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.
2009-2010 DI Rutter Computer Rankings Thread
- Thread starter LakersFan
- Start date
Re: 2009-2010 DI Rutter Computer Rankings Thread
For games played through January 24, 2010
For games played through January 24, 2010
Code:
Team Rating RPI.Rank RPI
1 Mercyhurst 1.4189 1 0.6302
2 Minnesota 1.1466 2 0.6068
3 UMD 0.7971 5 0.5660
4 Clarkson 0.7937 3 0.5772
5 New Hampshire 0.7628 4 0.5769
6 Harvard 0.5306 7 0.5459
7 Northeastern 0.5184 6 0.5605
8 Wisconsin 0.4870 10 0.5301
9 Connecticut 0.3648 8 0.5380
10 Providence 0.2939 9 0.5344Re: 2009-2010 DI Rutter Computer Rankings Thread
For games played through January 31, 2010
For games played through January 31, 2010
Code:
Team Rating RPI.Rank RPI
1 Mercyhurst 1.2958 1 0.6227
2 Minnesota 1.1652 2 0.6068
3 Clarkson 0.8577 3 0.5824
4 UMD 0.7932 5 0.5670
5 New Hampshire 0.7857 4 0.5783
6 Harvard 0.6337 7 0.5571
7 Northeastern 0.4873 6 0.5574
8 Wisconsin 0.4399 9 0.5257
9 Connecticut 0.4313 8 0.5455
10 Ohio State 0.2461 13 0.5164Re: 2009-2010 DI Rutter Computer Rankings Thread
I'd be curious to see what the rankings would look like if you treated the home and road versions of teams as separate entities. (as opposed to estimating a home-ice advantage across teams, which you say tends not to be significant).
I imagine "Minnesota home" would be far-and-away the best team. Among road teams, probably "UMD road" would do pretty well, as would "Mercyhurst road." My guess is "Minnesota road" would rank somewhere down with the middle-of-the-pack of the WCHA home teams, and nowhere close to the No. 1.
The good news for Minnesota though is that "Minnesota home" is the team that'll be playing the rest of the season going forward.
I'd be curious to see what the rankings would look like if you treated the home and road versions of teams as separate entities. (as opposed to estimating a home-ice advantage across teams, which you say tends not to be significant).
I imagine "Minnesota home" would be far-and-away the best team. Among road teams, probably "UMD road" would do pretty well, as would "Mercyhurst road." My guess is "Minnesota road" would rank somewhere down with the middle-of-the-pack of the WCHA home teams, and nowhere close to the No. 1.
The good news for Minnesota though is that "Minnesota home" is the team that'll be playing the rest of the season going forward.
Re: 2009-2010 DI Rutter Computer Rankings Thread
For games played through Feb 7th, 2010
For games played through Feb 7th, 2010
Code:
Team Rating RPI.Rank RPI
1 Mercyhurst 1.3239 1 0.6240
2 Minnesota 1.0389 2 0.5913
3 UMD 0.9102 3 0.5778
4 Clarkson 0.6955 4 0.5655
5 Harvard 0.6778 5 0.5640
6 New Hampshire 0.6375 6 0.5625
7 Wisconsin 0.4288 10 0.5252
8 Connecticut 0.4182 8 0.5443
9 Northeastern 0.3873 7 0.5479
10 Cornell 0.2686 9 0.5279scrambledlegs
Registered User
Re: 2009-2010 DI Rutter Computer Rankings Thread
Could you also publish 11-20? Thanks.
Could you also publish 11-20? Thanks.
ARM
Fan of chipmunk-like mascots.
Re: 2009-2010 DI Rutter Computer Rankings Thread
http://math.bd.psu.edu/faculty/rutter/WomensRankings.html
Click the link.
http://math.bd.psu.edu/faculty/rutter/WomensRankings.html
Re: 2009-2010 DI Rutter Computer Rankings Thread
For games played through Feb 21, 2010
For games played through Feb 21, 2010
Code:
Team Rating RPI.Rank RPI
1 Mercyhurst 1.3483 1 0.6240
2 UMD 0.9116 2 0.5778
3 Minnesota 0.8844 3 0.5753
4 Harvard 0.6764 4 0.5629
5 New Hampshire 0.6039 5 0.5594
6 Clarkson 0.5392 6 0.5511
7 Wisconsin 0.4213 11 0.5251
8 Connecticut 0.3997 7 0.5409
9 Cornell 0.3612 9 0.5380
10 Northeastern 0.3401 8 0.5381
11 St. Cloud State 0.2828 15 0.5129
12 Quinnipiac 0.2479 10 0.5339Re: 2009-2010 DI Rutter Computer Rankings Thread
For games played through Feb 28th, 2010
For games played through Feb 28th, 2010
Code:
Rank Team Rating RPI Rk. RPI
1 Mercyhurst 1.3885 1 0.6259
2 UMD 0.9384 2 0.5794
3 Minnesota 0.9163 3 0.5783
4 Harvard 0.7329 4 0.5680
5 New Hampshire 0.6021 5 0.5590
6 Clarkson 0.5551 6 0.5524
7 Connecticut 0.4309 8 0.5440
8 Cornell 0.4154 7 0.5445
9 Wisconsin 0.3527 14 0.5147
10 Northeastern 0.3068 9 0.5326Re: 2009-2010 DI Rutter Computer Rankings Thread
Would be interesting to see the 2009-2010 DI Rutter Rankings by league/conference...so there would only be 4 entries in the rankings. All games would count, except for those vs. independents.
Have you done this in prior years? If so, could you repost those results for a perspective?
My inclination is that the gap between leagues has narrowed considerably.
Would be interesting to see the 2009-2010 DI Rutter Rankings by league/conference...so there would only be 4 entries in the rankings. All games would count, except for those vs. independents.
Have you done this in prior years? If so, could you repost those results for a perspective?
My inclination is that the gap between leagues has narrowed considerably.
Last edited:
Re: 2009-2010 DI Rutter Computer Rankings Thread
Is that really the measure of conference strength that you want? It seems very silly to impose the assumption that Cornell=Union and then try to estimate a parameter.
What makes more sense to me is just calculating the round-robin winning percentages (the expected win pct of each if they played everyone once), and then just averaging those. I think that would give you a decent idea.
But at least you can aggregate these win percentages as you see fit using this method. (for instance, it's unclear to me how much you want to say the ECAC is relatively weaker than another conference because of a school like Union).
But yes, I agree with your hypothesis. Conferences have surely converged a bit this year. You had Vermont beat Clarkson and Brown beat Providence and RMU beat UMD (last-or-near-last teams beating 1st-or-near first teams). that pretty much sums up the year.
Is that really the measure of conference strength that you want? It seems very silly to impose the assumption that Cornell=Union and then try to estimate a parameter.
What makes more sense to me is just calculating the round-robin winning percentages (the expected win pct of each if they played everyone once), and then just averaging those. I think that would give you a decent idea.
But at least you can aggregate these win percentages as you see fit using this method. (for instance, it's unclear to me how much you want to say the ECAC is relatively weaker than another conference because of a school like Union).
But yes, I agree with your hypothesis. Conferences have surely converged a bit this year. You had Vermont beat Clarkson and Brown beat Providence and RMU beat UMD (last-or-near-last teams beating 1st-or-near first teams). that pretty much sums up the year.
IceIsNice
Huh? Whoopie Pies? WHERE???
Re: 2009-2010 DI Rutter Computer Rankings Thread
I'm just wondering if you couldn't get a decent ballpark comparison by simply computing the average of the RPI for each team in a given conference?
First do it for this year, then do it for the last couple of years, and see if the gaps have shortened....wouldn't that tell you something?
I'm just wondering if you couldn't get a decent ballpark comparison by simply computing the average of the RPI for each team in a given conference?
First do it for this year, then do it for the last couple of years, and see if the gaps have shortened....wouldn't that tell you something?
Re: 2009-2010 DI Rutter Computer Rankings Thread
Yeah, it would tell you something. But there's no statistical model that justifies RPI and that measure would be subject to all sorts of pitfalls. Also, do you have RPI data readily available over the years?
If anyone seriously wants to do this, this site http://it.stlawu.edu/~chodr/wchodr/past.html
has a statistical model and all the results year-after-year.
The "overall rating" in these models is "round-robin win percentage" like the one I described, and it's theoretically sound. So that'd be the most convenient thing to use where a lot of the work has already been done for you.
Yeah, it would tell you something. But there's no statistical model that justifies RPI and that measure would be subject to all sorts of pitfalls. Also, do you have RPI data readily available over the years?
If anyone seriously wants to do this, this site http://it.stlawu.edu/~chodr/wchodr/past.html
has a statistical model and all the results year-after-year.
The "overall rating" in these models is "round-robin win percentage" like the one I described, and it's theoretically sound. So that'd be the most convenient thing to use where a lot of the work has already been done for you.