2015 Pairwise Discussion & Predictions

[ne7minder]

Re: 2015 Pairwise Discussion & Predictions

So UMD is totally out. No chance. Right?

Pretty sure that UND will replace them but UND is at 8th so they should not make it either unless they can win the WCHA tourney*

*subject to the appearance of random seeding by the NCAA

 

[ARM]

Re: 2015 Pairwise Discussion & Predictions

So UMD is totally out. No chance. Right?

Well, there's always the possibility of litigation due to some form of bias by the PairWise Rankings, although there is a set precedent for Eastern Bias, so I don't know if that claim would fly in court.

 

[Numbers]

Re: 2015 Pairwise Discussion & Predictions

So, as of tonight, we would assume:
1-MN
2-BC
3-Harv
4-UW
5-QU
6-CCT
7-BU
8-CHA Champ

And, since there are no western teams in the lower part of the bracket, then you have to fly 2. And, you could (although the committee does not have any rules about this, but it does NOT affect anything else) avoid conference games.

CHA @ MN
QU @ UW

CCT @ BC
BU @ Harvard

For all the wrangling, that seems ok. UND makes it in by winning WCHA. That would kick BU out. UND would play at MN, CHA at BC, CCT at Harvard, and QU still at UW. Not as nice, but still ok. Especially since, in that case, UND would have recently beaten MN and/or UW. Give them a rematch then, seems ok.

 

[TonyTheTiger20]

Re: 2015 Pairwise Discussion & Predictions

No, they would not rearrange to avoid intraconference matchups. It affects bracket integrity. The women's handbook has two criteria: 1) Minimize flights (required) and 2) preserve bracket integrity as much as possible without adding flights.

 

[Numbers]

Re: 2015 Pairwise Discussion & Predictions

No, they would not rearrange to avoid intraconference matchups. It affects bracket integrity. The women's handbook has two criteria: 1) Minimize flights (required) and 2) preserve bracket integrity as much as possible without adding flights.

Correct, Tony. But look at those RPIs and tell me there really is a difference. It's my understanding the women's committee has more liberty to use their judgment rather than being bound to a difference in the 4th decimal place, like the men's committee is bound. I think they could find reason to seed BU ahead of CCT to avoid playing BU/BC and Harvard/Clarkson again.

All they have to do is say "Hey, the 2 h2h wins by BU seemed stronger to us than the .0006 edge in RPI." Or, it seemed to make more sense in this case to use the individual comparison as a tb, rather than such a small RPI edge. Or, they could say "We looked closely at the UMD/BU comparison, and BU's large RPI edge seemed more important than the slight edge in CommOpp and TUC that UMD had."

Or, am I wrong and they can't do that?

 

[Eeyore]

Re: 2015 Pairwise Discussion & Predictions

Any scenario that has both Minnesota and BC winning their tournaments still leaves BC with an RPI more than .005 better than Minnesota. Given how close the other criteria are, that may be enough for the committee to decide to set up a potential Minnesota/Wisconsin semi-final if that's what they really want.

 

[Numbers]

Re: 2015 Pairwise Discussion & Predictions

Any scenario that has both Minnesota and BC winning their tournaments still leaves BC with an RPI more than .005 better than Minnesota. Given how close the other criteria are, that may be enough for the committee to decide to set up a potential Minnesota/Wisconsin semi-final if that's what they really want.

Makes sense Eeyore. Any scenario with UM and BC winning their tourneys also leaves UW in limbo. For example, if Harvard loses their semi, and UW loses the final to Minny, what happens to the 3/4 seeds? That whole business is still way up for grabs. Also, what if UND beats UW, and then Minny wins. If QU beats HU, can QU climb to 4th, and host?

 

[TonyTheTiger20]

Correct, Tony. But look at those RPIs and tell me there really is a difference. It's my understanding the women's committee has more liberty to use their judgment rather than being bound to a difference in the 4th decimal place, like the men's committee is bound. I think they could find reason to seed BU ahead of CCT to avoid playing BU/BC and Harvard/Clarkson again.

All they have to do is say "Hey, the 2 h2h wins by BU seemed stronger to us than the .0006 edge in RPI." Or, it seemed to make more sense in this case to use the individual comparison as a tb, rather than such a small RPI edge. Or, they could say "We looked closely at the UMD/BU comparison, and BU's large RPI edge seemed more important than the slight edge in CommOpp and TUC that UMD had."

Or, am I wrong and they can't do that?

There is absolutely no directive nor mention to avoid intraconference matchups. They will keep bracket integrity because that's what the handbook tells them to do.

Ranking one team higher than another because of a big RPI difference despite losing the comparison is another story, however, because the handbook gives them that leeway. But that would be their justification -- teams were ranked differently, not because it gives the committee a more favorable bracket but becauae they deemed BU to be ranked higher than Clarkson.

Having said all that, I don't know what the threshold would be to tinker with the rankings. Personally I would prefer the rules be cut and dried rather than decided in a smoke filled room, despite the fact that such a decision (i.e. BC is #1 despite losing the comparison to UM because of RPI) would benefit BC this year.

 

[ARM]

Re: 2015 Pairwise Discussion & Predictions

I think the RPI difference has to be bigger than something like .005, which really isn't that large, before that would override the other criteria. If the committee would shift to only looking at RPI that easily, then it doesn't make sense to even have other criteria. Now if the RPI is several games different, then that is another matter, but in this case, it wouldn't be.

 

[TonyTheTiger20]

Re: 2015 Pairwise Discussion & Predictions

Oh I totally agree. I think the Clarkson/UMD pick would have been the one to watch for until the MIGHTY MIDJI came back to win, particularly given that Clarkson won its conference's regular season title.

 

[Numbers]

Re: 2015 Pairwise Discussion & Predictions

Oh I totally agree. I think the Clarkson/UMD pick would have been the one to watch for until the MIGHTY MIDJI came back to win, particularly given that Clarkson won its conference's regular season title.

And this is sort of the point. The 6/7 seed right now depends on two things.
First, who wins the UMD/BU comparison. Clarkson is currently ranked 6th because UMD wins that compare strictly according to TUCs and CommOpps. But, BU wins RPI by a full .020. So, if the committee wants, they can say that BU actually wins that compare. That being done, now BU is 6th and Clarkson is 7th. They don't need to tell us why; it is simply true that they can do that.
Second, if they decide that UMD still wins the compare, then the 6th/7th seed is still decided by RPI as a tiebreaker. I am not sure if that is coded into the rule book or not. If it is not, the RPI difference in this case is literally .0006. BU wins the compare with Clarkson. I believe the back room committee has the ability to decide that however they want. And, again, they don't have to tell us why. Or, they can tell us a reason , and it's not the real reason.

I understand there is nothing in the rule book about avoiding intraconference matchups. There is also nothing in the rule book about how, exactly, they committee is supposed to do these other tiebreakers. This is a committee with license to do whatever they want, except that they have to reduce flights if possible first. After that, they can find justification to do the minor details of the bracket according to how they want it to come out.

All that being said, however, my guess is that in the end, there will be no controversy.

However, while we are at it...... Suppose UND wins the WCHA (big upset, I know). Suppose Harvard beats Clarkson and BC beats BU in their conference finals. Do all of that, and now, UND comes in 7th, CHA 8th, and the controversy at 6th between BU and Clarkson becomes very important. I would hate to be on that committee. Their individual judgment of what is important would determine, not 'who plays where', but 'who is in'. Wow....

 

[ARM]

Re: 2015 Pairwise Discussion & Predictions

Suppose UND wins the WCHA (big upset, I know). Suppose Harvard beats Clarkson and BC beats BU in their conference finals. Do all of that, and now, UND comes in 7th, CHA 8th, and the controversy at 6th between BU and Clarkson becomes very important.

Don't worry too much about this hypothetical, because BU is going to beat BC. Grant already knows this -- he'll spend the rest of the week having the "Poulin Nightmare". His only hope is that Coyne takes out BU before BU gets to BC.

… or maybe BU beats BC in the NCAAs. These visions get so confusing, because sometimes the context is lacking.

Just kidding. Kind of. Can't you just see Poulin beating BC in her final game against the Eagles though? (* Play theme from "Jaws" here. *)

 

[dave1381]

Remember, if there's a tie in comparisons won, the committee tends to break it with who won the comparison among the two tied teams, not based on RPI, even though the PWR lists the teams in order of RPI. That's not in the women's handbook (or even the men's?) but it's in a men's PWR FAQ somewhere.

And BU clearly wins the comparison with Clarkson based on 2 H2H wins and common opponents. So Clarkson is the team that gets bounced, not BU, if someone outside the top 5 wins an autobid.

Kind of funny half of the bracket could be Beanpot redux with special guest CHA champ replacing Northeastern.

 

[vicb]

Re: 2015 Pairwise Discussion & Predictions

.........
And BU clearly wins the comparison with Clarkson based on 2 H2H wins and common opponents. So Clarkson is the team that gets bounced, not BU, if someone outside the top 5 wins an autobid.
..........

Even I as a huge Clarkson fan would have to agree with this. They beat us head to head and did better against common opponents, BU deserves the spot, not Clarkson.

 

[ARM]

Re: 2015 Pairwise Discussion & Predictions

Kind of funny half of the bracket could be Beanpot redux with special guest CHA champ replacing Northeastern.

Or Clarkson rather than a CHA champ. Another possibility, although not a strong one, is that Northeastern wins Hockey East and you wind up with the Beanpot field again, with the same bracket. A dream scenario for the committee -- there should be money in the budget to pay for those bus trips.

 

[Bumbles]

Or Clarkson rather than a CHA champ. Another possibility, although not a strong one, is that Northeastern wins Hockey East and you wind up with the Beanpot field again, with the same bracket. A dream scenario for the committee -- there should be money in the budget to pay for those bus trips.

If BU loses to Northeastern, would they not miss the NCAAs?

 

[Numbers]

Re: 2015 Pairwise Discussion & Predictions

So the conclusion I am getting here is that, really, BU is currently 6th, and Clarkson 7th. Right?

Thanks for the help.

 

[TonyTheTiger20]

Re: 2015 Pairwise Discussion & Predictions

I think the conclusion is we aren't really sure what the committee would do, but they would have a reasonable justification either way.

Never assume to know what the committee will do hahaha

 

[TonyTheTiger20]

Re: 2015 Pairwise Discussion & Predictions

(I realize the hypocrisy of me saying that right after unequivocally saying that they will not mess with bracket integrity to avoid intraconference matchups but my point still stands lol)

 

[Watson Rink]

Re: 2015 Pairwise Discussion & Predictions

Another possibility, although not a strong one, is that Northeastern wins Hockey East and you wind up with the Beanpot field again, with the same bracket. A dream scenario for the committee -- there should be money in the budget to pay for those bus trips.

I had a nightmare that if the QFs end up with two WCHA teams at #1 and #4 and two Eastern teams at #2 and #3, the committee will issue a ukase requiring the semis to be held at campus sites so that the Frozen Four will require fewer plane tickets!