What's new
By:
USCHO Fan Forum

This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

  • The USCHO Fan Forum has migrated to a new plaform, xenForo. Most of the function of the forum should work in familiar ways. Please note that you can switch between light and dark modes by clicking on the gear icon in the upper right of the main menu bar. We are hoping that this new platform will prove to be faster and more reliable. Please feel free to explore its features.

The 2015 Pairwise, Bracketology and History Thread

  • Thread starter Thread starter Priceless
  • Start date Start date
Re: The 2015 Pairwise, Bracketology and History Thread

umhockey91 said:
It would be a shame to see UMass-Lowell miss the tourney. I think that they are built for a deep run if they make it.

I know that PWR involves a lot of number crunching but I look at some of the mediocre records and wonder how a team like UML at 21-12-6 would be out while 19-18-1 St. Cloud would be in.
Click to expand...

That's because the entire scheduling and PWR system in determining at large and autobids as it stands today is seriously flawed. The team and fans I feel the most for is Robert Morris.

EDIT:

I would add from a statistical perspective, these are not outliers. These examples are merely symptomatic of a system that lacks discriminatory, conclusion and internal validity based upon an inaccurate sampling methodology. If one of my students submits statistical results for a research study based upon this system in a undergrad course...the grade is a "D".
 
Last edited:
Re: The 2015 Pairwise, Bracketology and History Thread

Somebody confirm this one. If Harvard and BU win, then Yale is in.
 
Priceless said:
Essentially what we have is this

In:
Mankato, Michigan Tech
North Dakota, Miami, Denver, Duluth, Omaha, St Cloud
Boston University, Boston College
ECAC Champ
AHA Champ

4 open spots

Still alive, playing, can be at-large
Minnesota, Harvard

Still alive, playing, need AQ
RIT, Mercyhurst
Colgate
Michigan
Lowell

Still alive, not playing:
Providence
Quinnipiac
Yale

I think Bowling Green is out. If anyone has a scenario where they make it, please post it.
Click to expand...

BG is in with wins by minnesota, Harvard, BU, Miami, Denver and tech
 
Priceless said:
If all higher seeds win tomorrow:

Fargo
#2 North Dakota vs RIT
Omaha vs Minnesota

South Bend
#3 Miami vs Harvard
Michigan Tech vs St Cloud

Providence
#1 Mankato vs Providence
Denver vs Colgate

Manchester
#4 Boston U vs Quinnipiac
Duluth vs Boston C
Click to expand...

This is a joke, right? If we finish with the #1 overall seed, we not only don't play the #16 seed, we get sent to Providence to play Providence?
 
Re: The 2015 Pairwise, Bracketology and History Thread

HarleyMC said:
Nope, not the core problem. Although I appreciate the speculations.:)
Click to expand...

I'm not saying the NCHC isn't the best league in the country but I struggle to believe they're that much better than the rest. I'll show my analysis once the regular season is over.
 
Re: The 2015 Pairwise, Bracketology and History Thread

Looks like Q is in. BG is still alive.
 
Re: The 2015 Pairwise, Bracketology and History Thread

HarleyMC said:
That's because the entire scheduling and PWR system in determining at large and autobids as it stands today is seriously flawed. The team and fans I feel the most for is Robert Morris.

EDIT:

I would add from a statistical perspective, these are not outliers. These examples are merely symptomatic of a system that lacks discriminatory, conclusion and internal validity based upon an inaccurate sampling methodology. If one of my students submits statistical results for a research study based upon this system in a undergrad course...the grade is a "D".
Click to expand...

I am guessing that you may have already answered the following question, sometime, somewhere, here in the forum, but I haven't looked.

What would be more fair than the system we have in place, now? I say KRACH, but it is sort of a Pairwise variant and I detect that you have something else entirely in mind.

I certainly find what college hockey does to be far less subjective than what goes on in D-1 basketball or football. Therefore, I find it very acceptable by comparison, particularly when one considers that it involves the NCAA, which I personally feel borders on being a totally profit motive driven cartel.
 
Re: The 2015 Pairwise, Bracketology and History Thread

MSchvatz44 said:
I have ran through numerous scenarios and cannot find one in which Quinnipiac does not make the tournament. Have you been seeing similar results?
Click to expand...

According to PWR, QU is in. Changes from yesterday seem to be:

BC is a lock
SCSU is a lock
QU is a lock
Colgate needs to win its conference tournament.
Mass.-Lowell needs to win its conference tournament.
Yale is up to about a 23% chance.
Vermont is out

http://bit.ly/1GBEDZI
 
Re: The 2015 Pairwise, Bracketology and History Thread

purpleinnebraska said:
This is a joke, right? If we finish with the #1 overall seed, we not only don't play the #16 seed, we get sent to Providence to play Providence?
Click to expand...

Sorry, but attendance matters. Technically, RIT could stay in Providence (386 miles) but the Providence regional would rival Joe Louis Arena last night for empty seats.
 
Re: The 2015 Pairwise, Bracketology and History Thread

Priceless said:
Sorry, but attendance matters. Technically, RIT could stay in Providence (386 miles) but the Providence regional would rival Joe Louis Arena last night for empty seats.
Click to expand...

I cannot see a way that MSU is not in South Bend as the #1 overall seed.
 
Priceless said:
Sorry, but attendance matters. Technically, RIT could stay in Providence (386 miles) but the Providence regional would rival Joe Louis Arena last night for empty seats.
Click to expand...
If attendance REALLY mattered, they would change the bidding process to get a regional.
 
Re: The 2015 Pairwise, Bracketology and History Thread

HarleyMC said:
That's because the entire scheduling and PWR system in determining at large and autobids as it stands today is seriously flawed. The team and fans I feel the most for is Robert Morris.

EDIT:

I would add from a statistical perspective, these are not outliers. These examples are merely symptomatic of a system that lacks discriminatory, conclusion and internal validity based upon an inaccurate sampling methodology. If one of my students submits statistical results for a research study based upon this system in a undergrad course...the grade is a "D".
Click to expand...

I don't think you're going to get scheduling down a sampling design in this world.

---

We (err, JimDahl) has the means to do a more proper investigation. Maybe its time for a sensitivity analysis of sorts. I'd take the schedule up to the NCAAs as given and then knock out and re-impute a fraction of the observations randomly and see how sensitive it is. How much does it change under a model that passes a smell test.

I've given up on calculating PWR for the time being and I don't have the drive to do it.

That being said, the "power" in any sports ranking method isn't particularly high and assessing a ranked score is not the most elegant task. A sample size of 40 would make any pollster unhappy.

---

If an undergrad studied the PWR under its various conditions in a rigorous manner, its probably publishable.
 
Last edited:
Re: The 2015 Pairwise, Bracketology and History Thread

Essentially what we have is this

In:
Mankato, Michigan Tech
North Dakota, Miami, Denver, Duluth, Omaha, St Cloud
Boston University, Boston College
Quinnipiac
ECAC Champ
AHA Champ

3 open spots

Still alive, playing, can be at-large
Minnesota, Harvard

Still alive, playing, need AQ
RIT, Mercyhurst
Colgate
Michigan
Lowell

Still alive, not playing:
Providence
Yale
Bowling Green
 
Priceless said:
If all higher seeds win tomorrow:

Fargo
#2 North Dakota vs RIT
Omaha vs Minnesota

South Bend
#3 Miami vs Harvard
Michigan Tech vs St Cloud

Providence
#1 Mankato vs Providence
Denver vs Colgate

Manchester
#4 Boston U vs Quinnipiac
Duluth vs Boston C
Click to expand...
I don't think the committee is going to risk an "all NCHC frozen four" like the WCHA had in 2005. Wouldn't surprise me at all if they slam three NCHC teams into one regional and leave one without an NCHC team. Something to think about.
 
Re: The 2015 Pairwise, Bracketology and History Thread

Shirtless Guy said:
I'm not saying the NCHC isn't the best league in the country but I struggle to believe they're that much better than the rest. I'll show my analysis once the regular season is over.
Click to expand...

Shirtless, the problems in this probability model are causal and systemic and has nothing to do with specific conferences. I'm absolutely serious here, if I were a coach of any team with 20+ wins and sitting tonight at PWR 16-25, I'd be tempted to kick the NCAA committee in the *** for what this statistically invalid system has done to rob my team out of an NCAA berth. It's THAT bad.
 
Priceless said:
Essentially what we have is this

In:
Mankato, Michigan Tech
North Dakota, Miami, Denver, Duluth, Omaha, St Cloud
Boston University, Boston College
Quinnipiac
ECAC Champ
AHA Champ

3 open spots

Still alive, playing, can be at-large
Minnesota, Harvard

Still alive, playing, need AQ
RIT, Mercyhurst
Colgate
Michigan
Lowell

Still alive, not playing:
Providence
Yale
Bowling Green
Click to expand...

What knocks minnesota out?
 
You must log in or register to reply here.
Back
Top