2012-13 KRACH RANKINGS - Updated 03/02 FINAL w/FINAL LPH RK

ghshockeyfan · Post by **ghshockeyfan** » Sat Jan 19, 2013 10:24 am

Other rankings for comparison that I'm aware of (in no particular order)...

Computer:
(Overall)
http://minnesota-scores.net/classqrf.ph ... 0&class=-2
(Class AA)
http://minnesota-scores.net/classqrf.ph ... 10&class=3
(Class A)
http://minnesota-scores.net/classqrf.ph ... 10&class=2
???OTHERS???

Human:
(by Class)
http://letsplayhockey.com/latest-rankin ... chool.html
???OTHERS???

ghshockeyfan · Post by **ghshockeyfan** » Sun Jan 20, 2013 11:20 am

Updated 01/20 AM

ghshockeyfan · Post by **ghshockeyfan** » Tue Jan 22, 2013 8:10 am

Updated 01/22 AM

ghshockeyfan · Post by **ghshockeyfan** » Wed Jan 23, 2013 8:49 pm

Updated 01/23 AM

MNHockeyFan · Post by **MNHockeyFan** » Wed Jan 23, 2013 10:31 pm

ghshockeyfan wrote:Updated 01/23 AM

Most Recent Ranking Links:
==========================================
SECTION:
http://www.bgoski.com/KRACH_SEC.htm

Interesing to look at how the leaders of the different sections compare as it related to the discussion in the "Section Realignment" thread.

ghshockeyfan · Post by **ghshockeyfan** » Thu Jan 24, 2013 8:14 am

MNHockeyFan wrote:
ghshockeyfan wrote:Updated 01/23 AM

Most Recent Ranking Links:
==========================================
SECTION:
http://www.bgoski.com/KRACH_SEC.htm
Interesing to look at how the leaders of the different sections compare as it related to the discussion in the "Section Realignment" thread.

I suppose you'd need to look at long term trends and also good projections for realignment based on strength of teams concentrated in any one current section... I looked at section 6AA and it appears it is like its own mini-state tourney with 6 of 8 teams ranked in the top 20 overall in the state I believe...

ghshockeyfan · Post by **ghshockeyfan** » Thu Jan 24, 2013 9:55 pm

Updated 01/24 AM

ghshockeyfan · Post by **ghshockeyfan** » Fri Jan 25, 2013 9:36 am

Updated 01/25 AM

Northender · Post by **Northender** » Sat Jan 26, 2013 11:25 am

Moundsview's team is likely the hardest working group and their record reflects this. A great group to watch over the next four weeks.

ghshockeyfan · Post by **ghshockeyfan** » Sat Jan 26, 2013 1:29 pm

Updated 01/26 AM

MNHockeyFan · Post by **MNHockeyFan** » Sat Jan 26, 2013 1:58 pm

ghshockeyfan wrote:SOS (Strength-Of-Schedule) is at:
http://www.bgoski.com/KRACH_SOS.htm

Interesting how the top 4 teams are members of the Lake Conference. In addition to having (getting?) to play each other twice, they choose to schedule very good teams out of conference as well. The other member, Hopkins, is at #16 so they've chosen an easier non-conference schedule.

Other top-ranked teams:

Mounds View is quite a ways down the list at #39. When it gets to sections it's probably a legitimate question whether their schedule has tested them enough, although they seem to have a knack of winning the close games.

BSM's SOS - also weak at #35 - is seriously dragged down by their games against North Suburban Conference teams. Out of conference their schedule has been pretty strong. FWIW their boys' team opted out of the North Suburban for hockey only this season, and their schedule has been very challenging, currently #6 on KRACH.

sinbin · Post by **sinbin** » Sat Jan 26, 2013 5:40 pm

The Lake conference SOS is no surprise since they play 1/3 of their season against ~Top 10 teams. Nor are BSM and MV's easy schedules surprises. The item that surprised me was the many Section 2AA teams near the top. I would expect Lake EP and Edina, but we also see Chan/Chaska, PL, and Jefferson in the Top 11. Do they really play in such a strong conference(s), or are they scheduling the big guns for non-conference play?

Goalie-Dad · Post by **Goalie-Dad** » Sun Jan 27, 2013 9:42 am

MNHockeyFan wrote:
ghshockeyfan wrote:SOS (Strength-Of-Schedule) is at:
http://www.bgoski.com/KRACH_SOS.htm
Interesting how the top 4 teams are members of the Lake Conference. In addition to having (getting?) to play each other twice, they choose to schedule very good teams out of conference as well. The other member, Hopkins, is at #16 so they've chosen an easier non-conference schedule.

Other top-ranked teams:

Mounds View is quite a ways down the list at #39. When it gets to sections it's probably a legitimate question whether their schedule has tested them enough, although they seem to have a knack of winning the close games.
BSM's SOS - also weak at #35 - is seriously dragged down by their games against North Suburban Conference teams. Out of conference their schedule has been pretty strong. FWIW their boys' team opted out of the North Suburban for hockey only this season, and their schedule has been very challenging, currently #6 on KRACH.

The SOS in girls high school is a non-factor in evaluating teams. Teams in large conferences, with weaker teams, have little or no opportunity to play teams outside of their conference and/or section. If all conferences had the same number of teams, then SOS maybe more relevant.

Secondly, looking at the MV schedule (probably the same for BSM), 18 of their 22 victories can against teams with a higher SOS ranking. This does not make sense in evaluating team strength.

ghshockeyfan · Post by **ghshockeyfan** » Sun Jan 27, 2013 10:14 am

It's a well known fact that SOS is to some degree out of the control of teams in a conference for their conference play.

What SOS can start to tell us is a bit about non conference choices. Maybe an interesting SOS would be for NC only - just to see what teams are really stepping up and taking on top opponents when given a choice.

SOS is a big factor in deciding the ranking of teams in any sport. To say otherwise seems to indicate that an undefeated season against the worst teams in the state is just as good as an undefeated season against the very best - and I think we can logically agree these are not equal achievements.

It's unfortunate that some teams don't have as many, if any, NC chances to up their SOS. But, we can't throw out SOS as a result - IMHO, regardless of the specific sport, its gender, etc.

Goalie-Dad · Post by **Goalie-Dad** » Sun Jan 27, 2013 10:50 am

ghshockeyfan wrote:It's a well known fact that SOS is to some degree out of the control of teams in a conference for their conference play.

What SOS can start to tell us is a bit about non conference choices. Maybe an interesting SOS would be for NC only - just to see what teams are really stepping up and taking on top opponents when given a choice.

SOS is a big factor in deciding the ranking of teams in any sport. To say otherwise seems to indicate that an undefeated season against the worst teams in the state is just as good as an undefeated season against the very best - and I think we can logically agree these are not equal achievements.
It's unfortunate that some teams don't have as many, if any, NC chances to up their SOS. But, we can't throw out SOS as a result - IMHO, regardless of the specific sport, its gender, etc.

Deciding the top teams in Girls hockey should not be based on a mathematical formula like SOS. For example Prior Lake is #8 in SOS ranking with only 4 wins this season (not picking on PL, just an example). It's a weak team with a strong schedule. Is PL that much better than MV and BSM?

ghshockeyfan · Post by **ghshockeyfan** » Sun Jan 27, 2013 11:19 am

Goalie-Dad wrote:Deciding the top teams in Girls hockey should not be based on a mathematical formula like SOS. For example Prior Lake is #8 in SOS ranking with only 4 wins this season (not picking on PL, just an example). It's a weak team with a strong schedule. Is PL that much better than MV and BSM?

"You have to play the best to be the best..." ...I don't think that's what we are saying here alone.

Instead, people are looking at SOS ranking in conjunction with the performance of teams to see who "play(ed) the best" teams and evaluating how they actually performed.

The actual ranking is here:
http://www.bgoski.com/KRACH_OA.htm

In fact, for this ranking system, we calculate SOS AFTER the teams are ranked based on their performance against their schedule.

Then, once a ranking is calculated based on their results (and not on who they played alone), we can retrospectively look back and say - on average, how strong were the opponents for team X?

That's what this is:
http://www.bgoski.com/KRACH_SOS.htm

More information about KRACH is at:

ghshockeyfan wrote:From a related thread:

http://www.ushsho.com/forums/viewtopic. ... ht=#605192

ghshockeyfan wrote:I'll be curious to see how KRACH computer ranking compares against some of the others mentioned here (Minnhock's PS2, USHSHO, MyHockeyRankings). KRACH is probably as "pure" as any algorithm in that it doesn't even care about score differential. That's why I prefer it. No additional human intervention.

Interestingly, it is supposed to give you odds on a game taking place. A team with a "10" rating value is considered the average team in the state regardless of Class designation. A team with, for example, a "100" rating is a 10:1 favorite over the average team. Similarly, you can compare any two teams based on their ratings.

SOS is just an average of opponents rankings played as of now.

http://www.ushsho.com/forums/viewtopic.php?t=29709

Most Recent Ranking Links NOTE THESE ARE BOYS RANKINGS LINKS:
==========================================
OVERALL:
http://www.bgoski.com/b/KRACH_OA.htm
CLASS AA:
http://www.bgoski.com/b/KRACH_AA.htm
CLASS A:
http://www.bgoski.com/b/KRACH_A.htm
SECTION:
http://www.bgoski.com/b/KRACH_SEC.htm
SOS (Strength-Of-Schedule) is at:
http://www.bgoski.com/b/KRACH_SOS.htm
==========================================

Added info. is at:
http://www.bgoski.com
And more info. on the girls forum at:
http://www.ushsho.com/forums/viewtopic.php?t=29597
http://www.ushsho.com/forums/viewtopic.php?t=27320
Here's the real details about KRACH:
http://www.mscs.dal.ca/~butler/krachexp.htm

And, our friends at USCHO speak to KRACH vs other options (yep, 7+ years ago):
http://www.uscho.com/2005/03/17/lets-get-kraching/

In addition to their ongoing rankings for both genders at both levels:
http://www.uscho.com/rankings/
http://www.uscho.com/rankings/krach/d-i-men/
http://www.uscho.com/rankings/krach/d-i-women/
http://www.uscho.com/rankings/krach/d-iii-men/
http://www.uscho.com/rankings/krach/d-ii-iii-men/
http://www.uscho.com/rankings/krach/d-iii-women/

Basic Explanation
KRACH — or “Ken’s Ratings for American College Hockey” — is the implementation for college hockey of a sophisticated mathematical model known as the Bradley-Terry rating system, first applied to college hockey by a statistician named Ken Butler.

This method is based on a statistical technique called logistic regression, in essence meaning that teams’ ratings are determined directly from their won-loss records against one another. A key feature of KRACH is that strength of schedule is calculated directly from the ratings themselves, meaning that KRACH, unlike many ratings (including RPI) cannot easily be distorted by teams with strong records against weak opposition.

The ratings are on an odds scale, so if Team A’s KRACH rating is three times as large as Team B’s, Team A would be expected to amass a winning percentage of .750 and Team B a winning percentage of .250 if it played each other enough times. The correct ratings are defined such that the "expected" winning percentage for a team in the games it’s already played is equal to its "actual" winning percentage.

An alternative definition of a team’s KRACH rating is as the product of its Winning Ratio (winning percentage divided by one minus winning percentage) with the weighted average of its opponents’ KRACH ratings. (The definition of the weighting factor makes this equivalent to the first definition of the KRACH ratings.) In addition to KRACH and RRWP, the table above lists each team’s Winning Percentage, Winning Ratio and Strength of Schedule (the aforementioned weighted average of their opponents’ KRACH ratings).

KRACH is provided for entertainment purposes only and is not used in any official way, nor is it endorsed by USCHO.com.
From the Hub:
http://www.mngirlshockeyhub.com/page/sh ... h-rankings
http://www.bgoski.com

About KRACH:
=================
KRACH is short for “Ken's Rating for American College Hockey.” Ken is Ken Butler, a statistician, and the mathematical model he used is known as the Bradley-Terry Rating System. The system and its details are well documented in great mathematical detail online (see KRACH explanation at Ken Butler's homepage).

..
The KRACH rating system is an attempt to combine the performance of each team with the strength of the opposition against which that performance was achieved, and to summarize the result as one number, a "rating", for each team. The higher the rating, the better the team.

This system accounts for strength-of-schedule (SOS) as it ranks the teams.

Teams with equal records and differing SOS can quickly be ranked as a result (i.e. team with stronger SOS is ranked higher than other teams with same record and weaker SOS).

The ratings are calculated purely from the game results (win, loss or tie), and do not use the goals scored at all. Overtime wins count as wins. Ties, for KRACH's purposes, count as half a win and half a loss.

KRACH only counts games against Minnesota State High School League opponents.
..

ghshockeyfan · Post by **ghshockeyfan** » Sun Jan 27, 2013 1:24 pm

Updated 01/27 AM

MNHockeyFan · Post by **MNHockeyFan** » Sun Jan 27, 2013 1:53 pm

Goalie-Dad wrote:Deciding the top teams in Girls hockey should not be based on a mathematical formula like SOS. For example Prior Lake is #8 in SOS ranking with only 4 wins this season (not picking on PL, just an example). It's a weak team with a strong schedule. Is PL that much better than MV and BSM?

Goalie-Dad, a team's SOS has nothing to do with how good they are. It only measures the strength of the teams they've played! SOS is purely an evaluation tool of the quality of the opposition that each team has played.

Using your Prior Lake example above, I don't think anyone has had them ranked in the Top 20 all season, nor has anyone claimed they are a better team than either MV or BSM, even though they've played a harder schedule than either of them.

P.S. You do have to give PL credit for playing such a difficult schedule, and I wouldn't be at all surprised if they've improved over the course of the season because of it.

ghshockeyfan · Post by **ghshockeyfan** » Tue Jan 29, 2013 10:21 am

Updated 01/29 AM

Marty · Post by **Marty** » Tue Jan 29, 2013 12:34 pm

They should not have a ranking number below the top 20 in "A" and the top 25 in "AA.

Once you get past the top 40% of the teams in each level, the balance becomes almost unsortable and mostly irrelevant.

luckyEPDad · Post by **luckyEPDad** » Tue Jan 29, 2013 2:17 pm

Marty wrote:They should not have a ranking number below the top 20 in "A" and the top 25 in "AA.

Once you get past the top 40% of the teams in each level, the balance becomes almost unsortable and mostly irrelevant.

The system uses an equation that derives a number that is easily sortable. The error is placing any importance in the order. Unlike LPH rankings, KRACH attempts to predict the results of a game played between any two teams. The important number in the KRACH list is the rating, not the ordinal number placed to the left of the sorted list. I don't much care that Mounds View currently has the highest KRACH rating. I am more interested in what the KRACH numbers said about last Saturday's meeting between Eden Prairie and Shakopee. I hadn't seen Shakopee play and had no idea how good they are. At the end of the second period it looked like KRACH was right, EP dominated. However, Shakopee turned it on in the third and nearly pulled off an upset.

Cut Above · Post by **Cut Above** » Tue Jan 29, 2013 2:33 pm

Must not have had any girl HS fans/volunteers for mnhockeyrankings.com to compare rankings?

http://myhockeyrankings.com/rank.php?y= ... &v=851&g=0

Two Totally Different Algorithms

ghshockeyfan · Post by **ghshockeyfan** » Wed Jan 30, 2013 10:39 am

Cut Above wrote:Must not have had any girl HS fans/volunteers for mnhockeyrankings.com to compare rankings?

http://myhockeyrankings.com/rank.php?y= ... &v=851&g=0

Two Totally Different Algorithms

My guess is they just need access to the data/scores. I've provided a tool to do this to other ranking entities and would be happy to assist them in getting the info.

The whole reason I started this KRACH thing way back when was that there wasn't anything for girls. It was all about the boys. The more the better IMHO.

ghshockeyfan · Post by **ghshockeyfan** » Wed Jan 30, 2013 10:49 am

Updated 01/30 AM

MNHockeyFan · Post by **MNHockeyFan** » Wed Jan 30, 2013 11:19 am

I was surprised to see the huge impact that Mounds View's loss last night had on their KRACH ranking. They dropped 443 points, from 819 to 376 (-54%)! I guess this is because White Bear Lake was ranked so far below them, at #58.

At the same time I was also surprised that the impact on WBL for their win was relatively minor. They only gained 2 points, going from 10.5 to 12. 5 (+19%). Their ranking did climb 15 places, from #58 to #43.

Would appreciate hearing your explanation, ghshockeyfan, as to why a huge upset like this, so late in the season, appears to impact the losing team (the favorite) so much more than the winning team (the underdog).

HighSchoolHockeyForums.com

2012-13 KRACH RANKINGS - Updated 03/02 FINAL w/FINAL LPH RK

Other rankings for comparison

Updated 01/20 AM 22039

Updated 01/22 AM 22643

Updated 01/23 AM 23118

Re: Updated 01/23 AM 23118

Re: Updated 01/23 AM 23118

Updated 01/24 AM 23638

Updated 01/25 AM 23830

Rankings

Updated 01/26 AM 24290

Re: Updated 01/26 AM 24290

Re: Updated 01/26 AM 24290

Updated 01/27 AM 24716

Updated 01/29 AM 25187

Updated 01/30 AM 25646