PDA

View Full Version : 2012 NCAA Tournament: Bracketology



Pages : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 [31] 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51

CLS
03-05-2012, 10:18 AM
I remember that year, Priceless. And, I agree, since the precedent has been set, that is probably how it would go.

However, and this is in the realm of "how can we get a better system?" rather than, "what is the bracket going to be?", it is my strong opinion that it was wrong then, and it would be wrong now, again, to do it like that. It seems like a basketball system being imposed on hockey, when the 2 tournaments are very different. In basketball, not so much difference between #16 seeds. Hockey - huge difference between #4s. And, this year, if things hold as they are now, BC is a prohibitive #1 overall.

Simply put, the PWR is not a definite enough tool, and the teams are too close, to say "This is the #8 team." "This is the #9 team."

Anyway, sorry for the rant. I know there are problems with trying something else. I would favor using PWR to choose the field, and giving the committee more flexibility in doing the bracket. But, then, there would be many complaints of back-room dealings, and quid pro quos, from many fans who wouldn't like the result.

Oh, well...It seems to me what you're really disagreeing with is that the committee has flexibility, but you don't agree with the way flexibility is exercised, namely by forbidding first round intra-conference matchups. That's the one piece of flexibility in adjusting the strict 1 v 16, 2, v 15 ... matchups they have that they consistently exercise. Or did I misunderstand what you were proposing? I think your point of view is reasonable, but I do like the no intraconference first round matchups rule. To me, part of the charm of the playoffs is to see some different teams play each other.

Also, if the PWR is not definintive enough to say "This is the #8 team." "This is the #9 team," then in my view, it's not definitive enough to say that BC is a prohibitive #1.

Priceless
03-05-2012, 10:52 AM
Is it conceivable that BU could pass BC by winning hockey East and be able to go to Worcestor instead of the Eagles because the way BU has been playing lately I think the committee should be able to take that into account and if we won hockey East I think the committee should consider it.

Yes, if BU should pass BC and both remain #1 seeds, the committee would send BU to Worcester and BC to Bridgeport.

The script for the weekend:


20120309 Pv 0 ML 1 nc
20120309 MA 0 BC 1 nc
20120309 NH 0 BU 1 nc
20120309 Mr 0 Me 1 nc
20120309 AA 0 Mn 1 nc
20120309 MT 0 CC 1 nc
20120309 Wi 0 DU 1 nc
20120309 BS 0 ND 1 nc
20120309 NO 0 SC 1 nc
20120309 Mk 0 MD 1 nc
20120309 Qn 0 Cg 1 nc
20120309 Da 0 Cr 1 nc
20120309 Ya 0 Ha 1 nc
20120309 RP 0 Un 1 nc
20120309 MS 0 Mm 1 nc
20120309 BG 0 FS 1 nc
20120309 Nt 0 Mi 1 nc
20120309 LS 0 WM 1 nc
20120309 Ct 0 AF 1 nc
20120309 HC 0 Mh 1 nc
20120309 RM 0 Ni 1 nc
20120309 By 0 RT 1 nc

Copy this for Saturday's games. Change the 0 to a 2 for a road victory - and remember to add a 3rd game to a series if you project a road win in any game. Exact game dates are not important (the UML-PC series starts on a Thursday, for example). Most sites will have a YATC script set up for the conference semis and finals next week.

Numbers
03-05-2012, 11:47 AM
It seems to me what you're really disagreeing with is that the committee has flexibility, but you don't agree with the way flexibility is exercised, namely by forbidding first round intra-conference matchups. That's the one piece of flexibility in adjusting the strict 1 v 16, 2, v 15 ... matchups they have that they consistently exercise. Or did I misunderstand what you were proposing? I think your point of view is reasonable, but I do like the no intraconference first round matchups rule. To me, part of the charm of the playoffs is to see some different teams play each other.

Also, if the PWR is not definintive enough to say "This is the #8 team." "This is the #9 team," then in my view, it's not definitive enough to say that BC is a prohibitive #1.

This is an excellent post. It helps me understand my own thinking. What I am really saying is that the committee has flexibility, but it is in the wrong pieces of the puzzle. As far as I understand it, the committee has the flexibility to say "#5,6,7,8 are all the same to us, so move them around however we want, to avoid intraconference matchups." I am OK with that.
But the committee lacks the flexibility to say, "#8, 9, 10 are so similar as to be interchangeable." They can't move the #9 or 10 into the #2 seed band. And, I don't like that. Sometimes those 7,8,9,10 seeds are only broken by the RPI tiebreaker. Or, the 12,13,14 are only broken by the RPI tiebreaker. And, that is a problem.

As for the BC is a prohibitive #1, I should be more clear. PWR is an imprecise tool when distinguishing between, for example, right now, BU and Minn. Their RPIs are close, the TUCs are close. In almost every year, I would agree that it is not a precise enough tool to say, "This team is #1." But, not this year. Look for a minute at BC's RPI. It is better by a good measure than anyone else's. That's why I call them a prohibitive #1 right now. Obviously, I say that, and then Mass-Amherst will proceed to knock them out of the HE tourney, and their RPI will fall, and I will have lots of egg on my face. But, right now, given the figures we have, not only in the comparisons, but in the results of games, RPI, etc, BC is #1, and by a good bit. At least in my view.

Right now, #2,3,4, not so much. I think it's pretty much a dead heat. If BC was not sitting above all of them, I wouldn't be worried at all about which of those faced the AHA champ. There is no way to say, clearly, which is better.

I hope that helps clarify my thoughts.

Numbers
03-05-2012, 11:59 AM
It seems to me what you're really disagreeing with is that the committee has flexibility, but you don't agree with the way flexibility is exercised, namely by forbidding first round intra-conference matchups. That's the one piece of flexibility in adjusting the strict 1 v 16, 2, v 15 ... matchups they have that they consistently exercise. Or did I misunderstand what you were proposing? I think your point of view is reasonable, but I do like the no intraconference first round matchups rule. To me, part of the charm of the playoffs is to see some different teams play each other.

Also, if the PWR is not definintive enough to say "This is the #8 team." "This is the #9 team," then in my view, it's not definitive enough to say that BC is a prohibitive #1.

CLS,
I should make one more comment. I like the idea of avoiding first round intraconference matchups, I really do. So, let me give you a for example: Let's say the year ends with a PWR like this:
1-BC; 2-FSU; 3-Mich; 4-Minn.....12-Cornell; 13-UND; 14-MichState; 15-WestMich; 16th team - Air Force. And, let's say that the difference in comparisons won between Cornell and Michigan State is (1) comparison.
There are several options:
A) BC v WestMich; FSU v Air Force; Mich v UND; Minn v MichState, ??? v Cornell
B) BC v Air Force; FSU v UND; Mich v West Mich; Minn v Mich State; ??? v Cornell, for example.
c) BC v Air Force; FSU v UND; Mich v Cornell; Minn v WestMich; ??? v Mich State

A is as the committee has done. I think it unfair to BC, especially if they with the HE tourney, because they will be a very solid overall #1.
B Protects BC, and Western and Michigan only played twice this year, and those games were a long time ago. This doesn't seem too bad to me. I know it's an intraconference matchup, but it could be worse...
C to me is best. It adjusts the seeds a little. But, it gives BC what BC deserves, and it avoids intraconference matchup, all for swapping #12 with #14. I don't think that is too bad. But, right now, the committee can't do that - it's not allowed directly by their rulebook.

That is the inflexibility that I don't care for.

Thanks.

davyd83
03-05-2012, 12:15 PM
Love to read you guys' thoughts!! Fun, isn't it? Tbh, I don't know which I would predict. I go both ways. Michigan, as a higher #1, should have priority to Green Bay. But, they have to fly. UMD can drive either way. Michigan's flight is cheaper into Mpls, than to GB, I would imagine, so tends to Mich in St Paul. But, that's tough assignment, Mich against Minn in Round 2.

Then, I think: Look, Michigan, UM-L, and UMD, are all tied. I know, the RPI is tiebreaker, but still, they are all tied. Likewise, BU and Minny are really tied. So, if I want to spin the seeds, I could just as well say that UMD is a 3, and Minn is a 6, and then I am fine again.

So, I simply would not know how the committee approaches this problem.

Why would Michigan have to fly to Green Bay? It's about 450 miles.

Priceless
03-05-2012, 12:29 PM
Why would Michigan have to fly to Green Bay? It's about 450 miles.

Anything over 400 miles is considered a flight by the NCAA. Once you get on a plane it doesn't matter where it lands. I've had no luck trying to convince my travel agent of that :D

JF_Gophers
03-05-2012, 12:42 PM
They should throw all 16 teams in a hat and draw out teams one at a time to make the matchups. Then place them at each location after that. They would draw the (up to) 4 host teams first to make sure hosts don't face each other, since that would be impossible.

Priceless
03-05-2012, 01:12 PM
They should throw all 16 teams in a hat and draw out teams one at a time to make the matchups. Then place them at each location after that. They would draw the (up to) 4 host teams first to make sure hosts don't face each other, since that would be impossible.

They could just do what the BCS does and have #1 play #2 and everyone else plays a meaningless exhibition.

CLS
03-05-2012, 01:13 PM
...They can't move the #9 or 10 into the #2 seed band. And, I don't like that. Sometimes those 7,8,9,10 seeds are only broken by the RPI tiebreaker. Or, the 12,13,14 are only broken by the RPI tiebreaker. And, that is a problem.

...

I hope that helps clarify my thoughts.It does, I think. So you could:

(a) let the committee move teams across bands (but if you did that, I'd think you'd want to specify a maximum number of positions a team could be moved; for example, 5 and 12 are in adjacent bands, but I don't think you'd want to allow flip flopping of them) or

(b) allow the flip flopping of teams if they're within x RPI points, without regard to bands?

(b) has some appeal to me, since it recognizes that large diffrence in PWR and RPI may have some validitity to them, but small differences don't (I'm thinking of gas pumps that tell you how much gas you've pumped, to the thousandth of a gallon :)). Seems reasonable to me, and also another reminder of why a reason I like at least attempting to play at neutral sites. The tools we have may have some validity as to determining who plays who, but not to determine an advantage as big as home ice.

Dirty
03-05-2012, 01:15 PM
They could just do what the BCS does and have #1 play #2 and everyone else plays a meaningless exhibition.

I wish they did that last year. :(

Fighting Sioux 23
03-05-2012, 02:01 PM
It does, I think. So you could:

(a) let the committee move teams across bands (but if you did that, I'd think you'd want to specify a maximum number of positions a team could be moved; for example, 5 and 12 are in adjacent bands, but I don't think you'd want to allow flip flopping of them) or

(b) allow the flip flopping of teams if they're within x RPI points, without regard to bands?

(b) has some appeal to me, since it recognizes that large diffrence in PWR and RPI may have some validitity to them, but small differences don't (I'm thinking of gas pumps that tell you how much gas you've pumped, to the thousandth of a gallon :)). Seems reasonable to me, and also another reminder of why a reason I like at least attempting to play at neutral sites. The tools we have may have some validity as to determining who plays who, but not to determine an advantage as big as home ice.

Keep in mind, that this set up would likely almost always be used to kick a team out of the Tournament and include a bigger name team. Imagine that North Dakota is #16 and Merrimack is #15 in the final PWR, and North Dakota is only .0001 behind in RPI that would flip that comparison. Who do you think the committee would want to see play Michigan in St. Paul? I suppose you could specify language that the only time you can move teams around is after the 16 teams have been selected for the tournament.

Fighting Sioux 23
03-05-2012, 02:07 PM
Why would Michigan have to fly to Green Bay? It's about 450 miles.

What would be awesome is if they traveled by boat...although I'm not sure if the rivers connecting Ann Arbor to Lake Erie are passable :p:D:D

Numbers
03-05-2012, 02:53 PM
Keep in mind, that this set up would likely almost always be used to kick a team out of the Tournament and include a bigger name team. Imagine that North Dakota is #16 and Merrimack is #15 in the final PWR, and North Dakota is only .0001 behind in RPI that would flip that comparison. Who do you think the committee would want to see play Michigan in St. Paul? I suppose you could specify language that the only time you can move teams around is after the 16 teams have been selected for the tournament.

You raise a good point, FS 23. The discussion we were having was always under the assumption that the PWR be used strictly to determine who is in. You can't let that be subjective.

Priceless
03-05-2012, 04:14 PM
Everything written here is based on what I have done with my scripts based on things that are LIKELY to happen. I do not do wacky scenarios, like RPI winning the ECAC or Mankato conquering the WCHA. There may very well be scenarios out there that do not match up with what is written below. If you can find them, more power to you. If they actually happen, you should have gone to Vegas. As Jim Dahl pointed out this morning, there are more than just a few scenarios out there.

I'll be editing this as I go.

Who is in:
Boston College - The lowest I can get them is 7th. I can't get both Lowell and BU ahead of the Eagles. If the Hockey East title came down to them, only the winner would finish ahead of the Eagles. BC's astronomical RPI and stellar record vs TUC's keep them ahead of most teams, even if they get swept at the hands of UMass. If they sweep UMass it will be very difficult to unseat BC from the #1 seed overall.

Michigan is not in, but only by the narrowest of margins. I can get them to 12. Unless all four major conferences produce an upset, they are in.

Lowell is not in by a large margin. I can get them down to 15th rather easily.

Duluth appears to have a floor of 12th, so they are in the same boat as Michigan.

Numbers
03-05-2012, 05:21 PM
It does, I think. So you could:

(a) let the committee move teams across bands (but if you did that, I'd think you'd want to specify a maximum number of positions a team could be moved; for example, 5 and 12 are in adjacent bands, but I don't think you'd want to allow flip flopping of them) or

(b) allow the flip flopping of teams if they're within x RPI points, without regard to bands?

(b) has some appeal to me, since it recognizes that large diffrence in PWR and RPI may have some validitity to them, but small differences don't (I'm thinking of gas pumps that tell you how much gas you've pumped, to the thousandth of a gallon :)). Seems reasonable to me, and also another reminder of why a reason I like at least attempting to play at neutral sites. The tools we have may have some validity as to determining who plays who, but not to determine an advantage as big as home ice.

CLS,
Actually, my preference, in a perfect world where the committee members could be trusted to be fair and honest (I don't think that exists, really), would be to start at the top of the PWR, kick the committee member from the conference of the team in question out of the room, and vote "Is this the #1 team?" I think there would be 4 votes, so give the PWR the 5th vote. If yes, move to #2. If no, move to the second team and see if they are the #1 team. Stipulate that you can only fall or rise 3 spots from the PWR (or something like that). Create the bracket as you go.

But, like I said, that's in a perfect world. In the real world, there are always deals going on. So, you couldn't do that. Instead, i would be in favor of allowing as you said, .0020 in the RPI or something like that, or 1 PWR comparison won. And, only for the sake of avoiding intraconference matchups, attendance, or the integrity of the #1 seed v AHA Champ game.

Also, there is one other thing about that. In my example above, I swapped the 12 and 14 seeds. I have to think about one other point: Since the #14 is now playing in the #12's usual place, what happens if the #12 and #13 win in round 1, and play in the Regional Finals? Who gets last change?

For that matter, as it is now, if 2 #2 seeds play in the National Semis, who gets last change?

Thanks for the help.

DaveStPaul
03-05-2012, 05:55 PM
They could just do what the BCS does and have #1 play #2 and everyone else plays a meaningless exhibition.

And make sure the polls have two SEC teams in those slots, even if they're both 7-5!

mookie1995
03-05-2012, 07:15 PM
I cannot imagine that the better #1 seed will be seeded with a #2 or #3 seed Minnesota.

i cannot imagine that they would worry about a 2nd round game when seeding for first round games and brackets.

Fighting Sioux 23
03-06-2012, 12:54 PM
College Hockey Weekly has their Bracketology up...

http://www.collegehockeyweekly.com/Current_Pairwise.html

Also, here is their forecast of what the tournament will look like...

http://www.collegehockeyweekly.com/Tournament-Forecast.html

Priceless
03-06-2012, 04:21 PM
Also, here is their forecast of what the tournament will look like...

http://www.collegehockeyweekly.com/Tournament-Forecast.html

Attendance would be great in Bridgeport with 3 CCHA teams and Union. :p

You also have an all-CCHA first round game. That's incredibly unlikely.

Maine and Air Force go to Bridgeport and Miami and Western go to Worcester. Or BU and Union trade places. But the 1-16 pairing is gone. Just ask Cornell.

LTsatch
03-06-2012, 05:29 PM
I wish they did that last year. :(
Something we agree on!