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.

2010-2011 Rutter DI Computer Rankings Thread

Re: 2010-2011 Rutter DI Computer Rankings Thread

ARM said:
Usually, we see these two tighten up as the season progresses. As some WCHA teams are running out of non-conference games, I'm not sure if they will line up or not. The key differences at this point are Minnesota, BSU, BU, and BC, because those are the rankings where one has a team in an 8-team field, the other has them on the outside. With 20 games to go, it is likely that there will be at most a couple of teams that would receive a different in/out fate with a different NCAA ranking system.
Click to expand...

Having followed these rankings over the years, it's pretty clear to me that RPI tends to suggests more parity between conferences than other statistical methods would suggest. To the extent that ARM observes they tend converge, it's only true because there has been more conference parity in the past recent years than this year.

For example, I can recall in the middle of last decade, some years where the men's WCHA was strong enough in the KRACH to get almost of its teams in the top 20, above the top ECAC team (I may be slightly exaggerating). But you wouldn't have known it from the RPI.

A good example from Women's hockey is that 2005 Wisconsin team that was 3rd in all the polls and KRACH but 5th in the Pairwise (and I believe also RPI).

That said, as Lakersfan suggests, we don't always have enough data points to make a strong conclusion about the relative strength of conferences. The RPI then tends to give teams with good records from weaker conferences more of a benefit of the doubt. I suspect this is one reason why you'll never see a superior statistical system adopted.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

If you want a ranking that takes into account margin of victory with solid statistical foundations, go here: http://it.stlawu.edu/~chodr/wchodr/current.html

Averaging margin of victory is kind of silly because it treats a goal scored against St Cloud the same way as a goal scored against Kim Martin.
 
Last edited:
Re: 2010-2011 Rutter DI Computer Rankings Thread

MICZamboni said:
Dr. Rutter...have you seen this thread on the men's side?

http://board.uscho.com/showthread.php?92761-MyHockeyRankings.com-Division-1-Computer-Rankings
Click to expand...

That thread being a wonderful example of why I stick to ranking women's college hockey. I have enough stress about trying to get my research published in a peer-reviewed journal. Speaking of which, look for an article about ranking women's college hockey teams in the spring issue of Chance magazine, or so I have been told.

And I second Dave's recomindation of Robin Locke's CHODR rankings. A nice counter to the win-loss only ranking systems.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

LakersFan said:
And I second Dave's recommendation of Robin Locke's CHODR rankings. A nice counter to the win-loss only ranking systems.
Click to expand...
As of today, the Rutter and CHODR rankings have the same top 10. In different order for the most part, but interesting that to some extent they agree on which teams belong in the discussion.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

For games played through 11/28/2010

Code: 
  	Team 			Rating 	
1 	Wisconsin 		1.5872 
2 	Cornell 		1.4013 	
3 	Minnesota Duluth 	1.2999 	
4 	Mercyhurst 		1.1730 	
5 	Minnesota 		0.9827 
6 	North Dakota 		0.7934 	
7 	Boston University 	0.7692 
8 	Bemidji State 		0.6116 	
9 	Boston College 		0.5324 	
10 	Ohio State 		0.5286
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

For games played through December 5th, 2010

Code: 
  	Team 			Rating 	
1 	Wisconsin 		1.5889 
2 	Cornell 		1.5552 	
3 	Mercyhurst 		1.2177 	
4 	Minnesota Duluth 	0.9945 
5 	Minnesota 		0.8221 
6 	Boston University 	0.7921 
7 	North Dakota 		0.6172 
8 	Boston College 		0.5998 	
9 	Bemidji State 		0.5601 	
10 	Ohio State 		0.4695
11 	Providence 		0.4564

If you look at the graphical summary of the ratings (http://math.bd.psu.edu/faculty/rutter/D1_history/season_history.html), you can see what the effect of the UMD/SCS tie was. Given the distance between UMD and SCS in terms of rating, the tie was as good as a win for SCS rating wise. All the top WCHA teams took a little tumble after that result, as results against UMD suddenly became a little less impressive.

At this point, I would consider Wisconsin and Cornell virtually tied. On the other of the top 10, Ohio State and Providence are close as well.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

LakersFan said:
If you look at the graphical summary of the ratings (http://math.bd.psu.edu/faculty/rutter/D1_history/season_history.html), you can see what the effect of the UMD/SCS tie was. Given the distance between UMD and SCS in terms of rating, the tie was as good as a win for SCS rating wise. All the top WCHA teams took a little tumble after that result, as results against UMD suddenly became a little less impressive.
Click to expand...
Wouldn't the negative effect of UMD being devalued be partially offset by an increase in the value of the H2H results with SCSU? I'd expect that the more damaging results for WCHA teams were Minnesota State's series versus Wayne State and Robert Morris, where the Mavericks went 0-3-1.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

Not exactly putting my finger on it but week's uscho poll seems to say something as to pollsters whistling up tel poles/masts, somewhat (maybe.) Two losses home/away with Harvard seemingly puts Dartmouth out of contention vis a vis Quinnipiac, Bemidji State, etc. I don't think so. Is a hodge podge though.
 
Last edited:
papulaisle said:
Not exactly putting my finger on it but week's uscho poll seems to say something as to pollsters whistling up tel poles/masts, somewhat (maybe.) Two losses home/away with Harvard seemingly puts Dartmouth out of contention vis a vis Quinnipiac, Bemidji State, etc. I don't think so. Is a hodge podge though.
Click to expand...
I agree the PC rank is puzzling.
But the harvard games are a large share of Dartmouth's season to date.
Also Bemidji has a lot more top 10 wins than any unranked team. if anyone has a case, it's them.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

dave1381 said:
I agree the PC rank is puzzling.
Click to expand...

I am not sure I understand this comment unless you mean you think PC should be rated higher. ;) With an almost 74% win/tie percentage, it seems appropriate.

I admit that Providence has lost a few that they shouldn't have (RMU, RPI), but they have also won all the games that they should have. As a matter of fact, comparing Dartmouth to PC's schedule, we have both beat the same teams and if you count McGill, we beat them twice when Dartmouth lost to them once.

And statisically, if you look at the teams that Dartmouth lost to we have a good record against the opponents they lost to. Boston College lost and tied St. Lawrence, we beat them. Harvard tied Yale and McGill and we beat them both.

All I am saying is that #11 doesn't seem that far fetched.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

papulaisle said:
Not exactly putting my finger on it but week's uscho poll seems to say something as to pollsters whistling up tel poles/masts, somewhat (maybe.)
Click to expand...
dave1381 said:
I agree the PC rank is puzzling.
Click to expand...
scrambledlegs said:
I am not sure I understand this comment unless you mean you think PC should be rated higher. ...
All I am saying is that #11 doesn't seem that far fetched.
Click to expand...
The topic seems to the USCHO poll, not the Rutter ranking (which is confusing given the comment was made in the "Computer Ranking Thread", but papulaisle is always good for some confusion.;)) The voters placed PC 8th, above teams like UND. And yes, PC has a better record than NoDak, but with at least as many bad losses without the wins over ranked teams. In that case, your only common opponent is BU, and the Sioux did better by splitting. Maybe voters feel that UND's best results came early and they have sinced tailed off a bit.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

ARM said:
The topic seems to the USCHO poll, not the Rutter ranking (which is confusing given the comment was made in the "Computer Ranking Thread",
Click to expand...
We should have a kangaroo court, and you should be the "ARM" of the Law to enforce it! ;)

Lakersfan, any chance that dynamic graphic tool is updated and available for the current season? Or are we sparing ARM and others from having to see the Gophers downward slope? (I'd make googly eyes here if only I knew how..... :( )
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

IceIsNice said:
Lakersfan, any chance that dynamic graphic tool is updated and available for the current season? Or are we sparing ARM and others from having to see the Gophers downward slope? (I'd make googly eyes here if only I knew how..... :( )
Click to expand...

I believe that the data is from 2010-2011 season, it is just the verbage at the top is wrong. For instance, last year, PC hovered around 0 :(
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

scrambledlegs said:
I admit that Providence has lost a few that they shouldn't have (RMU, RPI), but they have also won all the games that they should have.
Click to expand...
:confused: :confused: :confused:

I got a kick out of reading that sentence.
 
Last edited:
  • Like
Reactions: D2D
Re: 2010-2011 Rutter DI Computer Rankings Thread

For games played through December 12, 2010

Code: 
  	Team 			Rating 
1 	Wisconsin 		1.5715 	
2 	Cornell 		1.5229 	
3 	Mercyhurst 		0.9955 
4 	Minnesota Duluth 	0.9270 	
5 	Minnesota 		0.9221 
6 	Boston University 	0.9199 	
7 	Boston College 		0.6297 	
8 	North Dakota 		0.6290 	
9 	Bemidji State 		0.4869 	
10 	Providence 		0.4572

And I finally got around to changing the data on my graphical display of the season's ratings. Almost everything in the process is now automated, except changing dates. I need to work on that.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

For games played through January 2, 2011

Code: 
	Team 			Rating 	
1 	Wisconsin 		1.7077 	
2 	Cornell 		1.5365 	
3 	Minnesota Duluth 	0.9649 	
4 	Minnesota 		0.9625 	
5 	Mercyhurst 		0.9243 	
6 	Boston University 	0.9217 	
7 	North Dakota 		0.6725 	
8 	Bemidji State 		0.5074 	
9 	Providence 		0.5002 	
10 	Boston College 		0.4996

The uncertainty estimates are even more interesting, IMO.

Code: 
  	Team 		25th	75th Percent Top 8
1 	Wisconsin 	1 	2 	99.9
2 	Cornell 	1 	3 	98.7
3 	UMD 		3 	7 	88.1
4 	Minnesota 	3 	7 	88.0
5 	BU 		4 	7 	86.1
6 	Mercyhurst 	3 	7 	82.9
7 	North Dakota 	6 	10 	59.4
8 	Bemidji State 	7 	12 	37.3
9 	Providence 	7 	12 	36.8
10 	Boston College 	7 	12 	36.4
11 	Ohio State 	8 	14 	26.1
12 	Dartmouth 	9 	16 	19.2
13 	Quinnipiac 	10 	16 	15.1
14 	Northeastern 	10 	16 	11.7

You have the two top teams teams, the four teams fighting for home ice in the quarterfinals (UMD, UM, BU, and Hurst), North Dakota, three teams fighting for the last slot (BSU, PC, BC, maybe OSU), and then the teams outside looking in.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

LakersFan said:
You have the two top teams teams, the four teams fighting for home ice in the quarterfinals (UMD, UM, BU, and Hurst), North Dakota, three teams fighting for the last slot (BSU, PC, BC, maybe OSU), and then the teams outside looking in.
Click to expand...

In the PWR/RPI BU is in the 3rd place with quite a bit of breathing room over a pack of teams that includes UMD/Mercyhurst/Minnesota/BC battling to host the 4th quarterfinal.

Not quite sure I see the logic of describing the playoff race in terms of your rankings, when the reality is different. And I don't think it's likely that the PWR will converge to your ranking. The biggest difference in the rankings is the treatment of BU and BC, and I think the PWR/RPI will continue to inflate their ranking throughout the season.
 
Re: 2010-2011 Rutter DI Computer Rankings Thread

dave1381 said:
In the PWR/RPI BU is in the 3rd place with quite a bit of breathing room over a pack of teams that includes UMD/Mercyhurst/Minnesota/BC battling to host the 4th quarterfinal.

Not quite sure I see the logic of describing the playoff race in terms of your rankings, when the reality is different. And I don't think it's likely that the PWR will converge to your ranking. The biggest difference in the rankings is the treatment of BU and BC, and I think the PWR/RPI will continue to inflate their ranking throughout the season.
Click to expand...

Yes, the PWR will likely not converge with my (any other) rankings. What my rankings suggest is that UMD, UM, BU, and Hurst pretty close to each other in terms of quality and that the other two top seeds will likely come from this bunch (with BU and UMD currently having the advantage as of today). The PWR currently likes what BU has done so far, but it is very difficult to look at the PWR and determine how secure or insecure that position is. From a "where does my team stand" perspective, I think my uncertanity table gives some additional information.

By the way, looking closely at USCHO.com's PWR, it appears not to be awarding a point for record vs. TUC. When you click on an individual comparission, both are 0 when one is clearly higher (BU vs UM for example).
 
You must log in or register to reply here.
Back
Top