Jump to content
SiouxSports.com Forum

If the season ended 11/28/06


farce poobah

Recommended Posts

If the season ended today, the WCHA would get 3 of the #1 seeds. The change is WHICH WCHA teams.

Can they sustain the magic? Will UND stay on the bubble at #14? Will Bucky be on the sidelines?

Stay tuned.

1. Minnesota 30 0.6212 12-1-2 0.8667 0 0 2

2. New Hampshire 29 0.6151 9-2-1 0.7917 2 0 0

3. St Cloud St 27 0.5801 6-3-3 0.6250 0 0 0

4. Michigan Tech 27 0.5721 6-4-2 0.5833 2 0 0

http://siouxsports.com/hockey/rankings/pwr.php

Link to comment
Share on other sites

I don't think Tech will be able to sustain their current ranking; they will be better this year, but not that much better. I think St. Cloud is for real and will have an excellent chance of making the NCAA Tournament (and winning a game this time). Minnesota and New Hampshire are the most legitimate of the four teams, but again it's only November and a lot of things can happen before March rolls around. If our team can put together a string of performances like last Saturday, we'll get in as usual. Whether we can win #8 in St. Louis is a question we cannot answer at this time, but I have to lean to the negative on that right now. :silly:

End negativity. :ohmy:

Link to comment
Share on other sites

I really enjoy PWR and RPI, but I really hate what the NCAA has done with RPI. It is my belief that the PWR (mainly due to the changes in RPI, which is now even less of a reflection of a teams quality) will be a lot less volitile this year.

Agree with your observation. It seems likely that the lesser-rated conferences will do better in the rankings this year.

Link to comment
Share on other sites

I don't think Tech will be able to sustain their current ranking; they will be better this year, but not that much better. I think St. Cloud is for real and will have an excellent chance of making the NCAA Tournament (and winning a game this time). Minnesota and New Hampshire are the most legitimate of the four teams, but again it's only November and a lot of things can happen before March rolls around. If our team can put together a string of performances like last Saturday, we'll get in as usual. Whether we can win #8 in St. Louis is a question we cannot answer at this time, but I have to lean to the negative on that right now. :silly:

End negativity. :ohmy:

OK, here's reasons to be positive about the Sioux chances:

** Nothing has gone right so far this year, and we're still hanging around as a tournament team (on the bubble, but hey).

** When Phil gets healthy, he'll be fine. Not like he fell off a turnip truck and into the lineup; the guy is a proven winner.

** Johnny! and Oshie! haven't been scoring.

** Forney hasn't been 100% healthy yet. Flashes of good things, but not consistently good yet.

** Line combinations are just now getting settled.

We'll be in the mix at the end. Doesn't look like a title just yet, but then its a long road between now and March and a lot can happen between now and then.

Link to comment
Share on other sites

I really enjoy PWR and RPI, but I really hate what the NCAA has done with RPI. It is my belief that the PWR (mainly due to the changes in RPI, which is now even less of a reflection of a teams quality) will be a lot less volitile this year.

The changes to the RPI, while seemingly arbitrary, had two specific goals:

1) eliminate all the "negative" games

2) otherwise produce an ordering identical to that of the old formula

Given #2, the change was obviously only superficial -- though the weightings seem to make even less sense, the result is the same. Frankly, the previous dropping of "negative" games only from conference tournaments was completely arbitrary (and hard to implement). Choosing the weighting to mimic the old result over a year or two is silly, but considering that RPI is 75% SOS, not ridiculously out of line.

I do have the new RPI fully implemented (hopefully the above-linked page on this site will reflect it by the end of this weekend), and it humorously deviates wildly from the old formula early in the season (as there are currently about a dozen "negative" games despite there only being four by the end of last season), but it settles into the same result in the late season.

Link to comment
Share on other sites

1) eliminate all the "negative" games

By doing this I think they are also going to dull the upward effects of playing very good teams, but of course that may also be a hidden intent. One case that I can see happening is a team coming to Mariucci and getting a split. Despite this good effort, that teams RPI may likely drop for that weekend. This is due to the fact that now Minnesota's Opponents Records (currently 24th) are now more important to your team than Minnesota's record (currently 1st).

I don't know if you archived last years RPI charts, but I think they would be very interesting/useful to study in comparison to this years charts. Thanks again for all the great effort you put into this site.

Link to comment
Share on other sites

Not to discredit the SiouxSports page, but I like this comparison of all the rankings on the RPI hockey page better...

http://www.rpihockey.net/misc.rank1.shtml

If you all like KRACH, you can't be too disappointed with 7th! Personally, I like the RHEAL system the best. :D

No offense taken, they have a good site that has done a good job of calculating rankings for many years -- definitely the best source for a comparison of ordinal rankings across a broad set of interesting ranking schemes. Where the SiouxSports.com ranking pages differ is that I attempt to show significant additional detail behind just the "important" ratings (primarily RPI/PWR which are used to seed the NCAA tournament, though I also calculate KRACH which many college hockey fans consider to produce "the best" objective ranking). Each site is useful in a way the other isn't so I would certainly expect people to develop preferences based on what information they're seeking.

As a somewhat unimportant caveat, because they undoubtedly just haven't implemented it yet nor bothered to note it on the page, I don't think those RPI ratings are right. They seem to be using the new RPI weightings of .25/.21/.54, but not dropping games which hurt a team's RPI, as also changed this year. See all the footnotes on the above-linked RPI for examples of such games. On the other hand, they do seem to be a step ahead of me in not considering TUC until 10 TUC contests, but that's hardly surprising :) Soon, soon...

Link to comment
Share on other sites

Do you have to revisit the removal of games from the RPI calculation every week?

It seems like some of those game RPI's over 0.6000 are likely to help a teams RPI when their record falls to a more realistic level. Specifically the Gophs and UNH have those types of games being discounted currently.

Link to comment
Share on other sites

Do you have to revisit the removal of games from the RPI calculation every week?

It seems like some of those game RPI's over 0.6000 are likely to help a teams RPI when their record falls to a more realistic level. Specifically the Gophs and UNH have those types of games being discounted currently.

I agree that the number of dropped games should decline steadily. However, that will only benefit those teams whose RPIs have dropped low enough for those games to begin "helping", I guess providing some modest stabilization? Also keep in mind the "RPI contribution" of each game changes throughout the season as the opponent wins and loses additional games. Ugh.

Much like PWR has always been, RPI isn't really defined anymore until the last game finishes, because you don't decide which games to drop until then. Not coincidentally, the new weightings were picked precisely to minimize the number of such games at the end of the last two seasons to hopefully minimize that effect. Not bothering to drop any until the end, if that's what the rpihockey guys are doing, may be sensible.

I cannot believe you guys are worried about this at this stage of the season!
If I didn't worry about it now, I'd never get it finished by March :D
Link to comment
Share on other sites

I agree that the number of dropped games should decline steadily. However, that will only benefit those teams whose RPIs have dropped low enough for those games to begin "helping", I guess providing some modest stabilization? Also keep in mind the "RPI contribution" of each game changes throughout the season as the opponent wins and loses additional games. Ugh.

Much like PWR has always been, RPI isn't really defined anymore until the last game finishes, because you don't decide which games to drop until then. Not coincidentally, the new weightings were picked precisely to minimize the number of such games at the end of the last two seasons to hopefully minimize that effect. Not bothering to drop any until the end, if that's what the rpihockey guys are doing, may be sensible.

If I didn't worry about it now, I'd never get it finished by March :lol:

I think one of the "unintended consequences" of the RPI decrease rule might come later in the season is an increase in the instability of ratings - that last weekend, there will be more permutations of "if so-and-so loses, then their RPI falls", which will affect the RPI of teams they played .... (and theoretically, the RPI of their opponents opponents too).

All in all though, it is a change which seems likely to encourage power conference teams to schedule AHA, ECAC, and CHA teams. ...

Link to comment
Share on other sites

As promised, the new PWR.

At this point, I think the RPI and PWR duplicate precisely what the NCAA will use in its selection criteria. Though I'll finish cleaning up the RPI/PWR details screens and help pages soon, the numbers on the PWR team details screens should be correct already.

Cool. Jim, thanks for all your work on this. Much appreciated.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...