Joe Lunardi is a pretty good Bracketologist, I suppose, using the meaning of picking the Selection Committee's choices. But not necessarily the other meaning, i.e. making the actual picks in the tournament. I haven't checked his track record, but I assume it's no better than average.
That's why it's not to surprising to see him asking a ridiculous question: Would you take UConn? Or the rest of the field?
Insane question, let's face it. UConn is probably not even the best team in the country, let alone such an overwhelming favorite that you'd take them over all 67 other teams. That's a question that should have been asked maybe twice in the last 30 years: for 2015 Kentucky, and 1999 Duke. Go back a bit and yes, 1991 UNLV for sure. None of these teams won it all, but they should have been overwhelming favorites approaching the 50% level.
Really, to pick a team vs. the field I'd go back to 1976 Indiana and/or several UCLA teams. There were a few years where I'd say there was duo of teams that added up to 50%: 2005 with North Carolina and Illinois, and 2021 with Baylor and Gonzaga.
Lunardi invokes 2006-2007 Florida as his 2024 UConn equivalent:
The 2006-07 Florida Gators, defending national champions, reached Christmas that season with only two losses. They then embarked on a 17-game winning streak, leaving little doubt about the identity of the top team in the land.
He admits that Florida was returning all five starters from the season before, which is a huge difference that he glosses over with "nobody does that any more!" But is UConn the best team? Maybe every AP voter thinks so, but by this point they're voting mostly by résumé not by true strength. UConn is only #5 in our Strength power rating, behind Arizona, Houston, Purdue, and Alabama. Maybe that's wrong, but KenPom has them #2 behind Houston and the BPI has the Huskies #4 behind Houston, Purdue, and Arizona. So the consensus seems to be that UConn has a solid chance at the Final Four but they won't even be the favorite to repeat.
Repeating is difficult. Florida did it with all five starters back, and Duke did it in 1991-1992 with four of their starting five back. UNLV had everyone back in 1991 but lost in the Final Four to Duke. Arizona in 1998 had five starters back but lost in the Elite Eight.
UConn is more similar to some teams that didn't repeat despite being 1-seeds. Like 2017 Villanova, who went 31-3 the year after their championship but fell in the Round of 32. Other 1-seeded defending champs (since 1985) that failed to repeat include 1991 UNLV, 1994 North Carolina, 1997 Kentucky, 2001 Michigan State, 2002 Duke, 2011 Duke, 2022 Baylor, and 2023 Kansas.
Here's how much they returned from the previous year's squads and where they finished (for all 1-seed repeat attempts):
- 1991 UNLV (5 starters back, lost Final Four)
- 1992 Duke (4 starters back, National Champs)
- 1994 North Carolina (4 starters back, lost Round of 32)
- 1997 Kentucky (2 starters back, lost in Finals)
- 1998 Arizona (5 starters back, lost Elite Eight)
- 2001 Michigan State (2 starters back, lost Final Four)
- 2002 Duke (3 starters back, lost Sweet Sixteen)
- 2007 Florida (5 starters back, National Champs)
- 2011 Duke (3 starters back, lost Sweet Sixteen)
- 2017 Villanova (3 starters back, lost Round of 32)
- 2022 Baylor (2 starters back, lost Round of 32)
- 2023 Kansas (1 starter back, lost Round of 32)
- 2024 UConn (2 starters back, ???)
1995 Arkansas might count here since they were a 2-seed and made it to the title game; in fact, they seem more similar to 2007 Florida than UConn does. But as you can see the results are mixed, even with teams much more dominant at the time than UConn is today.
Let's expand the time back as far as seeds go, so 1979 to 1985 and we add just one team:
- 1985 Georgetown (5 starters back, lost in Finals)
So there's something to be said for a defending champ 1-seed having 4-5 starters back: those 6 teams had two champions, 1 runner-up, and a Final Four team. With 1-3 starters back, those 7 teams had a runner-up, a Final Four team, two that lost in the Sweet Sixteen and three that lost in the Round of 32. So it looks like actually returning the team that won the title last year is important. Overall, of 13 teams that tried to repeat and got a 1-seed, 6 made the Final Four, a 46% rate, and 2 have won, a 15% rate—a rate that almost exactly matches the odds of a 1-seed to win the title (24 teams in 38 x 4 attempts since 1985). 60 1-seeds have made the Final Four since 1985, a 39% rate. So being a returning champ 1-seed doesn't seem to impart any real advantage except maybe a slightly higher chance of making the Final Four. But check the number of returning starters first.
Last year's runner-up seems to do just as well if they get a 1-seed the following year:
- 1982 North Carolina (3 starters back - National Champs)
- 1989 Oklahoma (2 starters back - lost Sweet Sixteen)
- 1992 Kansas (2 starters back - lost Round of 32)
- 1993 Michigan (5 starters back - lost in Finals)
- 1998 Kentucky (2 starter back - National Champs)
- 2000 Duke (2 starters back - lost in Sweet Sixteen)
- 2013 Kansas (3 starters back - lost in Sweet Sixteen)
- 2015 Kentucky (3 starters back, lost in Final Four)
- 2017 North Carolina (3 starters back - National Champs)
- 2022 Gonzaga (2 starters back - lost in Sweet Sixteen)
*1984 Houston was runner-up as a 2-seed following their 1983 runner-up year
*1990 Duke was National Champ as a 2-seed following their 1989 runner-up year
*2007 UCLA made the Final Four as a 2-seed following their 2006 runner-up year
*2009 Michigan State made the Final Four as a 5-seed following their 2008 runner-up year
*2010 Butler was runner-up as an 8-seed following their 2009 runner-up year
So two national titles in 10 attempts, again not far from the 15% rate than 1-seeds have to win it all, given the sample size. But the 50% Final Four rate is pretty impressive.
And what about the whole Final Four?
- 1979 Notre Dame (3 starters back - lost Elite Eight)
- 1980 DePaul (3 starters back - lost Round of 32)
- 1982 Virginia (3 starters back - lost Sweet Sixteen)
- 1983 Houston (3 starters back, lost in Finals)
- 1983 Louisville (2.5 starters back, lost in Final Four)
- 1986 St. John's (2 starters back, lost Round of 32)
- 1989 Arizona (2 starters back, lost Sweet Sixteen)
- 1990 Arkansas (4 starters back, lost Elite Eight)
- 1993 Indiana (4 starters back, lost Elite Eight)
- 1998 North Carolina (4 starters back, lost in Final Four)
- 2000 Michigan State (3.5 starters back, National Champs)
- 2002 Maryland (4 starters back, National Champs)
- 2003 Oklahoma (3.5 starters back, lost Elite Eight)
- 2005 Duke (3 starters back, lost Sweet Sixteen)
- 2008 UCLA (4 starters back, lost in Final Four)
- 2009 North Carolina (4.5 starters back, National Champs)
- 2012 Kentucky (3 starters back, National Champs)
- 2013 Louisville (3.5 starters back, National Champs)
- 2014 Wichita State (2.5 starters back, lost in Round of 32)
- 2015 Wisconsin (3.5 starters back, lost in Finals)
*1989 Duke reached the Final Four as a 2-seed following their 1988 Final Four appearance
*1990 Duke was runner-up as a 2-seed following their 1989 Final Four appearance
*2003 Kansas made the Finals as a 2-seed following their 2002 Final Four appearance
Here we have 20 samples, and 5 national champs, a 25% rate. Seems pretty good. And again a 50% Final Four rate, besting the 39% rate for a typical 1-seed.
It appears that if we expand to include 2-seeds we'll capture some more results (I listed the ones I noticed right off the bat), but I suspect it wouldn't be as predictive as the rate for 1-seeds. It probably would exceed the rate for the typical 2-seed, though, so you might emphasize last year's Final Four teams if they get 1- or 2-seeds the following year.
The bottom line, though, is that UConn is nothing special in being a (very likely) 1-seed following a championship year, and the model for their success in the tournament is not 2007 Florida or 1992 Duke, but rather 2017 Villanova, 2022 Kansas, and 2023 Baylor all of whom lost in the Round of 32. That said, UConn is also the weirdest name in NCAA tournament history, winning out of the blue several years, getting to the Final Four whether they're pretty good or really good. So I'm not sure anyone can predict what UConn will do in the tournament.
But you'd be a fool not to take "the field" if given the chance.
2024 Week 15: Committee's Top 16 blesses the top four seeds
The Committee has spoken, and their Preliminary Top 16 consecrated all four of our #1 seed projections—including the overall top seed—and our top four seeds include 15 of the 16 teams they named. There are some discrepancies to note, so here we go:
asdfadfs
Purdue remained our projected overall top seed despite the crazy loss to Ohio State on Sunday. The Boilermakers were the Committee's top seed, too, but that was announced before they fell to the Buckeyes, who had lost 9 of their last 11 games prior. OSU was probably on some kind of "new coach high" is all we can figure about the 73-69 win. Purdue remained at the very top but their "#1 seed odds" fell from 97.2% last week to 94.5%. Meanwhile Houston's 1-seed odds rose by 1.0% after beating Texas Saturday. UConn and Arizona round out our top seeds, with the Huskies adding 12.4% to their 1-seed odds after beating DePaul by 36 and Marquette by 28. The Wildcats couldn't quite match that but they tried, beating Arizona State by 45 to raise their 1-seed odds from 66.3% to 70.2%. The four 1-seeds are taking shape, but UConn and Arizona probably need to win their conference championships to get there, where Purdue and Houston might not have to.
Iowa State comes up our next most likely 1-seed, but the Committee has them a 3-seed right now. Instead as a 2-seed they have Marquette (again, this was before Saturday's action where the Eagles were destroyed by UConn). If the Cyclones want to make it to a 1-seed they need to do something dramatic, like beat Houston on the road tonight (Monday 19th). That and winning the Big Twelve would put them in the mix for sure. Tennessee (10.9%) would need to win the SEC tournament but at least the Vols made it to 100% this week after two dominant wins (+64 points combined) over punching bags Arkansas and Vandy. Also getting to 100% is Kansas, who lost big (79-50) to Texas Tech on the road, but also picked up a key road win over Oklahoma. The Jayhawks are seen as having very little chance in the Big Twelve tournament so their 1-seed odds are minuscule. And North Carolina's 1-seed odds plunged from 11.9% to 5.9% after a loss to Syracuse last Tuesday. But we, like the Committee, still have them at a 2-seed.
The Committee also pretty much agreed with the Dance Chance in terms of 3- and 4-seeds, in particular that Alabama is the top 3-seed. Their 13.7% odds of getting a 1-seed reflect their Strength ranking (#4 currently after their 100-75 dispatching of Texas A&M) and the concomitant chance of winning the SEC tournament. Contrast that with Auburn, who is just a 4-seed by the Committee's early ranking, and who has just 1.7% shot at a 1-seed following their upset home loss to Kentucky. Previous to that, they crushed South Carolina 101-61 so Bruce Pearl's squad is running hot and cold rather than consistent. Duke just makes it into our 3-line after a couple workmanlike wins over Wake Forest and FSU.
Baylor is our top 4-seed while the Committee puts the Bears at a 3; we also have them a 3-seed if selections occurred today. They show a much better 1-seed shot (3.6%) than Kansas though, again due to Strength (#11 to KU's #17). San Diego State and Illinois both join our 4-seed line; the Aztecs are up from a 6-seed after beating 5-seeds Colorado State and New Mexico last week while the Illini added two less-significant wins. Both are 4-seeds by the Committee while Creighton was "considered" for a 4-seed but didn't get it. The Jays play UConn Tuesday night and a win would probably make the Committee reconsider.
Note Duke (0.3%) and SDSU's (0%) chance of a 1-seed. It looks like no Mountain West team is going to be considered no matter their finish, and if a team from the ACC gets a 1-seed it will be UNC. The same goes for the Big East (UConn or bust) and Big Ten (Purdue) and Pac-12 obviously with Arizona. Only the Big Twelve has several candidates and a (small) shot at two 1-seeds. The SEC has three strong teams and the tournament winner has a decent chance but there's no chance for two 1-seeds.
The other team in the Committee's Sweet (pre-Selection Sunday) Sixteen was Wisconsin, whom we have as a 6-seed. The Badgers lost at Iowa in overtime on Saturday, their 5th loss in their last 6 games so it makes me wonder what the Committee's been smoking (maybe some overripe cheese?). Two other teams "considered" for the top 16 were Clemson, who appropriately slips to a 5-seed here after a home loss to NC State, and Dayton, who we have as a "snapshot" 5-seed but project as a 6-seed ultimately.
BYU took a fall this week due primarily to a loss to Oklahoma State but also to a narrow 90-88 win over Central Florida that hurt their forward-looking Strength rating. The Cougars grade out as a 7-seed right now, and are projected to work their way up only to a 5-seed instead of a 3-seed (as projected last week) partly because of that decrease in Strength—and also because time is running out, they only have 6 regular-season games left! Their Strength is still solid enough (#9) that they show a nearly 1% chance of making it all the way to a 1-seed, something that would require them to win the Big Twelve tournament certainly. They have 4-seed Baylor and 2-seeds Kansas and Iowa State still on their schedule, the latter two on the road.
In Mountain West action, New Mexico and Colorado State both survived a loss to new conference "champ" San Diego State and held at 5-seeds due to big wins over Nevada and Utah State, respectively, with the Lobos, Rams, Aztecs, and Wolf Pack all at 20-6! The 21-5 Aggies fell to a 7-seed from a 5-seed after the 75-55 loss to Colorado State.
Kentucky and Texas Tech were the big winners here. The Wildcats are up from an 8-seed to a 6-seed after beating 3-seed Auburn on the road (as well as drubbing Mississippi at home). The Red Raiders gained almost 15% in odds and two seeds despite falling at Iowa State on Saturday (as expected); prior to that, they beat Kansas 79-50 and were set to show up as a 6-seed. Not quite!
St. Mary's also added double digit Dance Chance odds, mostly a bump in Strength from beating Pepperdine 103-59. This makes their odds of winning out better—even though the odds still say they drop one of their last four outings. And the Florida Gators, a bubble team for most of the year, is looking like a solid tournament team now at 83.2% after winning 7 of their last 8 games.
Washington State is another hot team right now; the Cougars have won 7 straight and 10 of 11; they face 1-seed Arizona on Thursday trying to sweep the series (they won 73-70 in the home contest). Also under-the-radar and moving from a 10-seed to an 8-seed is Nebraska, who beat Penn State at home (which is not impressive) by 19 points (which is impressive). The Cornhuskers' last five games are all winnable.
Lots of movement onto the 9-seed line, from below and above. Boise State and Gonzaga improved their lot, the Broncos beating Fresno State 90-66 and the Zags winning 91-74 and 102-76 (thought the latter home win, over #349-Strength Pacific, was a slight underperformance!). But Oklahoma and Wake Forest dropped down two seeds each, the former due to losses to 4-seed Baylor and 2-seed Kansas (at home), the latter due to losses to 3-seed Duke and 10-seed Virginia.
The Cavaliers also fell despite that win, as they were upset by Pittsburgh a few days earlier. And Florida Atlantic dropped a seed with another loss in the American Athletic, this time to South Florida. The Owls viability as an at-large team is starting to look shaky; right now they're probably safe without winning their conference tournament, but they can't lose too many more and still say that.
Bubble teams: First, let's look at the 800-lb elephant in the room (someone isn't feeding that elephant, btw) Indiana State, who lost twice last week and fell from a projected 6-seed all the way to an 12-seed play-in team, losing 33.7% of their at-large odds. It wasn't the road loss to 17-10 Southern Illinois that did the damage, but the home loss to 13-14 Illinois State. That loss dropped them to an 11-seed mid-week, down 26.5% already before the 2nd loss added to their woes. I think the Sycamores were distracted by their entry into the AP top 25. Whatever the cause, a team in a conference like the Missouri Valley going from 3 losses to 5 losses is a big deal. They're now 1 spot behind Drake and both aren't getting in. The Bulldogs enter our seeding for the first time in quite a while, and in current snapshot they're at a 9-seed.
Also entering the seedings is Pittsburgh, fresh off a win at Virginia and an 86-59 beatdown of Louisville. Dropping out are Texas A&M, who lost to Vanderbilt by a point and Alabama by 25 points, and are just 15-10. Their odds drop from 62% to 45%. Butler also has 10 losses now and their odds fell from 59% to around 40% following losses to Marquette and Creighton, both at home, the latter by 22 points. Those aren't games the Bulldogs were expected to win, but playing at home they had a decent chance in both, and their odds of making the tournament depends on winning some games like that; those opportunities are gone now.
South Carolina hid a big bump in the road, taking on loss #4 and #5 last week. The loss to Auburn wasn't unexpected but they fell by 40 points; maybe that caused a hangover to the 1-point home loss to LSU and that's what really hurt them. They're still in a lot better shape than Mississippi, the other SEC team with a schizm between Success and Strength. The Rebels broke a 3-game skid with a narrow home win over Missouri but they're still projected to come up a bit short (despite showing as Last Four In team in the current "snapshot").
Other teams on the wrong side of the waterline include Seton Hall, who makes a huge upward move this week from 23% to 47% odds after topping Xavier and St. John's. 17-9 might not sound so great but it's a lot better than 15-9. When a team's record is too close to .500 it's much harder to make a case as an at-large, especially when you add in the expected loss in the conference tournament. Grand Canyon would be in if the tournament selected today, but even one more loss puts them on the outside needing a WAC title to get in. For a minor conference, the difference between having 2 and 3 losses can be immense.
Colorado is also in the First Four Out at the moment, with Utah being the Last Team In. The Pac-12 has just one certainty (Arizona) and one team looking very good now (Washington State). The Buffaloes and the Utes play on Saturday in a very important Bubble game.
Here are some more teams making moves of note. McNeese State has been hanging around the bottom of the bubble ever since upsetting Michigan in Ann Arbor last December, but never making it to true at-large range. They've only lost once since then and if they can win out they might be in position for an at-large, but the conundrum for minor conferences is in play here: a loss in their tournament might put them back out, while winning the tournament makes them an auto-bid.
Ohio State scored the big win of the week, beating Purdue. The Buckeyes' odds had fallen to 3.9% mid-week after their loss to Wisconsin, but they're back in play—sort of—after beating the Boilermakers. They're still just 15-11 so unless their "new coach effect" carries through the Big Ten tournament they're still a long shot. But their last five games are pretty winnable—if, in fact, they've found new life rather than just a temporary jump-start. Penn State, on the other hand, dropped to 0.0% after falling to 12-14. The drop was pretty far, as #126 is well out of the 0.1% range. The Lions fell to Michigan State at home and Nebraska (by 19) on the road, so it's a number of factors—mostly that they project to 14-17 now and squandered the opportunity those two games afforded them.
North Texas fell to 0.0% as well, all the way from a reasonable 0.6%, also a big plunge. The Mean Green actually beat Memphis on Thursday, so the entire plunge to 0.0 was caused by the 71-62 loss to UAB. They were favored in that game, and when you're 14-10 you need to win those I guess. Speaking of the American, Memphis took another hit in odds, falling 14.1% to just 8.2% as the Tigers lost not only to the Mean Green but also to SMU, 106-79. Once a projected 7-seed at 10-2, the Tigers were cursed by Ken Pomeroy who declared them overrated, after which they went on a 4-game skid.
Big East busts: Xavier skidded almost 10% but it was St. John's who fell the most, dropping almost 25% to just 13.9% odds after losses to Providence and Seton Hall put them at 14-12. The Musketeers are even worse at 13-12 and 7.2%. You can't win em all but you have to win some of the tough ones, and neither team has lately: The Red Storm have lost 8 of 10 and Xavier 4 out of 7—though they did beat St. John's a few weeks ago.
The ACC actually has a team in the plus column: NC State, who beat 5-seed Clemson on the road, 78-77 after trailing almost all the 2nd half. They're offset by Miami, another team cursed by Pomeroy, who lost to Clemson and Boston College, their 4th loss in a row. They face Duke and North Carolina in their last 5 games, which is daunting but the opportunity of those games is responsible for most of their remaining 8.3% at-large hope.
And I was hoping to talk about UCLA this week and their 7-game win streak in Pac-12 play but they lost to Utah by a point, so instead of having a measurable at-large percentage, they're still in the <0.1% limbo. It's miraculous that they've held on this long after starting 6-10. They're clearly a better team now: they lost to Utah by 46 points a month ago.
Posted on February 19, 2024 at 07:52 PM in analysis, commentary | Permalink | Comments (0)
Reblog (0) | |