Pairwise and Bracketology 2013 Edition

[JB]

Re: Pairwise and Bracketology 2013 Edition

Sorry, didn't make it back over here this afternoon. As you discovered, I now have full probabilities by team/by ranking/by number of wins posted:

Remaining PWR possibilities

Cool I have always thought this was possible to do. I am too busy to input every possibility or sit down and program it. For me every possibility would be faster as I haven't done much real programming in years.

Always thought is would be a fun website add on. You can either play with you are the committee or input you team and get a list of who to route for. As each game ticks off the field could get tighter. Get to Saturday and get an instant output of who wants a tie in the ECAC consolation game.

With a program you could even look one game back to the best of 3 weekend or even 2 weeks back. Computer could do every possible outcome. Could UNH still be a lock if they had lost out or are the tie against Maine and win in Providence critical to keeping a 3-5-5 in the last 13 currently hapless team in.

 

[Numbers]

Re: Pairwise and Bracketology 2013 Edition

OK. I thought it would be fun to analyze the possibilities like this:
Right now there are 28 teams with hope, and 7 teams that have clinched.
What happens on Thursday immediately starts affecting that number:

For example: If Mankato wins on Thurs Aft, then Mankato is in for sure (makes 8), and UW and Alaska are both out (I think in the case of Alaska, because they need the Mk compare, and if Mk wins Thurs, AK can no longer get that).
What else happens in the way of 'outs' or 'ins' if Mankato wins?
And, What happens if UW wins? Can anything be concluded by that?

Then, what about the CC/UND game? Of course, if CC loses, season over. What else can be assuredly concluded from the outcome of that one?

I actually can't find any other direct results from these 2 games. Notre Dame is still beholden to Michigan's TUC status. DU and SCSU are against each other if SCSU loses on Friday, so Thursday doesn't decide anything, and the no one else can clinch with a 12 cutline.


Anyone have input?

 

[Numbers]

Re: Pairwise and Bracketology 2013 Edition

And, I know there are lots of RPI (the school) fans on this board. What needs to happen for RPI to make the field? I know that WI and CC losing on Thursday really helps. But, how much exactly?

And, also, WI and CC losing improves DU's chances. How much?

And, I think, although I am not sure, that if both Mankato and UND win, it clinches for SCSU, because it takes the WCHA out of what Priceless calls "the parlay." So, SCSU being 12th, would be safe. I believe they would hang on at 12th in this scenario, because they will assuredly win the compare with DU, regardless of their own results this weekend.

Edit again: First run proves me wrong. I was wrong about the DU/SCSU compare being in the bag for SCSU. Even if CC and Wi lose on Thurs, if Ct, Mich, Un, Prov win their tourneys, the cut line goes to 12, Niagara stays ahead of SCSU, DU wins the compare with SCSU because RPI, ComOpp and TUC all flip with one SCSU loss, and DU/SCSU end up tied in the PWR, with DU taking the last spot.

 

[goblue78]

Re: Pairwise and Bracketology 2013 Edition

Then, what about the CC/UND game? Of course, if CC loses, season over. What else can be assuredly concluded from the outcome of that one?

Anyone have input?

Another obvious implication is that the cut line possibility drops from 5 to 4, and, in practice, from 4 to 3. Thus, all cases in which your bottom seed probability is 13 you are safe. (This ignores the tiny possibility that Wisconsin can win out and still finish with a 17 PWR.)

 

[Priceless]

Re: Pairwise and Bracketology 2013 Edition

And, I know there are lots of RPI (the school) fans on this board. What needs to happen for RPI to make the field? I know that WI and CC losing on Thursday really helps. But, how much exactly?

And, also, WI and CC losing improves DU's chances. How much?

And, I think, although I am not sure, that if both Mankato and UND win, it clinches for SCSU, because it takes the WCHA out of what Priceless calls "the parlay." So, SCSU being 12th, would be safe. I believe they would hang on at 12th in this scenario, because they will assuredly win the compare with DU, regardless of their own results this weekend.

Edit again: First run proves me wrong. I was wrong about the DU/SCSU compare being in the bag for SCSU. Even if CC and Wi lose on Thurs, if Ct, Mich, Un, Prov win their tourneys, the cut line goes to 12, Niagara stays ahead of SCSU, DU wins the compare with SCSU because RPI, ComOpp and TUC all flip with one SCSU loss, and DU/SCSU end up tied in the PWR, with DU taking the last spot.

It is still likely that St Cloud - the #1 seed in the WCHA tournament - could be playing in the championship game Saturday night having still not clinched anything. Go figure.

 

[Numbers]

Re: Pairwise and Bracketology 2013 Edition

It is still likely that St Cloud - the #1 seed in the WCHA tournament - could be playing in the championship game Saturday night having still not clinched anything. Go figure.

Is this true if CC loses on Thursday night?

 

[Numbers]

Re: Pairwise and Bracketology 2013 Edition

Another obvious implication is that the cut line possibility drops from 5 to 4, and, in practice, from 4 to 3. Thus, all cases in which your bottom seed probability is 13 you are safe. (This ignores the tiny possibility that Wisconsin can win out and still finish with a 17 PWR.)

Do you say in practice because of the Niagara situation?

And interestingly my experimentation seems to indicate that even with both CC and UW losing on Thursday, and the cutline dropping to top-12 at least in, only Mankato clinches at that point.

I tried to explain above:
SCSU and DU can both end up just on the wrong side of the line even with CC/UW both out.
Notre Dame has a situation where Michigan's results are almost as important as their own.
Yale can lose twice, and so anything Thursday helps them, but doesn't clinch.
Niagara can't clinch without a win and/or the cutline lowering yet.
WMU, etc still need more help.

At least that's how I see it.

The picture won't get any clearer much until Friday when Michigan/Yale/Niagara, etc play, except for the teams on the ice on Thursday.

 

[JF_Gophers]

Re: Pairwise and Bracketology 2013 Edition

I don't think Niagara is eliminated simply because they didn't end up with 10 TUC games. I would assume they are eligible and that the league could get 2 teams in the tournament until you see definitively to the contrary. Also it looks like the "you are the commitee" program is including them in the tournament even if they lose the title game (I had them at #8 in that scenario).

10 TUC games is a pretty arbitrary way to eliminate a team considering they have less control over that than the winning and losing of games in general.

 

[goblue78]

Re: Pairwise and Bracketology 2013 Edition

Numbers: I say, "In practice" because once WCHA doesn't get an AQ, then the probability of 4 AQs is only 0.7% under KRACH. (Admitttedly, that's not down much from the approximately 1 percent probability while CC is still active.)

 

[Numbers]

Re: Pairwise and Bracketology 2013 Edition

Numbers: I say, "In practice" because once WCHA doesn't get an AQ, then the probability of 4 AQs is only 0.7% under KRACH. (Admitttedly, that's not down much from the approximately 1 percent probability while CC is still active.)

Got it: The odds of the "parlay" go to less than 1%. Cool.

 

[Numbers]

Re: Pairwise and Bracketology 2013 Edition

I don't think Niagara is eliminated simply because they didn't end up with 10 TUC games. I would assume they are eligible and that the league could get 2 teams in the tournament until you see definitively to the contrary. Also it looks like the "you are the commitee" program is including them in the tournament even if they lose the title game (I had them at #8 in that scenario).

10 TUC games is a pretty arbitrary way to eliminate a team considering they have less control over that than the winning and losing of games in general.

This is my understanding, too. And, I believe the understanding of Priceless, JimDahl and Craig P. and others who have posted here this year and in past years. I have seen lots on comments to the Bracketology blog on the main page about this, and I think that all who say "<10 games v TUC, no at-large bit" haven't looked close enough. The 10 games limit is the point at which that piece of the compare kicks in. It basically means that in Niagara's case, they are being evaluated solely on the basis of RPI.

 

[Craig P.]

Re: Pairwise and Bracketology 2013 Edition

The 10 games limit is the point at which that piece of the compare kicks in. It basically means that in Niagara's case, they are being evaluated solely on the basis of RPI.

And COP and H2H, where applicable.

 

[Felger]

Re: Pairwise and Bracketology 2013 Edition

It is still likely that St Cloud - the #1 seed in the WCHA tournament - could be playing in the championship game Saturday night having still not clinched anything. Go figure.

While at the same time UNH - the #5 seed in Hockey East - has already clinched, even though eliminated from their own tourney.

 

[Patman]

And COP and H2H, where applicable.

Right but if there isn't any H2H then the comparison is won solely on RPI

 

[80Miracle]

Re: Pairwise and Bracketology 2013 Edition

If St Cloud State misses the NCAA Tournament after winning (sharing with Minny) the toughest league in the country it might be time to "redo" the selection process. How about every league gets two bids (I know the stronger leagues hate this) with a Regular Season Champion and Playoff Champion (runner-up if same team). That should give you ten teams THEN contact MIT (just kidding) to Pairwise the remaining six teams to fill out the field. Just an idea

 

[JB]

Re: Pairwise and Bracketology 2013 Edition

If St Cloud State misses the NCAA Tournament after winning (sharing with Minny) the toughest league in the country it might be time to "redo" the selection process. How about every league gets two bids (I know the stronger leagues hate this) with a Regular Season Champion and Playoff Champion (runner-up if same team). That should give you ten teams THEN contact MIT (just kidding) to Pairwise the remaining six teams to fill out the field. Just an idea

This would take us back to the old CC rule. CC won WCHA once upon a time and missed the tourney. That old CC rule was for the 12 team tournament and if you pulled a double (win regular and conference tourney) you would get one of the 4 byes. That business went away when the tournament expanded to 16 teams.

If SCSU in the "toughest" conference had taken care of out of conference games for example against the 5th place hockey east team twice than it wouldn't be an issue. If they hadn't lost to Northern Michigan it might have helped too, the poor .500 OOC record with 2 win against Huntsville is the problem. SCSU has it in their power to get in WIN. Otherwise that toughest conference's record against say the 5th place hockey east team (0-3-1) will cause issues for some members... The system is fine SCSU can win their way in or they don't get in, everybody new the rule before the season.

 

[Numbers]

Re: Pairwise and Bracketology 2013 Edition

If St Cloud State misses the NCAA Tournament after winning (sharing with Minny) the toughest league in the country it might be time to "redo" the selection process. How about every league gets two bids (I know the stronger leagues hate this) with a Regular Season Champion and Playoff Champion (runner-up if same team). That should give you ten teams THEN contact MIT (just kidding) to Pairwise the remaining six teams to fill out the field. Just an idea

Nice idea, too. But, it also needs expressing that SCSU lost twice to New Hampshire, and twice to Northern Michigan, and only split with RPI. They did sweep Huntsville. But, collectively, those results balance down their sharing of the MacNaughton.

 

[goblue78]

Re: Pairwise and Bracketology 2013 Edition

Wasn't Yale an at-large team without 10 TUC games in 2009-10? Anybody got a database of old end-of-season PWR? IIRC (and there's absolutely no assurance that I do) that was part of the impetus for expanding the number of TUCs.

 

[Patman]

Nice idea, too. But, it also needs expressing that SCSU lost twice to New Hampshire, and twice to Northern Michigan, and only split with RPI. They did sweep Huntsville. But, collectively, those results balance down their sharing of the MacNaughton.

Agreed.. It has taken me years to emphasize with other Lowell fans that non-conference matters.

 

[unh1982]

Re: Pairwise and Bracketology 2013 Edition

If St Cloud State misses the NCAA Tournament after winning (sharing with Minny) the toughest league in the country it might be time to "redo" the selection process. How about every league gets two bids (I know the stronger leagues hate this) with a Regular Season Champion and Playoff Champion (runner-up if same team). That should give you ten teams THEN contact MIT (just kidding) to Pairwise the remaining six teams to fill out the field. Just an idea

If SCS loses twice to my team UNH , they dont deserve to make the ncaa , heck even UNH doesnt deserve to be there this year.