas won the title and 2 Big Eight teams were in the Final Four, 13 total Big Eight or Big 12 teams have made the Final Four. That's 13 out of the 120 Final Four teams over the past 30 years (10.8%). 2 total Big Eight or Big 12 teams have won the title over that 30-year period (Kansas in 1988 and 2008) (6.7%).

For seeding purposes, why does the Big 12 continue to get treated like a juggernaut conference, despite such terrible empirical/historical results? I'm as staunch a believer in current objective data as anyone, but at some point, perhaps after 30 years, don't you question whether other conferences are simply stronger than the numbers say relative to the Big 12, due to some macro-level issues?

For example, the ACC. 28 have made the Final Four (23.3%) and 9 have won the title (30.0%) over the same time period. The difference in the number of teams in each conference over the years does not come close to accounting for the difference in success rate, and it's debatable whether having more teams in the same geographical area in the same major conference (e.g., 4 teams in the state of NC) actually helps or hurts each of those programs field an elite team each season.

For a fairer example, based on numbers of teams and not being known as a basketball hotbed, the Pac-10 or Pac-12. 10 Final Fours (8.3%) and 2 titles (6.7%). That's a much closer comp for the Big 12 historically. Neither conference has been close to the ACC's level.

The annual consideration of Kansas for a 1-seed over teams that have fewer losses and are higher overall in RPI, Kenpom, Sagarin, etc., is a joke every year. The one time Kansas won it all since 1988, they had a pristine resume with 3 losses and were no worse than #2 on any reputable computer going into the Tournament.

That said, Kansas will obviously win the title this year.
 
I’m not sure what it is that gets folks to overrate the Big12 every single year. BracketMatrix updated this morning and still has Oklahoma as the top 7 seed (although I don’t know how often all of the input brackets are updated).
 
The Oklahoma thing is crazy. Three games underwater in their conference and they're apparently not even on the bubble.
 
TS9 said:
The Oklahoma thing is crazy. Three games underwater in their conference and they're apparently not even on the bubble.
Click to expand...

Yeah, but those losses are to teams in the Big XII, which is so good and competitive because they have to play teams like Oklahoma every game!
 
It’s like SEC West football! I’ve been trying to tell y’all that Arkansas football is a perennial top-10 team. They just can’t get the wins to show for it in that gauntlet of a division!
 
Maybe, just maybe, there's a reason Kansas wins their regular season title every year.
 
SeanMayTriedToEatMe said:
I'm going to be as biased in favor of Kansas as possible in the sample sizes I use for this.

Since 1988 (and including 1988), when Kansas won the title and 2 Big Eight teams were in the Final Four, 13 total Big Eight or Big 12 teams have made the Final Four. That's 13 out of the 120 Final Four teams over the past 30 years (10.8%). 2 total Big Eight or Big 12 teams have won the title over that 30-year period (Kansas in 1988 and 2008) (6.7%).

For seeding purposes, why does the Big 12 continue to get treated like a juggernaut conference, despite such terrible empirical/historical results? I'm as staunch a believer in current objective data as anyone, but at some point, perhaps after 30 years, don't you question whether other conferences are simply stronger than the numbers say relative to the Big 12, due to some macro-level issues?

For example, the ACC. 28 have made the Final Four (23.3%) and 9 have won the title (30.0%) over the same time period. The difference in the number of teams in each conference over the years does not come close to accounting for the difference in success rate, and it's debatable whether having more teams in the same geographical area in the same major conference (e.g., 4 teams in the state of NC) actually helps or hurts each of those programs field an elite team each season.

For a fairer example, based on numbers of teams and not being known as a basketball hotbed, the Pac-10 or Pac-12. 10 Final Fours (8.3%) and 2 titles (6.7%). That's a much closer comp for the Big 12 historically. Neither conference has been close to the ACC's level.

The annual consideration of Kansas for a 1-seed over teams that have fewer losses and are higher overall in RPI, Kenpom, Sagarin, etc., is a joke every year. The one time Kansas won it all since 1988, they had a pristine resume with 3 losses and were no worse than #2 on any reputable computer going into the Tournament.

That said, Kansas will obviously win the title this year.
Click to expand...


I suspect that the B1G is similarly bad at post season success - they also have 2 titles since 1988 - Michigan in 89 and Mich St in 2000(?), they do have more final fours I guess
 
Pretty good RPI site for keeping track of the new quadrant system:
http://www.warrennolan.com/basketball/2018/rpi-live

If we get fucked on the selection sunday, it's going to be our "weak' Q1 record in comparison to other teams. Need to win out and get our Q1 record up to 5-3 to have any chance at all at a 1 seed; not sure of the logistics but this makes it look like Xavier has a much weaker resume than I thought.
 
DurhamSon said:
Pretty good RPI site for keeping track of the new quadrant system:
http://www.warrennolan.com/basketball/2018/rpi-live

If we get fucked on the selection sunday, it's going to be our "weak' Q1 record in comparison to other teams. Need to win out and get our Q1 record up to 5-3 to have any chance at all at a 1 seed; not sure of the logistics but this makes it look like Xavier has a much weaker resume than I thought.
Click to expand...

Worth noting that on neutral courts, a win over the Top 50 counts as "Group 1". Right now, that would include teams like Miami, Louisville, FSU, and Syracuse, so even a first-round ACCT win could add to the count.
 
Here are the committee’s top 16 from 2 weeks ago, and what they have done since then. (The exact ranking might be a little off but this is how I remember it.)

Team (results since 2/11)
Virginia (WWW)
Villanova (LWWL) Losses at Providence and Creighton. Beat Xavier away.
Xavier (WLW) Loss to Villanova.
Purdue (LWW) Loss at Wisconsin.
Auburn (WLWL) Losses at South Carolina and Florida.
Kansas (WWWW) Beat Texas Tech away and Oklahoma at home.
Cincinnati (WLLW). Losses at Houston and to Wichita St. at home.
Duke (WWWW) Beat Clemson away. Named in federal investigation into chain restaurant meals.
Clemson (LLLW) Losses at FSU and VT and at home to Duke.
Michigan St (WWW)
North Carolina (WWW)
Texas Tech (WLLL) Losses at Baylor and Oklahoma State and home against Kansas. Beat Oklahoma.
Tennessee (WLWW) Loss at Georgia.
Arizona (WWL) Loss at Oregon. Lost coach, may lose best player, became national pariah and scapegoat.
Ohio St. (LLWW) Losses at Penn State and Michigan)
Oklahoma (LLLW) Losses at Texas Tech and Kansas and home against Texas.

Using the committee’s own rankings as the most relevant guidepost, and assuming that they will do exactly what the polls and bracketologists do (i.e., move teams up and down when they win or lose rather than taking a fresh look), I think the new order is this.

1. Virginia. Let’s assume they prefer Atlanta to Boston, so South #1
2. Kansas. Gross but the other options are hard to defend at the moment. Midwest #1
3. Xavier. 2-0 record against St. John’s makes the difference here. This is really a placeholder for Villanova when they win in MSG. East #1
4. Duke. West #1
5. Michigan St. Assuming they can stay scandal free for two weeks, they are East #2
6. Villanova. Again this is probably Xavier’s spot, but right now it isn’t. West #2
7. UNC. I don’t quite buy that they have played their way into the 1-seed convo, there’s still the Wofford loss to deal with. For now, Midwest #2
8. Purdue. South #2.
9. Tennessee. South #3. If this plays out, the South is ripe for upsets.
10. Auburn. Midwest #3.
11-16. Who knows? I think teams can still play their way into the 3. Looking at Michigan and Kentucky and West Virginia in particular. This would be a logical place for Gonzaga too.
 
This is a great breakdown. I do think that Villanova (2 losses at the time of the bracket reveal) was probably way out in front of everyone other than UVA at the time, so I'm not sure they're really off the one line as of today. Especially with the @Xavier win and still having only 4 losses.

It's become clear that Tony Bennett's choice of Charlotte or Pittsburgh is going to be of large importance for Duke, assuming you think Duke is much worse in road games than neutral site games. I just checked and learned that you cannot fly direct from Charlottesville to Pittsburgh. In fact, some of the flights include Charlotte layovers. This is good.
 
You must log in or register to reply here.

Chat users

  • No one is chatting at the moment.

Chat rooms

  • General chit-chat 0

Forum statistics

Threads
1,065
Messages
424,060
Members
624
Latest member
Bluegrass Blue Devil
Back
Top Bottom