2013-2014 Women's D-I PairWise Contentions and Affirmations

[OnMAA]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Well done, fellows. Thank you. Unlikely, though, eh?

There is not that much of a difference amongst the top four in the ECAC. Seen Quinnipiac play a few times in the past, and IMHO they have as good a shot as any of the other three teams. I think they match up well against both Harvard and Cornell, so if they somehow can get past Clarkson, I see them having a good shot at it.

 

[Numbers]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Yeah all this theorizing out of the window after a few upsets. What if RIT beats MU. What if UVM wins HE. What if NoDak wins the WCHA. What if Quinnipiac wins the ECAC. If those things ALL were to happen you might as well put all your pre-arranged brackets in the garbage can and start with a blank sheet of paper. (partly kidding off-course)

Well, now that is interesting. Let's see. If RIT beats MU, I don't think much changes. If Quinn wins HEA, then they are in, perhaps as #7, maybe as #8. If NoDak wins WCHA, then they are in as #8, and either Quinn or Mercy is #7. If UVM wins HEA, too, them someone else is out, and who that is depends on results.

If Minny wins, Quinn wins, and BC wins, then it's 1-UM, 2-UW, 3, 4, 5, 6 = Clark, Corn, Harv , BC in some order, 7&8-Mercy and Quinn. That's an easy, but boring bracket.

Add NoDak winning and you get 1-UM, 2-UW, 3,4,5,6 as above, 7-Quinn, 8-NoDak. Also an easy and boring bracket. NoDak's reward for beating Minny is to play them again.

Add upset in HEA and you get 1-UM, 2-UW, 3,4,5 (3 of Clark, Corn, Harv, BC. Who gets left out?); 6-Quinn, 7-NoDak, 8-HEA winner. And, the bracket goes NoDak @ Minn (bus ride); HEA @ UW; and the 4 remaining eastern schools.

 

[brookyone]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

I probably sleep better than you guys. I just wait and see what happens.

 

[OnMAA]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

If Quinn wins HEA, then they are in

The only problem with that option is that it is impossible for Quinn to win HEA.
(Remember that Quinnipiac joined the ECAC to take the spot vacated by UVM when UVM jumped from the ECAC to HE)

 

[robertearle]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Because a 15-4-2 CoP record is 'better' than an 18-2-2 record. !Ay Caramba!

'Old' vs 'new' COP calculation also flipping the Quinnipiac vs Robert Morris comparison; in one case RMU benefiting from the change, in the other being hurt by it.

 

[driventoit]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

I probably sleep better than you guys. I just wait and see what happens.

A true sage. (Or at least one who can be appreciated from the cheap seats.)

 

[dave1381]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

If the last spot comes down to RMU vs. Quinnipiac, and Quinnipiac loses to Clarkson, I expect it would be RMU.

What works for RMU are the H2H win over Quinnipiac, the smallish Quinnipiac RPI advantage, and RMU doing relatively well in the comparison vs. Mercyhurst compared to Quinnipiac.

The burden is on Quinnipiac to really distinguish itself in the other criteria to overcome that H2H loss to RMU. That means beating Clarkson, which give Quinnipiac enough of an RPI edge to even be considered over Mercyhurst.

 

[ARM]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

RIT will become a TUC if it wins the CHA final on Saturday. Adding RIT to the TUC list could scramble some numbers, but shouldn't impact the RMU/MU compare, because in that scenario, each would be 3-2 versus the Tigers.

 

[Eeyore]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

If the last spot comes down to RMU vs. Quinnipiac, and Quinnipiac loses to Clarkson, I expect it would be RMU.

Maybe, but it would be a shame if it did. If that happens, Quinnipiac would have a .0049 edge in RPI which isn't large but it is significant, and a very small edge in record vs. Common Opponents. RMU's only claim to win the pair would be an edge in record vs. TUC of either .6154 or .545 (depending upon whether Vermont beats BC and remains a TUC or loses and falls off the cliff) against Quinnipiac's .500. That looks like a large edge but it is the product of pretty much the nightmare scenario of why the Record vs. TUC criterion is so badly flawed.

Based upon the numbers right now, RMU has played 13 games against TUCs. By game, those 13 games were against the following rankings by RPI (with Sacred Heart removed): 7,7,7,7, 11, 11, 14, 14, 14, 14, 16, 18, 18. That's an average ranking of 12.15 and a mean RPI of .5312.

Quinnipiac would have played 10 games vs. TUCs, with rankings of: 2, 2, 2, 4, 4, 5, 6, 6, 15, 15. That's an average ranking of 6.10 and a mean RPI of .5793.

You simply can't draw a meaningful comparison between these two teams with their raw winning percentage vs. TUCs because they haven't played anything resembling the same schedule vs. TUCs. The committee would be saying that Quinnipiac not only has to have a better RPI and record vs. Common Opponents than Robert Morris does but that they also have to have a better record vs. TUCs despite playing a much more difficult selection of them.

That's idiotic. No matter what happens tomorrow, no one should think that RMU wins the comparison between these two teams. If they do it means that they don't understand what the numbers are actually saying.

Edit: Keep in mind that the games between Quinnipiac and RMU are dropped when calculating their records vs. TUCs for the comparison between them. It's one of the sensible things involved in this process, since those games are already accounted for in the HtH points criterion.

 

[Numbers]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Can anyone find a way to get Clarkson or Cornell out?

I looked this way - assume upsets in all 3 tourneys. Then, 5th is the cutline. I can't find a way to get either of those schools down to 6th with Conference upsets by any reasonable consideration of the criteria.

Comments?

PS - Eeyore, I appreciate your analysis of RMU v Quinn above.

 

[dave1381]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Eeyore, I don't understand how you're factoring in RMU's head-to-head win over Quinnipiac.

I agree the TUC criteria is messed up so it'll probably get less weight. The question is more whether Quinnipiac's common opponent edge and RPI edge are large enough to overcome the head-to-head. I don't think Quinnipiac has enough edge if it loses to Clarkson.

I figure you need something close to a .0100 edge in RPI (which is exactly a 1 game-difference in standings for teams playing a 30-game season... more like .0080 for a longer season) to be considered equal to overcome a 1 game head-to-head deficit.

I think Quinnipiac would lose out for the same reasons that Dartmouth beat Clarkson in 2008 for the last slot. I do pity Rick Seeley who also coached that Clarkson team.

 

[Eeyore]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Eeyore, I don't understand how you're factoring in RMU's head-to-head win over Quinnipiac.

I wasn't because I must have blinked at the wrong moment and thought that the two teams split a pair rather than the other game being a tie.

I agree the TUC criteria is messed up so it'll probably get less weight. The question is more whether Quinnipiac's common opponent edge and RPI edge are large enough to overcome the head-to-head. I don't think Quinnipiac has enough edge if it loses to Clarkson.

Which I still think would be a shame. Here, too, the way the criterion works overstates Robert Morris' advantage since the two teams played a second game that ended in a tie. RMU does not have a 1.000 winning percentage in head-to-head matchups but that's the way the criterion reads.

Another way of saying this is that I think the way that PWR uses head-to-head matchups is poor. It puts an overwhelming emphasis on the smallest subset of the games. In this case, everything else about the teams' records suggests that Quinnipiac is the stronger team but a single game could override that.

 

[dave1381]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Regarding my overall feelings, in KRACH, Quinnipiac and RMU are 9th and 10th. The rankings imply that we would expect Quinnipiac to beat RMU 52% of the time. If Quinnipiac loses to Clarkson, that probably drops a percentage point or two. When the teams are so close in the various imperfect measures, I have absolutely no problem with giving disproportionate weight to the head-to-head results.

 

[ARM]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Here, too, the way the criterion works overstates Robert Morris' advantage since the two teams played a second game that ended in a tie. RMU does not have a 1.000 winning percentage in head-to-head matchups but that's the way the criterion reads.

You're looking at this totally wrong; H2H isn't a percentage, it's a +/-. RMU is +1 over Q. UM is +4 over UW, and +2 over UND. No percentage. And qualitatively, the teams didn't tie the second game, they tied the first. Then RMU came out on Quinnipiac's ice and won convincingly. So the H2H result isn't as inconsequential as you're trying to make it. I think that RPI may hold sway and Q will get the nod. But if that is not the case, each team will have many results where it could have done more. The ECAC is strong this year, but the Bobcats were only the fourth-best team in the league if they lose today, and I don't know that the ECAC was necessarily four-teams strong. They tied St. Cloud and lost to RPI. Nine ties! As my wife would say, "They should try harder to win things." I'm okay if Q gets in, and I'm okay if RMU does. I don't see any injustice either way. That's why we have a committee instead of a formula.

 

[ARM]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Because a 15-4-2 CoP record is 'better' than an 18-2-2 record. !Ay Caramba!

That hinges heavily on the Ohio State games. Mercyhurst went 0-1 and gets .0000, while RMU was 1-0 and gets a 1.0000.

 

[robertearle]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

That hinges heavily on the Ohio State games. Mercyhurst went 0-1 and gets .0000, while RMU was 1-0 and gets a 1.0000.

Mercyhurst and RMU had seven common opponents. They had the exact same record against two of them (St Lawrence and Penn State); Mercyhurst had a better record - no matter how you look at things - against three of them (Syracuse, RIT, Lindenwood); Mercyhurst had what is in my mind a better record against Maine - 2-0 vs 1-0; the ONLY team that Robert Morris did better against is Ohio State.

As I said when you got confirmation that this was/is how the committee was doing the COP comparisons, you can cook up 'bad' examples for either way. This to me is a 'bad' example of the new mechanism. And who gets the #8 slot might well hinge on it.

 

[robertearle]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Mercyhurst and RMU had seven common opponents. They had the exact same record against two of them (St Lawrence and Penn State); Mercyhurst had a better record - no matter how you look at things - against three of them (Syracuse, RIT, Lindenwood); Mercyhurst had what is in my mind a better record against Maine - 2-0 vs 1-0; the ONLY team that Robert Morris did better against is Ohio State.

As I said when you got confirmation that this was/is how the committee was doing the COP comparisons, you can cook up 'bad' examples for either way. This to me is a 'bad' example of the new mechanism. And who gets the #8 slot might well hinge on it.

...and Robert Morris will continue to have the 'better' COP, even if Mercyhurst beats RIT today. Because the Mercyhurst win over RIT would take their record from 3-1 to 4-1, adding only 0.05 to their total.

In fact, Mercyhurst would have to beat RIT *NINE TIMES* in a row for the effect on their COP to overcome the single loss to Ohio State.

Seems 'wrong' to me...

 

[Skate79]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Three WCHA teams on one side of the bracket and three ECAC on the other? No thanks. I'd rather see a little more crossover in a perfect world.

Where's the third ECAC team? BU plays in Hockey East.

 

[ARM]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Where's the third ECAC team? BU plays in Hockey East.

That was from three months ago so it is hardly relevant, but Harvard, Clarkson and Cornell are all ECAC teams, and they were with BU in that example we were discussing in December.

 

[ARM]

Re: 2013-2014 Women's D-I PairWise Contentions and Affirmations

Mercyhurst and RMU had seven common opponents. They had the exact same record against two of them (St Lawrence and Penn State); Mercyhurst had a better record - no matter how you look at things - against three of them (Syracuse, RIT, Lindenwood); Mercyhurst had what is in my mind a better record against Maine - 2-0 vs 1-0; the ONLY team that Robert Morris did better against is Ohio State.

As I said when you got confirmation that this was/is how the committee was doing the COP comparisons, you can cook up 'bad' examples for either way. This to me is a 'bad' example of the new mechanism. And who gets the #8 slot might well hinge on it.

Remember that the committee reserves the right to weight the criteria differently. So they could agree with you and say that one OT loss to Ohio State should not cost the comparison to a Mercyhurst team that did better than RMU in the CHA season and tournament. Humans can do that, but a USCHO program that just goes by raw math is not capable of making those subjective decisions. That's why it says it "attempts to mimic" what the committee will do, not "here's what the committee will decide."