Sunday, January 22, 2012

W-7 BP68

It's been a busy week at the top of the bracket. I've had the same four teams as 1 seeds since April, and I talked this week for the first time about switching it up by dropping North Carolina out due to the Dexter Strickland injury. But I decided to keep them there because Duke hasn't impressed lately either, and I'm not going to become convinced that Duke is going to win the ACC until I see UNC struggle without Strickland. The core of their team is still the front court and Kendall Marshall.

You might ask why I think North Carolina is better than Kansas, and the answer is that I don't. I think Kansas is better. The problem is that Kansas has much, much more competition than UNC does. They have to get past Missouri, Baylor and Texas. The odds of Kansas failing to win their conference's regular season and/or tournament title are higher than for UNC, despite Kansas being the better team. That's why I still have Kansas on the second line.

I really struggled filling in those 12 seeds. Once again there is a clear dichotomy. All the teams I have seeded as a 10 or better seem like legitimate Tournament teams, and I feel confident having all of them in the field. But the at-large teams on the 12 line? Not good. And the collection of the best teams outside the Tournament? Putrid. I had to drop Northern Iowa out, and so I decided to put in Iowa State in their place. But I thought this evening that I'd be dropping Stanford out as well, and just couldn't find anybody to put in their place. NC State? Minnesota? Marshall? Eh. I'm leaving Stanford there for now.

But as I said last week, this is the thinnest bubble I can ever recall at this point in the season. I honestly think that if the season ended now, the Selection Committee would only have serious debate on filling the final four or five spots, and they'd only have around 10-12 teams to choose from. Normally the bubble is two or three times that size at this early stage. I think that's just a random quirk more than anything else.

Among the automatic bids, I switched my pick in the MEAC, putting in Norfolk State for Morgan State.

I started the Full Bubble last week, and so I will be eliminating teams each new bracket the rest of the way. This bracket? I only eliminated four teams: Rice, Lehigh, College of Charleston and North Dakota State. There are a couple dozen teams that are almost assuredly going to be eliminated in the next two or three weeks. Once they collect the losses that I expect them to get, they'll be out. Last week I started the Full Bubble with 83 teams not in the Field of 68 that had a chance at an at-large bid. With four teams eliminated that's down to 79. It will be down to around 8-10 on Selection Sunday.

Below are my disclaimers:

If I projected your favorite team below where you think it deserves to be, it's because I hate your favorite team. If I projected a team above where you think it deserves to be, it's because I secretly love them and have an incredibly blind bias in their favor.

On a more serious note, this is a projection of the final bracket on Selection Sunday, and not a listing of how I think teams would be seeded if the season ended now. There's a difference.

Here we go:


2. KANSAS (BIG 12)
2. Duke
2. Florida
2. Georgetown

3. Texas
3. Missouri
3. Michigan State

4. Baylor
4. Indiana
4. Marquette
4. Vanderbilt

5. UConn
5. Alabama
5. Wisconsin

6. Virginia
6. Saint Louis
6. Michigan

7. Louisville
7. San Diego State
7. West Virginia

8. St. Mary's
8. Purdue
8. Kansas State
8. Illinois

9. Wichita State
9. Florida State
9. New Mexico
9. Seton Hall

10. BYU
10. Cincinnati
10. Dayton
10. Xavier

11. Southern Miss
11. California
11. ARIZONA (PAC-12)

12. Northwestern
12. Mississippi State
12. Stanford
12. Iowa State

13. OHIO (MAC)




Teams seriously considered that just missed the cut:
NC State, Virginia Tech, Minnesota, Marshall, Cleveland State, Northern Iowa

Decent resumes, but not good enough:
Miami (Fl), Notre Dame, La Salle, UMass, St. Joseph's, South Florida, Iowa, Oklahoma, Texas A&M, Drexel, VCU, Central Florida, Missouri State, Colorado State, Wyoming, Colorado, Oregon, Washington, Arkansas, LSU, Mississippi, Nevada

Long shots, but still in the at-large discussion:
Clemson, Maryland, Duquesne, Richmond, St. Bonaventure, Pittsburgh, Providence, Rutgers, Villanova, Weber State, Nebraska, Penn State, Oklahoma State, Georgia State, UW-Milwaukee, Akron, Drake, Illinois State, UCLA, Oregon State, Washington State, Auburn, Tennessee

Still alive, but pretty much need a miracle:
Boston College, Georgia Tech, Wake Forest, Charlotte, Fordham, George Washington, DePaul, St. John's, Texas Tech, Old Dominion, UTEP, Tulane, Tulsa, Valparaiso, Loyola-Maryland, Ball State, Buffalo, Kent State, Indiana State, Air Force, Boise State, TCU, Wagner, Georgia, South Carolina, South Dakota State, Denver, Utah State


Anonymous said...

I have commented in the past on your projections on Texas and Missouri. I didn't think I crossed the line (never called names, never used rough language), but you've deleted my posts. If I have offended you, I am sorry. That being said...

Even if you're projecting your own opinion of how the rest of the season will play out, you're apparently projecting based on an alternate reality. And I've seen you mention that you are a numbers guy, which I totally respect. I am as well. So let's go side-by-side here:

Missouri: 18-1, 5-1 Big 12.
Kenpom: #7
Sagarin: #1. That's right, #1.
Sagarin predictor: #7
Quality road wins: @Baylor, @Iowa State
Quality home/neutral wins: Illinois, Cal, and...Texas
Other top 100 home/neutral wins: Oklahoma, Villanova, and ND

They have only two more games where they're legitimate underdogs, both vs. Kansas. Even going with a less than optimal scenario, saying they drop 4 more games in conference, that would make them 13-5 in conference, 26-5 overall. Even if they went 1-1 in the Big 12 tourney, they'd be 27-6 overall, and that's a pretty standard #2 seed record.

Texas: 12-7, 2-4 Big 12.
Kenpom: #24
Sagarin: #44
Sagarin predictor: #25
Quality road wins: none
Quality home/neutral wins: Temple
Other top 100 home/neutral wins: NONE

Texas hosts Missouri, and yes, they can legitimately win that game. But they still have to go to Baylor and KU, and host Baylor, K-State, and Iowa State. Kenpom projects 9-9 in the Big 12, but for argument's sake, let's say they go 11-7 in conference. That would leave them 21-10 entering the Big 12 tourney. If they made the final of the tourney and lost, they would be 23-11.

I am not denying that Texas is a good team, or has the potential of being a good team. But at a certain point of the season, you HAVE to look at what has actually happen, use numbers and projections to visualize what CAN happen, and blend the two. I've given you a pessimistic projection of Mizzou at 27-6 overall and an optimistic projection of Texas at 23-11. At best, that's a 6-8 seed in the tourney. in the world can you project Texas as a 3 and Missouri as a 3, BELOW THEM ON THE S-CURVE???? It defies sense and logic. Anxiously awaiting your appeal. And if you just defend it by saying "this is my projection", you're showing bias to both current results and the expert numerical analysis that you use well on your site.

Jeff said...

As long as you make an argument based on statistics and facts, I'll always leave the post up. I leave plenty of posts very critical of me.

First of all, I can basically guarantee that every single Division I team is going to lose at least one game that they're favored in the rest of the way. I think Missouri is likely to lose four more games in the regular season, and then another game in the Big 12 tournament. You can argue that would make them a 2 seed, I'm putting them at a 3 seed. I'm not sure that's really a huge disagreement, do you?

As for Texas, you have to remember how important those final ten games are, particularly the conference tournament. Let me give you a few examples:

Last year's UConn team entered the Big East tournament knowing that a first round loss would have put them on the bubble (they were only 9-9 in conference play). They won the conference tournament and earned a 3 seed.

The 2005-06 Syracuse team entered the Big East tournament looking like an NIT team (7-9 in conference play), and they won the conference tournament and earned a 5 seed.

That same year, Iowa was on the bubble just two weeks before the Big Ten tournament (they were 9-5 in the Big Ten with zero quality non-conference wins and only three Top 25 wins in conference play). They won their final two regular season games at home in games they were favored and then won the Big Ten tournament and earned a 3 seed.

If Texas finishes 11-7 in Big 12 play and makes it to the Big 12 tournament finals (and I think they're the most likely team other than Kansas to win the conference tournament) then that will probably earn them a 3 seed. Look at their remaining schedule and keep in mind just how close they've been to big wins in their last three games, and ask yourself if it's really that unlikely that they'll win 9 of their last 12 games?

Jeff said...

Look, I understand that I'm predicting future results that are different from the results that have happened up to this point. If the season ended now Texas would be a bubble team. I know that on Missouri message boards people are saying I'm the dumbest person ever because I don't realize that Missouri has more quality wins and better computer numbers than Texas right now. Duh. I watch the games, I see the results.

But the reality is that some teams that have struggled the past couple of months are going to finish strong, and vice versa. I'm trying my best to project that. Sometimes I'll be right and sometimes I'll be wrong.

I remember during the 2009-10 season, Rhode Island spent weeks as in the top ten of the RPI in December and January, and basically every other bracketologist has them somewhere between a 4 and 7 seed. I refused to put them in the Field of 68. Every week I was pounded with Rhode Island fans telling me I was the dumbest person ever (some using more colorful language), I was ripped on Rhode Island websites/message boards, and told that I was too arrogant for believing Rhode Island could possibly lose more than a couple games down the stretch. But I believe that they were a bad team that had just gotten lucky, and I was proven right. They lost five of their final seven regular season games and ended up in the NIT.

After I turned out to be right about Rhode Island, nearly every fan just disappeared - not one came and admitted they were wrong. But my favorite part was that one guy came back and insisted that I was still wrong because there's no way I could have known that they were going to collapse like that, and so I made a bad projection and just got lucky that some fluke stuff happened.

I'm not going to pretend I'm not right all the time. That same 2009-10 season I missed totally on North Carolina and kept them in my bracket longer than anybody else and was proven wrong when by the start of March they had finally played themselves out of Tournament contention.

But that's what I do on this site. I make projections and try to explain them. You are free to disagree. But it doesn't mean that I'm "not looking at what actually happened" or that I have some terrible bias either in favor or against one team or another. It just means we disagree. It's okay.

Big Guy said...

This whole Texas thing is just a joke. I only check in to this site to see when you'll actually swallow your pride and admit you were completely wrong on Texas this year. 23-11 without a Big 12 tournament victory will not be a 3 seed and there is no reason whatsoever to believe they will even get to 23-11 (best case scenario).

Even in the examples you give in the past of teams getting a 3 seed with so-so records, they won their conference tournament, yet you go on to say that Texas only has to REACH the conference tournament to get a 3 seed. WHAT?!?!

Furthermore, if you recall, those UConn and Syracuse teams went on simply MAGICAL runs to win their conference tournament, defying long, long odds to win their conference tournament. WHat makes you believe that Texas (TEXAS!!!!) will have a similar magical run? Where is the logic there?

Sorry my dude, but you are just being stubborn. The first poster had an absolutely brilliant and logical post yet you still you think Texas will somehow put together some miracle run. I mean, it could happen but the world could end tomorrow too.

Anonymous said...

The humongous differences between last year's UConn and this year's Texas:

1) Last year's UConn team won a loaded tournament in Hawaii in November, against some great teams. UConn was also in the top 10 in December/January, showing that they had the talent to deserve a 3 seed after they put it together in NYC.
2) Last year's Big East was absolutely loaded, maybe the best season of any conference ever, and 9-9 there is a LOT different than 9-9 in this year's Big 12.

Thus, UConn last year and your projection of UT this year are far from equal.

I really do think Mizzou will be at least 13-5 and a likely 2, and that Texas will be staring at an 8-11 seed. But I tried to take my personal bias out of it and present a case to show that even if things work out well for Texas, their best case scenario is a 6 or 7.

Anonymous said...

And I've seen you present the cases where you are right and everyone else is wrong. I believe you when you say that happened. You seem to watch a lot of CBB and care a lot, and you generally seem knowledgeable.

What I have NOT ONCE seen you acknowledge is the cases where you are wrong, which we all are. When I am presented with facts that challenge my opinions, I usually relent, concede, and learn something from the person who has presented a strong argument.

Anonymous said...

First, let me say that I LOVE your site - never commented here before but I check in daily because I find your analysis to be very in-depth and useful. That said, I almost fell out of my chair when I saw that you STILL have Texas as a number 3 seed, ahead of not only Missouri, but about 40 other teams whose resumes are now, and very likely will be at the end, better than Texas.

I was going to do a full-blown argument but I see that the first poster beat me to it. I appreciate that you responded to him. But I think that response just underscores the fact that you have a huge blind spot (not calling it bias) when it comes to Texas.

You say that you expect Missouri to lose five more games. That would put them, at worst (assuming 1st game loss in B12 tourney), at 26-6. You then emphasize the importance of the last ten games. Included in Texas's last ten games are hosting Baylor, Missouri, Kansas State ... and playing on the road at Kansas. Based on their play to date, and the results, I don't understand how you could POSSIBLY assume anything better than 8-2 in the last ten ... and of course, Texas is at Baylor next Saturday. So 11-7 in conference is an optimistic, but reasonable, prediction - as you suggest.

That would leave us with Missouri at 26-6, 13-5 and Texas at 21-10, 11-7. Texas would have 6 total wins over current RPI Top 100; Missouri would have 11 (both figures assume wins over all teams worse than 100).

Setting aside the conference tournament, those numbers would probably earn Missouri a 3 seed, and Texas around an 8 or 9.

Now, say Texas really does win the conference tournament (more on that in a second). If they do so by beating Baylor, Missouri, AND Kansas (extremely unlikely) - then they are 24-11, 9 wins over top 100, and MAYBE they could jump to a 5 or 6 seed. But jumping them ahead of Missouri, and Baylor (already has 7 wins over Top 100, likely to end with 11-12) ... it's just not going to happen.

But maybe most importantly: What in the world has Texas done to suggest to you that "they're the most likely team other than Kansas to win the conference tournament"??? Close losses are just that - losses. (And their loss at Missouri wasn't close ... Missouri was up 16 in the first half.) At some point, a team has to actually WIN a few of those games. I'll agree that Texas has been better than their record suggests, and I would even agree they will be a dangerous sleeper-type team come March. But saying Texas has a better chance to win the B12 tournament than Missouri or baylor ... and projecting them a 3 seed ... you just can't justify that by looking at the numbers OR the watching the team play.

Anyway, keep up the (mostly) great work!

Big Guy said...

What I meant to say was that you stated Texas only has to reach the conference tournament FINAL and that should be good enough to get them a 3 seed. Just wanted to clarify.

Jeff said...

Alright, this is getting absurd. Let me explain a few things that you guys got wrong.

First of all, the difference between this year's Texas team and last year's UConn team is that this year's Texas team is rated better in Pomeroy and the Sagarin PREDICTOR than last year's UConn team was at this point. This year's Texas team is more talented and looks better to me.

The fact that UConn won a November tournament is meaningless. I've presented the data here before - there is zero correlation between success in November tournaments and success in March tournaments. And remember that UConn needed a miracle to get out of their first game at that November tournament against a mediocre Wichita State team. The odds of UConn winning last year's Big East tournament were much lower than the odds of Texas winning this year's Big 12 tournament.

And no, the Big East was not the greatest conference ever last year. That was nonsensical media hype. Last year's Big Ten was clearly better than last year's Big East. I'd take this year's Big 12 top to bottom over last year's Big East.

And by the way, did you not notice my most recent post in this comments section? I wrote a whole paragraph about a prediction I got wrong the same year I nailed Rhode Island.

Look, I understand how bracketology is. This is my sixth season of doing this. I understand that if I get 9 out of 10 right, the 9 I got right will be forgotten while the fans of the 10th will come on here and tell me I'm an idiot and all my predictions are laughable.

That's all fine. I choose to do this. But I've explained this Texas thing over and over and over again. Maybe I'll get it right, maybe I'll get it wrong. But I'm not going to keep repeating this argument ad nauseum.

Anonymous said...

The committee doesn't see things the way you do. They buy into the hype that EXPERTS create based on real results. The Big East getting 11 teams last year was justified, whether you believe it or not. Results of the Big East teams in the tournament aside, that was all deserved.

On January 22nd last year, UConn was Kenpom's 19th best team. So much for Texas being higher than UConn was at this time last year.

I agree that the Big Ten and Big 12 are the two best conferences in CBB this year. But I cannot get behind saying that they are better than the Big East. Those 11 Big East tournament teams all FINISHED the season in the top 42 of Kenpom. That's incredible depth.

Last year's Big Ten was very balanced; the bottom of the conference was clearly better than the South Florida, Rutgers, Seton Hall type teams. But a valid 11/16 > a valid 7/11 in tourney teams.

This debate is straying into greater topics now. I'll stop my posting before everything becomes circular, but I'll reiterate my stance on UT: even if things go well for them--finishing the regular season and conference tourney at 11-4--they're still going to be a 23-11 team without the same "feathers in their cap" that UConn had last year. They will not be a 3. The committee would never let them happen.

Well, unless DeLoss Dodds called his friends at ESPN, started the Longhorn Network propaganda machine, and paid Doug Gottlieb to angrily say great things about Texas. Maybe then the committee will "have" to listen.

Jeff said...

Those are some pretty impressive conspiracy theories.

The Big East deserved the teams they got in the Tournament, but:

conference strength /= number of teams making the tournament

But again, I'm not going to spend all day debating Texas. I put a lot of time into rating well over 100 different teams. If all I end up getting wrong is ONE of them, I did a pretty damn good job.

Go print out this bracket, and go print out the Bracket Matrix right now. Check back on Selection Sunday. I guarantee you mine will be more accurate.

Anonymous said...

That's a very bold claim. If I had a printer, I would.

One other thing (I know, I know, I promised)...UConn was 16-2 at this point last year, and #5 in the nation (polls); UT is 12-7. Along with the Kenpom rating for UConn being better at this point of the season last year than Texas is this year, are you still prepared to say UT right now is better than UConn was at this point last year?

I respect your conviction, and I love how much you care about CBB.

Anonymous said...

Mizzou is the second most likely team to win the Big 12 Tourney. It's in KC, a place where they've throttled two decent teams already this year and will have the home crowd for any game not against KU or KSU. UT won't have the second best chance, let alone the third (Baylor) or fourth (KSU) or maybe even 5th (ISU).

Jeff said...

Guys, enough. There are 344 other teams in Division I. I've said more about the University of Texas in the past 24 hours then I ever want to talk about any other school. That's enough for one day.

I'm happy to talk about any other team in the bracket.

Ken Miller said...

I was going to stop in to make mention of the slowly improving young Longhorns, but I can see that we have covered that pretty extensively today.

How about the Horizon? Butler has done well in league play and seems to have righted the ship after a brutal non-conference schedule, but it is worth remembering that Butler and Valpo still have to make the two more challenging league road trips. Regardless of who wins this conference tournament, does the league look down enough this year to be a 14 seed?

Jeff said...

It's not inconceivable that the Horizon could end up with a 14 seed. I think it's unlikely though, for three reasons:

1) Cleveland State has a strong resume.

2) Butler will get love from the Selection Committee because of their last two trips to the title game.

3) There will be some surprise automatic bid winners in other conferences. Every year there are a couple of these, and they bump all the 13, 14 & 15 seeds up a line.

Tom said...

I don't think Arizona's getting in, Jeff. They've blown every opportunity this year:

- up 7 at Florida late
- up 2 vs. Oregon late
- up 2 vs. Colorado late

(on and on)

team doesn't have "it" this year - can't close. I hope you're right and they get the auto-bid (or preferably rattle off 7 or 8 straight wins) but it looks bleak.

Jeff said...

Tom, I agree with you that the odds are against Arizona earning an at-large bid. They're still my pick to win the conference tournament, but I've got them as an 11 seed. You always get bumped up a couple of lines for winning a tournament final, particularly a BCS conference.

The problem for Arizona is that they've been inconsistent. And being inconsistent is never good in a conference full of potential bad losses, with no potential quality wins anywhere.

Anonymous said...

So Butler loses another conference game to fall to 12-10 overall and 6-4 in conference. I can't see the committee giving them anything better than a 15 seed at this point (maybe a 14 if they figured it out and a bunch of randoms won their conference tournaments).

I don't really care who you project from the Horizon because any team that comes out of that league is going to have a weak seed and little chance of winning their first round game, and therefore whatever team you pick is little more than a place-holder, but would you agree Cleveland State clearly has the best NCAA resume from that league at this point?

Anonymous said...

Also, Wichita State is sitting at #10 in Pomeroy, #13 in Sagarin, #13 in the Sagarin predictor, and is #16 in the nation in rebounding margin. Just based on the numbers, have you ever seen a more dangerous team that seems to play in the 8/9 game?

I'm sure you'll agree that Florida State is probably more dangerous, but their computer numbers can't compare to those...

Jeff said...

Hi guys, I'll try to answer both of those questions in one post.

On Butler/Cleveland State, I don't even think there's a question which team in the Horizon has the best resume. It's Cleveland State. That's why I have them on the bubble while Butler doesn't even have a realistic path to the bubble. I'll re-evaluate this weekend to see if I still want to make Butler the favorite in the conference tournament and, if so, if I'm going to drop them a line in the seedings... but to answer your question again I definitely agree that Cleveland State has the best resume so far.

As for Wichita State, I do agree that they'd be an extremely dangerous 8/9 seed. If I were a 1 seed, I'd much rather see a team like Illinois or Seton Hall. Right now, Creighton is my pick to win the conference, but if Wichita State proves me wrong and wins it then I could easily see them sliding up to something like a 5 seed.