Re: March Madness: The Thread

[Andy]

Meanwhile, Gonzaga isn't doing so hot.

Rick Jackson's 3rd foul should've helped them out a lot. Whoops.

[Outshined_One]

Meanwhile, Gonzaga isn't doing so hot.

Rick Jackson's 3rd foul should've helped them out a lot. Whoops.

 

I think they're asleep.

[Andy]

Anyway, with that, I give. Pound away. Whatever. I would've thought if anyone understood how annoying it is to see your league dismissed because of a tiny sample size of postseason performance, it'd be Big Ten football fans. Apparently I was way off.

 

You take these things very seriously.

Yep. Just call me the Erik of college sports, I guess.

[Guest]

There's a reasonable chance that none of the big east teams could make the sweet sixteen, which would be awesome. Of their eight teams, they have three remaining in the tourney. Syracuse plays Gonzaga, which could be an outstanding game with Gonzaga potentially having an edge on the inside with Onuaku being out for the game. Third seeded Pitt plays a tough Xavier team and West Virginia will run up and down the court with Missouri.

 

If those three teams all lose, it would be a stunning exit for the Big East after being seeded 1, 2, 2, 3, 3, 6, 6 & 9. It isn't just that they got eight teams in the tourney. It's the number of high seeds the teams were given that makes this so remarkable.

 

Of course, if all three teams still alive win today, only four of the teams (that's half of them) will have underperformed their seeding to date. :)

Big East = 8 out of 16 teams in = 50 percent

Big 12 = 7 out of 12 teams in = 58 percent

ACC = 6 out of 12 teams in = 50 percent

 

BTW, the ACC could easily also be done by the Sweet 16 (certainly more likely than the Big East).

 

The Big East has the same fraction of teams in the dance as the ACC does, and LESS than the Big 12.

Way to completely ignore the overseeding point.

If you guys would [expletive] pay attention to me, I've been among the first to tell you that ND and Marquette were overseeded (I'm the one who bracketed those two teams into our selection committee bracket as 7 seeds). But you're all so excited to jump all over the Big East for losing a couple of games that you're willfully ignoring that.

 

Also, in my OWN S-curve for that bracket, I had ND slotted as a 10 seed. If you're trying to beat down someone who thought the Big East was a freaking superconference, you've got the wrong person.

The only teams you're talking about are ND and Marquette (and saying they're only overseeded by one single slot). However, the Big East's fifth team was still a freaking #3 seed. So the committee is saying that five of the top 12 teams in the country played in the BE. That's what I call severely overrating the conference. And it played out that way.

[Guest]

I think I underrated Syracuse, though.

 

Or at least Wesley Johnson.

[Andy]

Georgetown was a relatively unprecedented case. They played the best schedule in the country, the toughest Big East schedule, and racked up a ton of good wins with only one really bad loss (@Rutgers). They weren't a run of the mill middle-of-the-road team. I doubt we'll ever see anything like that again. Pittsburgh and Nova would be better examples of overseeded Big East teams than the Hoyas.

 

And of course, they got blasted by the #9 seed in the MAC. They make no sense.

[bukie]

Conference performance relative to seed, thus far (using simple higher seed wins expectations):

 

Pac 10 (8, 11); Expected wins: 1; actual wins: 3 (+2)

WCC (8, 10); Expected wins: 1; actual wins: 3 (+2)

MVC (9); Expected wins: 0; actual wins: 2 (+2)

ACC (1, 4, 7, 9, 9, 10); Expected wins: 3; actual wins: 4 (+1)

Horizon (5); Expected wins: 1; actual wins: 2 (+1)

Ivy League (12); Expected wins: 0; actual wins: 1 (+1)

OVC (13); Expected wins: 0; actual wins: 1 (+1)

CAA (11); Expected wins: 0; actual wins: 1 (+1)

MAC (14); Expected wins: 0; actual wins: 1 (+1)

Big Ten (2, 4, 4, 5, 11); Expected wins: 4; actual wins: 4 (0)

SEC (1, 4, 6, 10); Expected wins: 5; actual wins: 4 (-1)

Big 12 (1, 2, 3, 5, 7, 8, 10); Expected wins: 9; actual wins: 7 (-2)

Mountain West (3, 7, 8, 11); Expected wins: 4; actual wins: 2 (-2)

Atlantic 10 (5, 6, 7); Expected wins: 3; actual wins: 1 (-2)

Big East (1, 2, 2, 3, 3, 6, 6, 9); Expected wins: 8; actual wins: 3 (-5)

More information I already knew.

 

Look, this performance blows. I've said it before, I'll say it again even if Cuse, Pitt and WVU all make the Elite Eight and even if Cuse and WVU make the title game.

 

That doesn't invalidate an entire season's worth of data that made pretty clear that the top 3 leagues in America were the Big 12 first, and then the ACC and Big East 2nd and 3rd in some order. Unless you're a moron.

I didn't do this to rub in the Big East thing, I was just doing it to do it. I was curious how many leagues have over-underperformed (it's a zero-sum game, as the default is everyone performing exactly to expectations). All the same, a 9 beating an 8 shouldn't be worth as much weight as, say, a 14 beating a 3 or a 13 beating a 4, so a better weighting system would provide a better result. Maybe I'll do something like that if I get really curious.

[Andy]

Ohio State and WVU can either make me happy or destroy my bracket today. A&M will be important too since I have the Aggies in the Elite 8.

[Guest]

Conference performance relative to seed, thus far (using simple higher seed wins expectations):

 

Pac 10 (8, 11); Expected wins: 1; actual wins: 3 (+2)

WCC (8, 10); Expected wins: 1; actual wins: 3 (+2)

MVC (9); Expected wins: 0; actual wins: 2 (+2)

ACC (1, 4, 7, 9, 9, 10); Expected wins: 3; actual wins: 4 (+1)

Horizon (5); Expected wins: 1; actual wins: 2 (+1)

Ivy League (12); Expected wins: 0; actual wins: 1 (+1)

OVC (13); Expected wins: 0; actual wins: 1 (+1)

CAA (11); Expected wins: 0; actual wins: 1 (+1)

MAC (14); Expected wins: 0; actual wins: 1 (+1)

Big Ten (2, 4, 4, 5, 11); Expected wins: 4; actual wins: 4 (0)

SEC (1, 4, 6, 10); Expected wins: 5; actual wins: 4 (-1)

Big 12 (1, 2, 3, 5, 7, 8, 10); Expected wins: 9; actual wins: 7 (-2)

Mountain West (3, 7, 8, 11); Expected wins: 4; actual wins: 2 (-2)

Atlantic 10 (5, 6, 7); Expected wins: 3; actual wins: 1 (-2)

Big East (1, 2, 2, 3, 3, 6, 6, 9); Expected wins: 8; actual wins: 3 (-5)

More information I already knew.

 

Look, this performance blows. I've said it before, I'll say it again even if Cuse, Pitt and WVU all make the Elite Eight and even if Cuse and WVU make the title game.

 

That doesn't invalidate an entire season's worth of data that made pretty clear that the top 3 leagues in America were the Big 12 first, and then the ACC and Big East 2nd and 3rd in some order. Unless you're a moron.

I didn't do this to rub in the Big East thing, I was just doing it to do it. I was curious how many leagues have over-underperformed (it's a zero-sum game, as the default is everyone performing exactly to expectations). All the same, a 9 beating an 8 shouldn't be worth as much weight as, say, a 14 beating a 3 or a 13 beating a 4, so a better weighting system would provide a better result. Maybe I'll do something like that if I get really curious.

 

The other thing to note is that those expected wins include today but the actual wins don't yet. The Big East could be at -2 if Cuse, WVU, and Pitt win.

[Exile on Waveland]

Conference performance relative to seed, thus far (using simple higher seed wins expectations):

 

Pac 10 (8, 11); Expected wins: 1; actual wins: 3 (+2)

WCC (8, 10); Expected wins: 1; actual wins: 3 (+2)

MVC (9); Expected wins: 0; actual wins: 2 (+2)

ACC (1, 4, 7, 9, 9, 10); Expected wins: 3; actual wins: 4 (+1)

Horizon (5); Expected wins: 1; actual wins: 2 (+1)

Ivy League (12); Expected wins: 0; actual wins: 1 (+1)

OVC (13); Expected wins: 0; actual wins: 1 (+1)

CAA (11); Expected wins: 0; actual wins: 1 (+1)

MAC (14); Expected wins: 0; actual wins: 1 (+1)

Big Ten (2, 4, 4, 5, 11); Expected wins: 4; actual wins: 4 (0)

SEC (1, 4, 6, 10); Expected wins: 5; actual wins: 4 (-1)

Big 12 (1, 2, 3, 5, 7, 8, 10); Expected wins: 9; actual wins: 7 (-2)

Mountain West (3, 7, 8, 11); Expected wins: 4; actual wins: 2 (-2)

Atlantic 10 (5, 6, 7); Expected wins: 3; actual wins: 1 (-2)

Big East (1, 2, 2, 3, 3, 6, 6, 9); Expected wins: 8; actual wins: 3 (-5)

More information I already knew.

 

Look, this performance blows. I've said it before, I'll say it again even if Cuse, Pitt and WVU all make the Elite Eight and even if Cuse and WVU make the title game.

 

That doesn't invalidate an entire season's worth of data that made pretty clear that the top 3 leagues in America were the Big 12 first, and then the ACC and Big East 2nd and 3rd in some order. Unless you're a moron.

I didn't do this to rub in the Big East thing, I was just doing it to do it. I was curious how many leagues have over-underperformed (it's a zero-sum game, as the default is everyone performing exactly to expectations). All the same, a 9 beating an 8 shouldn't be worth as much weight as, say, a 14 beating a 3 or a 13 beating a 4, so a better weighting system would provide a better result. Maybe I'll do something like that if I get really curious.

 

Yeah, I think a more in-depth analysis is needed. Upsets don't happen in a vacuum. I don't really care if a 9-seed beats an 8-seed; that's not really under-performing much. But a 14-seed over a 3-seed, or a 9-seed over a 1-seed, that's under-performing.

[Andy]

The performance relative to expectations is a cruel game for multi-bid leagues. The Big 12 has performed admirably in the tournament - 2 teams already in the Sweet 16, 2 more with a chance to play for it. And yet they're at -2 because of Texas and Kansas.

[UK1679666180]

The only thing worth noting is now that KU has been upset, UK clearly has the advantage of being the most talented team left. Unfort., they are also one of the youngest.

 

 

(Fwiw, I still have them losing to WV)

[Guest]

I've seen the weighted expectations before, like the average number of games an X seed should win in the tournament. But I don't think it considers the seed of opponents beyond the first round, and I think you need to be at the end of the tournament to be able to do much measurement.

[bukie]

I've seen the weighted expectations before, like the average number of games an X seed should win in the tournament. But I don't think it considers the seed of opponents beyond the first round, and I think you need to be at the end of the tournament to be able to do much measurement.

I think it's possible to do results, but predictor work is tricky.

[Andy]

This is a good time to mention that Wes Johnson said he would've voted for Andy Rautins as Big East player of the year. That guy's a weapon. And the #1 reason that my pick of KSU to the Final Four is making me nervous right now.

[Outshined_One]

I feel bad for Gus Johnson. He's had some real clunkers so far.

[Exile on Waveland]

I've seen the weighted expectations before, like the average number of games an X seed should win in the tournament. But I don't think it considers the seed of opponents beyond the first round, and I think you need to be at the end of the tournament to be able to do much measurement.

 

That's probably not a bad idea. For instance, if George Mason would have beat Florida in the 2006 Final Four, surely that couldn't have been much "under-performance" by Florida considering what the Patriots had accomplished to that point. But, on paper, it would have been just any old No. 11 beating a No. 3.

[Guest]

I've seen the weighted expectations before, like the average number of games an X seed should win in the tournament. But I don't think it considers the seed of opponents beyond the first round, and I think you need to be at the end of the tournament to be able to do much measurement.

 

That's probably not a bad idea. For instance, if George Mason would have beat Florida in the 2006 Final Four, surely that couldn't have been much "under-performance" by Florida considering what the Patriots had accomplished to that point. But, on paper, it would have been just any old No. 11 beating a No. 3.

 

I don't even think it's that complex, at least what I'm referring to. What I've seen(mostly in reference to coaches), is the average # of games won for anyone rated a X seed at the beginning of the tournament. So if 3 seeds typically win 2.35 games(which encompasses all previous 2nd-Champ round opponents), Georgetown is -2.35 for the tournament.

[Careless]

Is it really smart for CBS to have a 1-8/9 game be the only game going on when there's going to be overlap later in the day?

[Guest]

Is it really smart for CBS to have a 1-8/9 game be the only game going on when there's going to be overlap later in the day?

 

I bet all the Gonzaga fans appreciated the 9 AM local tip time to see their team get taken to the woodshed.

[Outshined_One]

Four games going on at once?

 

Groan. Would it have killed CBS to have one of these games on at the same time as Cuse/Gonzaga?

[Sosa21MVP]

I think the Big Ten goes 2-2 today. Ohio State and Michigan State win, Wisconsin and Purdue lose. I realize I'm not exactly going out on a limb there.

[Guest]

I think the Big Ten goes 2-2 today. Ohio State and Michigan State win, Wisconsin and Purdue lose. I realize I'm not exactly going out on a limb there.

 

Hoping for a 4-0 day from the Big 10.

[Outshined_One]

GTech/Ohio State has been impressively sloppy so far.

[Outshined_One]

Sweet, my CBS affiliate is showing Wisconsin/Cornell.

 

What should my MMOD game be: Mizzou/WVU, MSU/Maryland, or GTech/Ohio State?