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.

Not so ez????

Re: Not so ez????

TimU said:
Ever hear of Lynah Rink? How about Bill Cleary or Hobey Baker?

And Grand Forks? Give me a break. The population of Canton, NY is like 6,500 people and it's **** cold there, too. And if an alum built us a brand new $50 million rink next year, we'd be able to recruit a little, too.
Click to expand...
pfft, take your Cleary and Baker, I'll have Mayasich, Brooks, Esposito, Armstrong...

Someone obviously hasn't been to Grand Forks in winter. Siberia is more hospitable than this god forsaken hell. But I'm sure it is -27 with a 30 mph wind in Canton right now too.
 
Re: Not so ez????

bigblue_dl said:
pfft, take your Cleary and Baker, I'll have Mayasich, Brooks, Esposito, Armstrong...

Someone obviously hasn't been to Grand Forks in winter. Siberia is more hospitable than this god forsaken hell. But I'm sure it is -27 with a 30 mph wind in Canton right now too.
Click to expand...

Man, I'm going to have to pack my swimsuit for this weekends game. Sounds like it's going to be nice weather!
 
Re: Not so ez????

Dirty said:
Man, I'm going to have to pack my swimsuit for this weekends game. Sounds like it's going to be nice weather!
Click to expand...
I'm so sick of this ****ing town and this weather. It wouldn't be so bad if I didn't have to work outside in it for 6 hours each day. I haven't been able to feel the tips of my fingers since Tuesday morning, and my cheeks are so frostbitten that they're turning black. I look like Rob ****ing Green.
 
Re: Not so ez????

bigblue_dl said:
pfft, take your Cleary and Baker, I'll have Mayasich, Brooks, Esposito, Armstrong...

Someone obviously hasn't been to Grand Forks in winter. Siberia is more hospitable than this god forsaken hell. But I'm sure it is -27 with a 30 mph wind in Canton right now too.
Click to expand...

Have you been to Canton? The forecast for Sunday morning is 21 below. And yes, the wind howls there as well.
 
Re: Not so ez????

TimU said:
Have you been to Canton? The forecast for Sunday morning is 21 below. And yes, the wind howls there as well.
Click to expand...

Is that the high? Down in Syracuse, the high for Saturday is supposed to be 10F. Good thing I'll be out watching the Engineers. ;)
 
Re: Not so ez????

Fighting Sioux 23 said:
I'll let you figure that one out...

Average KRACH rating:
WCHA - 258.84
ECAC - 170.53
Hockey East - 149.61
CCHA - 148.77
AH - 31.87
UAH - 16.2
Click to expand...

Just to point out, the average Krach rating can't be right if you want to assess conference strength since it's so vulnerable to outliers. You can see that in your example, since Yale's gaudy Krach raises the ECAC so much. To take an extreme example, suppose one team had the perfect goaltender who never gave up a goal (his body was exactly cage-sized and he just sat there). That team would never lose a game and, since they'd never really have to bother with defensemen, would never tie one either. Their Krach would be effectively infinite, as would the conference average but you wouldn't want to say their conference was necessarily the best simply because they had the most awesome player ever to live (and I mean that in two senses.) That's why you use medians, as I did above. I think a fair statement about the strength of a conference is a statement about the strength of its middle teams, not the top and bottom per se.
 
Re: Not so ez????

Dirty, quit deleting posts and reposting, you're messing me up. Want to get lunch today? I should be headed your way in time for lunch.
 
Re: Not so ez????

bigblue_dl said:
I firmly believe that people who haven't experienced grand forks winter, just don't understand how brutal it is. Like I said in another thread, the radiator for the engine I'm working on was frozen solid on Tuesday, yes the coolant was mixed correctly.
Click to expand...

This.

I've spent winters in a number of places, but nothing was as bad as the winters I've experienced when I lived in Grand Forks. The story I tell most often is of the World Jr. Championships in 2005, and the Fins and Sweeds kept complaining about how cold it was, and "how could people live here" :eek:
Dirty said:
http://www.weather.com/outlook/heal...ND&sfld2=Canton, NY&clocid1=USND0146&clocid2=

Grand Forks wins!
Click to expand...

Is "wins" the word you were really looking for?
 
Re: Not so ez????

goblue78 said:
Just to point out, the average Krach rating can't be right if you want to assess conference strength since it's so vulnerable to outliers. You can see that in your example, since Yale's gaudy Krach raises the ECAC so much. To take an extreme example, suppose one team had the perfect goaltender who never gave up a goal (his body was exactly cage-sized and he just sat there). That team would never lose a game and, since they'd never really have to bother with defensemen, would never tie one either. Their Krach would be effectively infinite, as would the conference average but you wouldn't want to say their conference was necessarily the best simply because they had the most awesome player ever to live (and I mean that in two senses.) That's why you use medians, as I did above. I think a fair statement about the strength of a conference is a statement about the strength of its middle teams, not the top and bottom per se.
Click to expand...

I'm not a statistician, and it sounds like you very well may be. I understand the concept behind taking the middle teams, but wouldn't a sample size of 10-12 teams be too small to get any sort of reasonable statistical determination?

That's why I figured out the Average KRACH. Perhaps a better average would be taking their rank and determining the average rank of a conference. That way, any infinite outlier would be maxed out, and one player wouldn't account for an entire conference being rated higher than another. Again, I'm not THAT into stats, so please correct me if I'm wrong.
 
Re: Not so ez????

You're not wrong (and I am a statistician.) A sample of 10-12 teams is not too small... it just lowers the expected accuracy. For technical reasons, the right thing to do is divide by the square root of the number of teams, so that a 12 team league's median is about twice as accurate as three-team league (almost got down to UAH there.) A fifty team league would be about twice as accurate as twelve-team league. But that's the variation you have to expect in looking at 12 teams who've played about 20 games each, Without everyone playing everyone (and even that would be subject to random jitters) the median is the most robust measure.

You mention using ranks... that's a good idea (are you sure you aren't a statistician?). That's what the median really is. In a 12 team conference, it's roughly the average krach between the 6th and 7th ranked team. In an 11 team conference (CCHA) it's the Krach of the 6th ranked team. Where the uncertainty comes in is that the Krach ranking of that team is only based on twenty games or so, so we need to take the fact that Krach is an imperfect measure into account. That's what the mean absolute deviation divided by the sqaure root of the number of teams does.

I'm not saying that this is the only measure you can use, but I've tried several and they all come up with about the same result. The WCHA is a little better than the other three major conferences, who are all quite close to each other, and are in turn considerably better than the Atlantic. (Of course, that doesn't stop RIT from winning tournament games, but that's not the question we're asking.)

That was essentially the same result Brian Sullivan got in his blog post this week, just by looking at interconference W-L records, which shouldn't be too surprising, since that's the raw material for Krach as well. The advantage of this method over the raw W-L is that it gets around situations in which the interleague matchups happened to oddly match good teams in one conference uniquely against bad teams from the other -- the W-L record won't line up with quality. Sullivan's results suggest that they do this year.
 
Re: Not so ez????

SJHovey said:
Canton

Grand Forks

Both look pretty balmy.
Click to expand...


it was 27 below here this morning, and I had a very large buck, and a Pileated Woodpecker come up to the bird feeder already this morning. I can hardly believe the miracle of engineering it takes for a 3/4 ounce junco or chickadee to stay warm enough over night not to freeze solid.
 
You must log in or register to reply here.
Back
Top