Re: 2009-2010 Polls thread
Particularly puzzling given Providence's two game set against Wisco, which was all PC
Particularly puzzling given Providence's two game set against Wisco, which was all PC
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 Polls thread
- Thread starter IceIsNice
- Start date
ARM
Fan of chipmunk-like mascots.
Re: 2009-2010 Polls thread
The Badgers got a vote for every time Wisconsin was mentioned on the NBC Olympic hockey telecast.
A Badger win and a tie is all PC? Whoa, a new case of eastern bias.Particularly puzzling given Providence's two game set against Wisco, which was all PC
Re: 2009-2010 Polls thread
I was at both games, and I distinctly remembered game 1 as being a Tie - shootout win for PC. I was going to check the scores before I posted this, but didn't because I thought sure PC won the second. My bad!
So, you're right, I must have succumbed to the aforementioned eastern bias, because as all here know, I'm not likely to just throw point Providence's way
So yeah, now the ranking makes sense.
A Badger win and a tie is all PC? Whoa, a new case of eastern bias.
I was at both games, and I distinctly remembered game 1 as being a Tie - shootout win for PC. I was going to check the scores before I posted this, but didn't because I thought sure PC won the second. My bad! So, you're right, I must have succumbed to the aforementioned eastern bias, because as all here know, I'm not likely to just throw point Providence's way
So yeah, now the ranking makes sense.
scrambledlegs
Registered User
Re: 2009-2010 Polls thread
Not picking on you (since you did give us the benefit of the doubt
) but I don't think Providence's ranking makes sense.
Since we played Wisconsin in November, we have gone 9-3-3 with 9 of these games being against teams ranked in the top 10. Our record against ranked teams during this time frame has been 6-2-1.
On the other hand, Wisconsin has gone 8-7-1 with only 4 games being against ranked teams (all of them UMD) and they have gone 1-2-1. So obviously, I think our results are better.
In addition, to move them up to #8 over Northeastern and Cornell makes you wonder if the voting is in fact some recognition of name rather than results.
Sure, we shouldn't have lost to UConn, but they at least were #7 vs Mankato (who we beat) is unranked.
So, you're right, I must have succumbed to the aforementioned eastern bias, because as all here know, I'm not likely to just throw point Providence's way
So yeah, now the ranking makes sense.
Not picking on you (since you did give us the benefit of the doubt
) but I don't think Providence's ranking makes sense. Since we played Wisconsin in November, we have gone 9-3-3 with 9 of these games being against teams ranked in the top 10. Our record against ranked teams during this time frame has been 6-2-1.
On the other hand, Wisconsin has gone 8-7-1 with only 4 games being against ranked teams (all of them UMD) and they have gone 1-2-1. So obviously, I think our results are better.
In addition, to move them up to #8 over Northeastern and Cornell makes you wonder if the voting is in fact some recognition of name rather than results.
Sure, we shouldn't have lost to UConn, but they at least were #7 vs Mankato (who we beat) is unranked.
Last edited:
Re: 2009-2010 Polls thread
The simplest possible explanation for why Wisconsin moved up to 8 in the USCHO poll -- because Wisconsin is No. 8 in the PWR now. There were a lot of mixed results, plenty to confuse the voters, and they just put Wisconsin 8th because they were 8th in the PWR. Simple as that.
The USCHO voters are more familiar with the PWR than the USA Today voters, explaining that distinction.
And the reason why splitting Mankato moved Wisconsin up? Because the NCAA criteria is a totally awful system that doesn't punish you enough for splitting a team like Mankato. Wisconsin's RPI is nowhere near No. 8 (No. 12), yet the PWR rewards records against teams with RPI > .500 and common opponents and H2H, and the Mankato loss only shows up in 1 of the 4 criterion in its comparisons against non-WCHA teams (which are really the only ones that matter to Wisconsin at this point).
I wish the USCHO voters would not simply regurgitate the PWR when it behaves in a completely illogical and despicable fashion such as this.
The simplest possible explanation for why Wisconsin moved up to 8 in the USCHO poll -- because Wisconsin is No. 8 in the PWR now. There were a lot of mixed results, plenty to confuse the voters, and they just put Wisconsin 8th because they were 8th in the PWR. Simple as that.
The USCHO voters are more familiar with the PWR than the USA Today voters, explaining that distinction.
And the reason why splitting Mankato moved Wisconsin up? Because the NCAA criteria is a totally awful system that doesn't punish you enough for splitting a team like Mankato. Wisconsin's RPI is nowhere near No. 8 (No. 12), yet the PWR rewards records against teams with RPI > .500 and common opponents and H2H, and the Mankato loss only shows up in 1 of the 4 criterion in its comparisons against non-WCHA teams (which are really the only ones that matter to Wisconsin at this point).
I wish the USCHO voters would not simply regurgitate the PWR when it behaves in a completely illogical and despicable fashion such as this.
ARM
Fan of chipmunk-like mascots.
Re: 2009-2010 Polls thread
More than any other season I can remember, I think that winning % is about as good a measurement as anything else. Find a way to win, no matter how good or bad the opponent.
I understand what you are saying. And some computer systems are going to view certain losses as much worse than others. Subjectively, though, I don't think that there is much difference this year from losing to a team like Mankato, or St. Cloud, or Ohio St, or even UND. And I'd put Wisconsin right in that same soup, because outside of a couple of home series against UMD and Minnesota, they've been roughly a .500 team. All of those WCHA teams have strengths and weaknesses; they've had stretches where they played well, and times when they've been bad. So in my mind, the damage isn't in losing to Mankato or any other particular opponent, it is more about losing so often given that the WCHA is down a tad and they didn't play a particularly impressive non-conference schedule.
More than any other season I can remember, I think that winning % is about as good a measurement as anything else. Find a way to win, no matter how good or bad the opponent.
Re: 2009-2010 Polls thread
Rutter also has Wisconsin at #8 (PC at #15), although I would guess that their ranking there is influenced quite a bit by the prior.
Although, I thought the prior was supposed to matter less as the season progresses?
Rutter also has Wisconsin at #8 (PC at #15), although I would guess that their ranking there is influenced quite a bit by the prior.
Although, I thought the prior was supposed to matter less as the season progresses?
ARM
Fan of chipmunk-like mascots.
Re: 2009-2010 Polls thread
I think their results vs Minnesota and UMD help them a lot in Rutter.
Re: 2009-2010 Polls thread
Right, and to a large extent, I'm complaining more about how the rankings changed on the margin, rather than the fact Wisconsin is No. 8. The fact that a loss to a really strong opponent can hurt you much more than a loss to a weak opponent is an absurd feature of the NCAA's system.
ARM
Fan of chipmunk-like mascots.
Re: 2009-2010 Polls thread
Looking at the criteria of the PWR, where does it start to go wrong? Assume for the moment that we are all okay with comparing teams one pair at a time in the first place (not a valid assumption, but it simplifies the problem like ignoring friction and wind resistance.) I don't think that too many have a problem with head to head, although we may disagree on the weighting (e.g., there could be a max of one H2H point, you win it, lose it, or draw, so the H2H point is awarded to the pair as 1-0, 0-1, or 0-0.) Common opponents is a good idea, although there are cases where it breaks down, like if two teams both play teams ranked #1 and #31, but one plays the former more while the other's schedule is heavy on the latter. RPI is probably a flawed measurement that could be improvement, but usually by the end of the season, it doesn't depart much from KRACH or Rutter. I used to like the idea of a L16, but felt it penalized teams that played in a tough conference, because most of the nonconference games fell out. So a team with a cupcake schedule down the stretch always looked hotter than teams that had been beating each other up. Maybe something like a comparison of KRACH or RPI points gained over the final 16 games would be better.
The biggest problem IMO has always been TUC. On one sense, it is a nice idea in that you want teams in the tournament that will be competitive against other good teams. The problem is that the in-or-out cutoff at .500 RPI is so arbritrary and so dependent on results not involving the two teams in question. Beyond that, it is flawed in the same way that COP is, incapable of detecting if a team is playing the majority of the games against teams at the top or bottom of the range. To me, it would greatly improve the system if TUC was dropped in favor of something like KRACH or Rutter.
Or we could go to using just KRACH or Rutter as a standalone measurement. The biggest flaw there is that it would give us very little to discuss regarding selections.
Potentially, that is true. But it is also true that a loss to a weak opponent can be very damaging. We often discuss the categories in the individual PWR comparisons as though though a point is a point, but of course, that is far from the truth. RPI is the Super Point, in that it breaks ties for both individual comparisons and those with an equal number of PWR comparisons won. So it is a slippery slope trying to hold your place in the PWR if you start to fade in the RPI.
Looking at the criteria of the PWR, where does it start to go wrong? Assume for the moment that we are all okay with comparing teams one pair at a time in the first place (not a valid assumption, but it simplifies the problem like ignoring friction and wind resistance.) I don't think that too many have a problem with head to head, although we may disagree on the weighting (e.g., there could be a max of one H2H point, you win it, lose it, or draw, so the H2H point is awarded to the pair as 1-0, 0-1, or 0-0.) Common opponents is a good idea, although there are cases where it breaks down, like if two teams both play teams ranked #1 and #31, but one plays the former more while the other's schedule is heavy on the latter. RPI is probably a flawed measurement that could be improvement, but usually by the end of the season, it doesn't depart much from KRACH or Rutter. I used to like the idea of a L16, but felt it penalized teams that played in a tough conference, because most of the nonconference games fell out. So a team with a cupcake schedule down the stretch always looked hotter than teams that had been beating each other up. Maybe something like a comparison of KRACH or RPI points gained over the final 16 games would be better.
The biggest problem IMO has always been TUC. On one sense, it is a nice idea in that you want teams in the tournament that will be competitive against other good teams. The problem is that the in-or-out cutoff at .500 RPI is so arbritrary and so dependent on results not involving the two teams in question. Beyond that, it is flawed in the same way that COP is, incapable of detecting if a team is playing the majority of the games against teams at the top or bottom of the range. To me, it would greatly improve the system if TUC was dropped in favor of something like KRACH or Rutter.
Or we could go to using just KRACH or Rutter as a standalone measurement. The biggest flaw there is that it would give us very little to discuss regarding selections.
ARM
Fan of chipmunk-like mascots.
Re: 2009-2010 Polls thread
"Perhaps it was at that moment that the hockey gods began to align the stars to ensure a first-round meeting of Wisconsin and Mercyhurst..."
ManBehindTheCurtain
Great and powerful wizard
Re: 2009-2010 Polls thread
The defending champion against the number 1 contender.
May the gods make it so.
This would be a natural attraction that everyone should want to see:"Perhaps it was at that moment that the hockey gods began to align the stars to ensure a first-round meeting of Wisconsin and Mercyhurst..."
The defending champion against the number 1 contender.
May the gods make it so.
IceIsNice
Huh? Whoopie Pies? WHERE???
ARM
Fan of chipmunk-like mascots.
MICZamboni
Mr. Know It All
Re: 2009-2010 Polls thread
With Mercyhurst's track record against Wisco, I would have waited till the NCAA's are over before I made a comment like this.
With Mercyhurst's track record against Wisco, I would have waited till the NCAA's are over before I made a comment like this.
Re: 2009-2010 Polls thread
3/1/10 uscho.com poll
1 Mercyhurst (15) 27-2-3 150 1
2 Minnesota-Duluth 26-8-2 133 2
3 Minnesota 24-7-5 121 3
4 Harvard 20-6-5 98 4
5 New Hampshire 19-7-5 88 5
6 Clarkson 22-10-5 65 6
7 Connecticut 20-8-7 58 7
8 Cornell 17-8-6 42 9
9 Northeastern 17-9-7 26 8
10 Quinnipiac 19-10-8 14 NR
Others Receiving Votes: Providence 13, Ohio State 5, Boston University 4, Wisconsin 1
Quinnipiac moves into the top ten despite being eliminated by RPI.
3/1/10 uscho.com poll
1 Mercyhurst (15) 27-2-3 150 1
2 Minnesota-Duluth 26-8-2 133 2
3 Minnesota 24-7-5 121 3
4 Harvard 20-6-5 98 4
5 New Hampshire 19-7-5 88 5
6 Clarkson 22-10-5 65 6
7 Connecticut 20-8-7 58 7
8 Cornell 17-8-6 42 9
9 Northeastern 17-9-7 26 8
10 Quinnipiac 19-10-8 14 NR
Others Receiving Votes: Providence 13, Ohio State 5, Boston University 4, Wisconsin 1
Quinnipiac moves into the top ten despite being eliminated by RPI.