2012 NCAA Tournament: Bracketology

[Fighting Sioux 23]

College Hockey Weekly has released their 2012 NCAA Tournament Bracketology (also their Forecast on what they think the tournament will actually look like). I'm sure a few of the other Bracketologies will be coming out soon from other sites as well. Anyway, I thought I would post a thread where we can discuss all the different bracketologies (and of course post our own).

Enjoy!

Here is the direct link for the CHW Bracketology...

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

Here is the direct link for the CHW Tournament Forecast...

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

 

[GopherBigGuy911]

Re: 2012 NCAA Tournament: Bracketology

College Hockey Weekly has released their 2012 NCAA Tournament Bracketology (also their Forecast on what they think the tournament will actually look like). I'm sure a few of the other Bracketologies will be coming out soon from other sites as well. Anyway, I thought I would post a thread where we can discuss all the different bracketologies (and of course post our own).

Enjoy!

Here is the direct link for the CHW Bracketology...

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

Here is the direct link for the CHW Tournament Forecast...

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

Oh how sweet that would be.... To be able to end No Dak's season in the X to move on to the Frozen Four and earn redemption for the last time they had that chance....

 

[Fighting Sioux 23]

Re: 2012 NCAA Tournament: Bracketology

Oh how sweet that would be.... To be able to end No Dak's season in the X to move on to the Frozen Four and earn redemption for the last time they had that chance....

It would be an interesting matchup that is for sure.

FWIW, the voting for the projected tournament was done by the same group of 8 that voted for the midseason All-American team, and the midseason awards.

Here was my ballot:

#1 Seeds: Ohio State, Boston College, Merrimack, Minnesota Duluth
#2 Seeds: Colorado College, Western Michigan, Union, Notre Dame
#3 Seeds: Minnesota, Boston University, Cornell, Michigan State
#4 Seeds: Denver, Atlantic Hockey Champion, UMass Lowell, Colgate

 

[GopherBigGuy911]

Re: 2012 NCAA Tournament: Bracketology

It would be an interesting matchup that is for sure.

FWIW, the voting for the projected tournament was done by the same group of 8 that voted for the midseason All-American team, and the midseason awards.

Here was my ballot:

#1 Seeds: Ohio State, Boston College, Merrimack, Minnesota Duluth
#2 Seeds: Colorado College, Western Michigan, Union, Notre Dame
#3 Seeds: Minnesota, Boston University, Cornell, Michigan State
#4 Seeds: Denver, Atlantic Hockey Champion, UMass Lowell, Colgate

Huh - you don't think No Dak has their act together yet?

 

[Fighting Sioux 23]

Re: 2012 NCAA Tournament: Bracketology

Huh - you don't think No Dak has their act together yet?

While I think North Dakota will be much improved in the second half of the season, I think they will fall just short of making the NCAA Tournament. The WCHA is not very strong this year, and that is shown by the current pairwise (and only 3 WCHA teams in the top 20 currently, with CC just barely making it in as the #15 overall seed). If North Dakota is able to continue to get great goaltending, and get healthy for the second half, we can be as good as anyone in the country. That being said, probably only three, maybe four WCHA teams will make the NCAA Tournament, and I think Duluth, CC and Minnesota will be virtual locks. That leaves Denver, North Dakota and Omaha to fight over one potential spot...assuming that there are no big upsets come tournament time. I think Denver will have the best second half of those three teams (I thought Denver would win the WCHA at the beginning of the season), and will sneak in as one of the last at-large selections for the NCAA Tournament.

 

[Priceless]

Re: 2012 NCAA Tournament: Bracketology

Which Pairwise are they using? How the COp component is calculated was changed. The USCHO Pairwise were corrected after I pointed out the flaw in the system. SS and Slack.net still use the incorrect calculation. For example, SS has NMU losing the COp with LSSU .625 to .500 however the margin is actually 2.500 to 2.000 because of the new way the COp is calculated.

Right now they happen to match, but I expect there will be differences as we get closer to the official selection.

 

[Fighting Sioux 23]

Re: 2012 NCAA Tournament: Bracketology

Which Pairwise are they using? How the COp component is calculated was changed. The USCHO Pairwise were corrected after I pointed out the flaw in the system. SS and Slack.net still use the incorrect calculation. For example, SS has NMU losing the COp with LSSU .625 to .500 however the margin is actually 2.500 to 2.000 because of the new way the COp is calculated.

Right now they happen to match, but I expect there will be differences as we get closer to the official selection.

I'm not sure which pairwise they were using. I would guess USCHO or CHN, but I'm not 100% certain. As there are no other Sioux fans in the group, I would doubt it is siouxsports, but it's possible.

 

[Patman]

Re: 2012 NCAA Tournament: Bracketology

Which Pairwise are they using? How the COp component is calculated was changed. The USCHO Pairwise were corrected after I pointed out the flaw in the system. SS and Slack.net still use the incorrect calculation. For example, SS has NMU losing the COp with LSSU .625 to .500 however the margin is actually 2.500 to 2.000 because of the new way the COp is calculated.

Right now they happen to match, but I expect there will be differences as we get closer to the official selection.

as long as A>B if and only if f(A)>f(B) who cares

2.5/2=.625/.5=1.25

I would say the whole number figuring is much more confusing... but its the same as long as you divide by the total number of common opponents

 

[LTsatch]

Re: 2012 NCAA Tournament: Bracketology

Cornell gets left out due to a lesser OOC schedule on both predictions. Can't see why they would not be in there at this point.

 

[gopher wes]

Re: 2012 NCAA Tournament: Bracketology

Looking forward to more releases in the second half of the season. Always love watching the bracketology unfold.

Thanks for the work you guys do over there and posting updates here for us to see.

 

[Fighting Sioux 23]

Re: 2012 NCAA Tournament: Bracketology

Cornell gets left out due to a lesser OOC schedule on both predictions. Can't see why they would not be in there at this point.

Cornell is a #4 seed in the Tournament Forecast (Northeast Regional).

 

[Fighting Sioux 23]

Re: 2012 NCAA Tournament: Bracketology

Looking forward to more releases in the second half of the season. Always love watching the bracketology unfold.

Thanks for the work you guys do over there and posting updates here for us to see.

Thanks. We will be doing a bracketology (as well as a tournament forecast) weekly until the end of the season.

 

[Priceless]

Re: 2012 NCAA Tournament: Bracketology

as long as A>B if and only if f(A)>f(B) who cares

2.5/2=.625/.5=1.25

I would say the whole number figuring is much more confusing... but its the same as long as you divide by the total number of common opponents

There can be differences. Earlier this year the comparison between Maine and Northeastern was different which flipped the pairing. I did analysis of the last few NCAA brackets and found that it would impact the bubble teams. I think (on my other computer I could look it up) it was Vermont who got left out of the field using this new formula.

Edit: It was a different change that impacted UVM. Here is my analysis from last summer:

After analyzing the Pairwise from last year not a single comparison changed among teams that were in the NCAA tournament or on the bubble. I examined the comparisons where flipping the winner of the COp component would make a difference (ie 1-1, 3-2 etc) and there were no changes. The team that won the COp component in the original Pairwise won with this new formula. So I'm not sure why this rule had to be changed.

The example provided by the NCAA:

This example provided by the NCAA shows how the new system of calculating results against common opponents would be scored.

Team A is 3–0 against Team C (1.000);
Team B is 1–0 against Team C (1.000);
Team A is 0–2 against Team D (0.000);
Team B is 1–2 against Team D (0.333);

Team B would be credited with having won the common opponents category with a 1.333 total in the individual opponent percentages compared to 1.000 for Team A. They both had a 1.000 winning percentage against Team C, and Team B had a better winning percentage than Team A against Team D (.000 for Team A and .333 for Team B). Under the current structure, Team A would be credited with having won the common opponents category as they were 3–2 (.600) against common opponents, while Team B was only 2–2 (.500).

So using the example provided before

against Colorado College--Michigan was 1-0, Minnesota-Duluth was 0-1-1

Michigan = 1.0000 UMD=.2500

against Lake Superior--Michigan was 2-0, Minnesota-Duluth was 0-0-1

Michigan = 1.0000 UMD = .5000

against Michigan Tech--Michigan was 1-0, Minnesota-Duluth was 4-0

Michigan = 1.0000 UMD = 1.0000

against Minnesota--Michigan was 0-1, Minnesota-Duluth was 1-1-2

Michigan = .0000 UMD = .5000

against Nebraska-Omaha--Michigan was 1-1, Minnesota-Duluth was 1-1

Michigan = .5000 UMD = .5000

against Northern Michigan--Michigan was 2-0, Minnesota-Duluth was 1-0

Michigan = 1.0000 UMD = 1.0000

against Wisconsin--Michigan was 0-0-1, Minnesota-Duluth was 3-1.

Michigan = .5000 UMD = .7500

So COp is Michigan 1+1+1+0+.5+1+.5= 5.0000 > UMD = .25+.5+1+.5+.5+1+.75= 4.5
Michigan wins the comparison 5 > 4.5833

Michigan still wins the overall comparison 3-0

It's going to be fun to see how long it takes the programming wizards at SS.com and slack.net to accurately duplicate the new format.

 

[CHFAN222]

Re: 2012 NCAA Tournament: Bracketology

The biggest question the second half of the season has to be how Boston University will respond to its mid season departures. I also would be cautious on Merrimack. They have struggled as of late. If Northeastern continues to play like they have been playing than they have a decent shot of snagging that final at large seed especially if they win the Minnesota tournament.

 

[css228]

Re: 2012 NCAA Tournament: Bracketology

Cornell gets left out due to a lesser OOC schedule on both predictions. Can't see why they would not be in there at this point.

No they make the tourney forecast as a four seed. Besides they're probably the safest bet to win the tourney.

 

[Fighting Sioux 23]

Re: 2012 NCAA Tournament: Bracketology

College Hockey Weekly has released their 2012 NCAA Tournament Bracketology for Week 2 (also their Forecast on what they think the tournament will actually look like).

Enjoy!

Here is the direct link for the CHW Bracketology...

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

Here is the direct link for the CHW Tournament Forecast...

http://www.collegehockeyweekly.com/Tournament-Forecast.html[/QUOTE]

 

[CHFAN222]

Re: 2012 NCAA Tournament: Bracketology

Interesting forecast.

Just wondering could you add a little section that shows which teams have been mathematically eliminated from gaining an invitation to the tournament and would have to win their conference tourney to get in? I'd imagine we will start seeing quite a few times knocked out over net two to three weeks.

 

[Patman]

Re: 2012 NCAA Tournament: Bracketology

Interesting forecast.

Just wondering could you add a little section that shows which teams have been mathematically eliminated from gaining an invitation to the tournament and would have to win their conference tourney to get in? I'd imagine we will start seeing quite a few times knocked out over net two to three weeks.

It would be very difficult to determine that

 

[JimDahl]

Re: 2012 NCAA Tournament: Bracketology

Which Pairwise are they using? How the COp component is calculated was changed. The USCHO Pairwise were corrected after I pointed out the flaw in the system. SS and Slack.net still use the incorrect calculation. For example, SS has NMU losing the COp with LSSU .625 to .500 however the margin is actually 2.500 to 2.000 because of the new way the COp is calculated.

Right now they happen to match, but I expect there will be differences as we get closer to the official selection.

Nope, you're misinterpreting. As Patman tried to note, SS and CHN are showing the average winning percentage vs. COPs instead of just the sum.

I personally think average winning percentages, e.g. .625 > .500, are a lot more intuitive than sum of winning percentages, e.g. 2.5 > 2.0. But the outcome will always be the same -- if A>B then A/n > B/n.

as long as A>B if and only if f(A)>f(B) who cares

2.5/2=.625/.5=1.25

I would say the whole number figuring is much more confusing... but its the same as long as you divide by the total number of common opponents

 

[jtwcornell91]

Re: 2012 NCAA Tournament: Bracketology

Nope, you're misinterpreting. As Patman tried to note, SS and CHN are showing the average winning percentage vs. COPs instead of just the sum.

I personally think average winning percentages, e.g. .625 > .500, are a lot more intuitive than sum of winning percentages, e.g. 2.5 > 2.0. But the outcome will always be the same -- if A>B then A/n > B/n.

That's right. The pairwise at TBRW?/slack.net also uses average winning percentages, although it's not spelled out as nicely in terms of team-by-team breakdowns as on CHN. For example, Denver wins the comparison with Michigan State thanks to common opponents, which are:
Air Force: both 1-0
MTU: Denver is 0-1-1, MSU is 1-0
BC: Denver is 1-0, MSU is 0-1
Under the old system, Denver would be 2-1-1 (.625) and lose to Michigan State's 2-1 (.667). But in the new system, Denver's averaged winning percentage is (1.000+.250+1.000)/3 = .750, while MSU's is still .667, so they win the criterion and with it the comparison. I didn't have the time to make the TBRW/slack version pretty, so it says "2-1-1 .7500 COp .6667 2-1" which is somewhat confusing, since it implies we're comparing 2-1-1 to 2-1, rather than .7500 to .6667. Similarly, USCHO's version says "2-1-1 2.250 1 COp 0 2.000 2-1-0" which is the same thing multiplied by 3 (the number of opponents). The CHN version leaves out the (irrelevant) overall record and includes an opponent-by-opponent breakdown. But all of them compare equivalent quantities, whether it's 2.25 vs 2.00 or .750 vs .667.