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.

2019-20 pwr

Re: 2019-20 pwr

At this point, it is foolish to even pay any attention to this metric. Until there is more connectivity in the competion graph, the teams with insular schedule are getting a huge boost..
 
Re: 2019-20 pwr

NUProf said:
At this point, it is foolish to even pay any attention to this metric. Until there is more connectivity in the competion graph, the teams with insular schedule are getting a huge boost..
Click to expand...

Yeah, there are too few cross-over games at the D-3 level for either the PWR (or it's twin, the RPI) to mean much...

Yet, that's far better than the smoky-room BS, and we still have the far superior option of employing the KRACH.

(If only.)

;)
 
Last edited:
Re: 2019-20 pwr

NUProf said:
At this point, it is foolish to even pay any attention to this metric. Until there is more connectivity in the competion graph, the teams with insular schedule are getting a huge boost..
Click to expand...

Well, too be fair, the only teams with insular schedules who are getting a boost are the better teams in weak conferences. The top-to-bottom stronger conferences tend to eat their own young.
 
Last edited:
Re: 2019-20 pwr

Fishman'81 said:
Well, too be fair, the only teams with insular schedules who are getting a boost are the better teams in weak conferences. The top-to-bottom stronger conferences tend to eat their own young.
Click to expand...

Explain to me why you think Ken's metric is so superior. Give me a reason for your faith in this measurement. The lack of schedule connectivity creates problem with that metric as well. When the only path connecting two teams has a bridge, comparisons will not be reliable. Ken's metric is no better or worse than the RPI, only different with similar flaws for similar reasons.
 
Re: 2019-20 pwr

NUProf said:
Explain to me why you think Ken's metric is so superior. Give me a reason for your faith in this measurement. The lack of schedule connectivity creates problem with that metric as well. When the only path connecting two teams has a bridge, comparisons will not be reliable. Ken's metric is no better or worse than the RPI, only different with similar flaws for similar reasons.
Click to expand...

The KRACH is simply a better predictor of results than the PWR/RPI.
 
Re: 2019-20 pwr

Fishman'81 said:
The KRACH is simply a better predictor of results than the PWR/RPI.
Click to expand...

On what basis do you make that claim? That's my question to you. It has flaws that are similar to the RPI and for the same reasons. Comparisons between the two regions are flawed with both metrics because the small number of E/W games gives the results of those games that are played excessively heavy influence in determining the comparisons between the two regions.
 
Re: 2019-20 pwr

NUProf said:
On what basis do you make that claim? That's my question to you. It has flaws that are similar to the RPI and for the same reasons. Comparisons between the two regions are flawed with both metrics because the small number of E/W games gives the results of those games that are played excessively heavy influence in determining the comparisons between the two regions.
Click to expand...

I understand that. There is a serious dearth of comparative D-3 data in many, many cases. None of the available metrics have a lot of grist for the mill at this level.

But the KRACH has just held more water for years now, and that's clear. Certainly it's the best metric of the three, and you can look that up over the past several years.
 
Last edited:
Re: 2019-20 pwr

I am asking for a quantitative explanation as to why you keep saying that "KRACH has just held more water for years now." Is there evidence based data that shows that? I have read claims like that with no supporting documentation at all for years.
 
Last edited:
Re: 2019-20 pwr

NUProf said:
I am asking for a quantitative explanation as to why you keep saying that "KRACH has just held more water for years now." Is there evidence based data that shows that? I have read claims like that with no supporting documentation at all for years.
Click to expand...

Well, you can look up how well the KRACH has predicted outcomes vs. how the PW has, if you like... I have a ball-game to watch. ( And I haven't kept a log, but I'm completely confident that it has.)

And, as an aside, do you actually believe that the PWR is a more thorough metric? Please. Of course you don't. The KRACH employs far more data, and you know that. "Statistical regression", as you you used to say all the time.

Now, you're a PW guy? Come on.
 
Last edited:
Re: 2019-20 pwr

LakerEagleLover said:
The fact its just the two of you bantering here is all anyone needs to know....:D
Click to expand...

True enough.

This D-3 thread is fairly moribund, and I seriously doubt that uscho itself will even continue to exist at all, three years down the road.

It was fun while it lasted, though.

But it has to be said that Prof attempts to prey on the general ignorance of the population, even on this humble forum.
 
Last edited:
Re: 2019-20 pwr

The "way to early" 2020 NCAA field looks like this.
First the Pool A 1st place teams. Their PWR ranking is included in ()

CCC - Univ. of New England (15)
MASCAC - Plymouth State (57)
MIAC - Augsburg (24)
NCHA - Lake Forest (2)
NEHC - Norwich (3)
NESCAC - Trinity (1)
SUNYAC - Oswego (31)
UCHC - Wilkes (7)

The remaining 4 teams are the highest remaining teams in the PWR
1 Trinity 83 10-1-0 0.9091 1 0.6514* 1
2 Lake Forest 82 13-2-1 0.8438 3 0.6241* 2
3 Norwich 81 12-2-1 0.8333 4 0.6224* 3
4 Canton 80 8-2-2 0.7500 11 0.6043* 4
5 UW-Eau Claire 79 11-2-1 0.8214 6 0.5977* 5
6 Utica 78 10-2-2 0.7857 7 0.5954* 6
7 Elmira 76 9-3-1 0.7308 13 0.5934* 7
7 Wilkes 76 11-2-0 0.8462 2 0.5930* 8
9 Babson 75 8-3-2 0.6923 18 0.5906 9
10 Hobart 73 10-3-2 0.7333 12 0.5902* 10
10 Geneseo 73 12-3-1 0.7812 8 0.5866* 11
12 UW-Superior 72 12-2-1 0.8333 4 0.5829* 12

Seeding the tournament
1 Trinity
2 Lake Forest
3 Norwich
4 Canton
5 Eau Claire
6 Utica
7 Elmira
8 Wilkes
9 Univ. of New England
10 Augsburg
11 Oswego
12 Plymouth State

I'll do another one in a week.
 
Re: 2019-20 pwr

28 Middlebury 55 7-6-0 0.5385 35 0.5346 28
28 St. Norbert 55 8-8-2 0.5000 39 0.5344 29
30 UW-Stevens Point 53 9-6-1 0.5938 32 0.5340* 30

How the mighty have fallen
 
Re: 2019-20 pwr

By my quick look at the PWR this week, I come up with a 10-2 split. Stevenson replaces Wilkes as the UCHC rep. The teams would be seeded this way:

W1 - Lake Forest (1)
W2 - Augsburg (27)

E1 - Norwich (1)
E2 - Trinity (3)
E3 - Utica (4)
E4 - Babson (4)
E5 - Geneseo (6)
E6 - Canton (7)
E7 - Stevenson (10)
E8 - University of New England (15)
E9 - Oswego (30)
E10 - Plymouth State (54)


First round:

Plymouth State @ Babson (I switched Plymouth State and Oswego for easy of travel). Strictly by the numbers, Oswego would play here)
Oswego @ Utica
University of New England @ Geneseo (499 miles)
Stevenson @ Canton (459 miles)

Interesting travel for the first round here... 3 New England schools, 4 New York schools, and one in Maryland... would they drastically play with the matchups to try and reduce travel a little? (For instance, you could switch Canton and U. of New England (496 miles). I would love to have seen what they did with this lineup... I doubt the final setup will only have 2 western teams.
 
Last edited:
Re: 2019-20 pwr

PrezdeJohnson09 said:
I think he’s actually got it correct. Oswego and Stevenson technically lead their conferences by the current standings.

That makes the Pool C bids:

Utica, Babson, Geneseo and Canton. All of which are above Eau Claire and Marian currently.
Click to expand...

Thanks Prez...I didn't look that deep..,,
 
Re: 2019-20 pwr

spwood said:
Interesting travel for the first round here... 3 New England schools, 4 New York schools, and one in Maryland... would they drastically play with the matchups to try and reduce travel a little? (For instance, you could switch Canton and U. of New England (496 miles). I would love to have seen what they did with this lineup... I doubt the final setup will only have 2 western teams.
Click to expand...

As long as the match ups are within 500 miles, they usually try for bracket integrity first with the caveat being what type of travel may occur in the second round.
 
Re: 2019-20 pwr

I have been watching these posts for a few years, and they always interest me. However, I am not a math major (actually American History!), so my question: Is there someplace I can go to find an explanation of KRACH, PairWise and the RPI and, explain it to me like I am a 6-year-old?
 
You must log in or register to reply here.
Back
Top