Jump to content
SiouxSports.com Forum

Recommended Posts

Posted
This is one reason why the pairwise system is flawed. You sometimes have a situation in the first round of the playoffs where it is better for your PWR to beat a team in 3 games (to ensure they remain a TUC) as opposed to sweeping them in 2. My guess is that it may be the same way next weekend.

yeah, the pairwise needs to be overhauled. i mean, as long as you beat all the tuc teams, you can lose to every non-tuc team you play in a season and it won't affect your pwr.

  • Replies 246
  • Created
  • Last Reply

Top Posters In This Topic

Posted
Week to week there's a lot to complain about the PWR. At the end of the year, it's usually pretty fair, although not perfect.

By fair I mean as compared to the KRACH which week to week is probably as good as it will get, but not perfect. But what else are you going to go by?

personally, i think the pairwise could be updated so that all 58 teams are compared to all of the other 57 teams. why not rank every team 1 - 58? that way every game truely counts.

Posted
personally, i think the pairwise could be updated so that all 58 teams are compared to all of the other 57 teams. why not rank every team 1 - 58? that way every game truely counts.

I thought the RPI which is a big part of the PWR considered your entire record. I could be wrong about that.

Posted
I thought the RPI which is a big part of the PWR considered your entire record. I could be wrong about that.

i don't know if that is true or not, but rpi is only a tie-breaker. your pwr is determined by the number of tucs (teams under consideration) that you beat in each comparison.

Posted
i don't know if that is true or not, but rpi is only a tie-breaker. your pwr is determined by the number of tucs (teams under consideration) that you beat in each comparison.

The top 25 RPI teams are the 25 Teams Under Consideration. Then, head-to-head RPI matchup of each team in the top 25 garners a point. There are 4 total criteria for points (RPI, Common Opponent Record, Record Against Teams Under Consideration (minimum 10 games), and head-to-head record).

For example:

Team A:

RPI: .5831

Record Against Teams Under Consideration: 12-3-5

Record Against Common Opponents (Common Opponents being record against the same teams Team B has Played): 3-4-2

Head-to-Head Record against Team B: 1-2-1

Team B:

RPI: .5683

Record against TUC: 10-6-2

Record Against Common Opponents: 5-1-3

Head-to-Head Record against Team A: 2-1-1

-Team A gets a point for RPI, better record against TUC, and 1 point for 1 win against Team B

-Team B gets a point for better Record against Common Opponents, and 2 points for 2 wins against Team A

-Both teams have 3 points against each other

-If teams have an equal amount of points, the first tie-breaker is head-to-head record

-Team B gets the Pairwise point

-If both teams share the same head-to-head record, then the tie-breaker is RPI...which would then give it to team A

So, each team gets matched up this way against the other 24 teams under consideration, and the Pairwise then ranks those 25 teams by whoever has the most head-to-head wins.

Therefore, RPI is much more essential than just a tie-breaker. It determines which teams are considered (meaning that those bubble teams at the bottom of the list in RPI are essential...like the Gophers...because one week they could be a TUC, and the next week not...shaking up everyone's point totals against each other). RPI also is big because of that point factor for each head-to-head matchup.

Sorry for the lenghty post, but I hope it clears that up on how important RPI can be...besides just being a tie-breaker.

From USCHO.com on how RPI is determined:

"The Ratings Percentage Index is one tool used to select teams for the national collegiate ice hockey tournament. Only results from games between two teams that each play 20 or more games against Division 1 opponents are used. Factors involved are 1) the team's winning percentage; 2) the average winning percentage of the team's opponents; and 3) the average winning percentage of the team's opponents' opponents. These factors are multiplied by 25%, 21%, and 54% respectively."

Posted

Now that we are at #5 in the PWR with 19 comparisons won, can we move up?

http://siouxsports.com/hockey/rankings/pwr...ls.php?teamid=0

We currently are losing comparisons with

Denver: 0-4 head-to-head make that comp next to impossible to flip

Miami: they have too big a margin in all the categories (RPI/TUC/COP) to realistically flip.

BC: even if we win 4 and they go 2 and out it will still be hard to flip this one.

Alaska: TUC and COP are close - we could flip this one.

Wisconsin: best case is to face them in F5 and beat them - this would even out the H2H. If we go on a run we could overtake them in RPI and maybe COP, then we could flip this one as well.

One huge key is to have UMTC stay a TUC after we beat them Fri/Sat.

Thoughts?

Posted
Now that we are at #5 in the PWR with 19 comparisons won, can we move up?

Thoughts?

OK so here is my NERD take on where we stand.

The short answer is yes, we can quite easily move up. We are almost certainly going to lose comparisons to Denver, Wisconsin, and Miami, but we can still get a #1 seed with #4 overall. Right now we are 2 comparisons behind BC, and are losing 5 total. The three above and BC and Alaska.

I we flip the BC comparison we gain 2 and pass them up because we will likely have the better RPI. The TUC system is really dumb and arbitrary because there is a cutoff at 25. If you look, BC wins this but almost all of there wins are against the teams currently ranked 23-25. I would expect that some or all of these teams will slip (one of which is actually playing BC). Then UND very well could flip TUC, that matchup, and become #4 overall. If BC wins this weekend it helps the Sioux and if BC loses this weekend it helps the Sioux.

Alaska is in a similar situation where there just aren't many TUCs they have played. In Northern Michigan can eliminate Alaska this weekend, look for that comparison to flip.

Bottom line is that the Sioux likely control their own destiny for a #1 seed. This is important because I think that there are only 2 teams in the nation that can beat the Sioux if the Sioux play their best game. Denver and Wisconsin, that doesn't mean they will, but these are the class of the NCAA and if we get a #1 seed we can avoid them until Detroit.

My guess is that if UND wins out we will be the #1 seed in St. Paul.

Bottom line, people sitting in a smoke filled room SUCKS, PWR SUCKS less, KRACH rules and takes human bias out of the picture.

Posted
OK so here is my NERD take on where we stand.

The short answer is yes, we can quite easily move up. We are almost certainly going to lose comparisons to Denver, Wisconsin, and Miami, but we can still get a #1 seed with #4 overall. Right now we are 2 comparisons behind BC, and are losing 5 total. The three above and BC and Alaska.

I we flip the BC comparison we gain 2 and pass them up because we will likely have the better RPI. The TUC system is really dumb and arbitrary because there is a cutoff at 25. If you look, BC wins this but almost all of there wins are against the teams currently ranked 23-25. I would expect that some or all of these teams will slip (one of which is actually playing BC). Then UND very well could flip TUC, that matchup, and become #4 overall. If BC wins this weekend it helps the Sioux and if BC loses this weekend it helps the Sioux.

Alaska is in a similar situation where there just aren't many TUCs they have played. In Northern Michigan can eliminate Alaska this weekend, look for that comparison to flip.

Bottom line is that the Sioux likely control their own destiny for a #1 seed. This is important because I think that there are only 2 teams in the nation that can beat the Sioux if the Sioux play their best game. Denver and Wisconsin, that doesn't mean they will, but these are the class of the NCAA and if we get a #1 seed we can avoid them until Detroit.

My guess is that if UND wins out we will be the #1 seed in St. Paul.

Bottom line, people sitting in a smoke filled room SUCKS, PWR SUCKS less, KRACH rules and takes human bias out of the picture.

I agree with most of your analysis, except I don't see UND getting the St. Paul regional as the final #1 seed. It definitely depends on a few other things, but I see the committee sending us out East if we are fortunate enough to sneak into the #4 spot.

Posted
OK so here is my NERD take on where we stand.

The short answer is yes, we can quite easily move up. We are almost certainly going to lose comparisons to Denver, Wisconsin, and Miami,

DU and miami yes are unflippable however wisc still could be with good fortune. I think if we could sweep this weekend and wisc lose one and we end up playing wisc at the final five and beat them that could be enough to flip it. for us.

Posted

The whole pairwise system needs to be scraped or drastically remade.

It is an absurd system where it may be better to lose a game than win due to the TUC cliff.

Let's look at the current UND/Yale comparison.

RPI: UND 0.5541 to Yale 0.5460 UND wins

COP: UND 1-2-1 to Yale 2-0-1 Yale wins

TUC: UND 9-10-4 to Yale 4-2-2 No point awarded since Yale has less than 10 games against TUC.

Since it is 1-1 the tie is broken by RPI so UND wins the comparison.

So it is OK to use an extremely small sample to award an entire point for COP but TUC needs at least 10 games ? Where is the logic in that ?

Since Yale is at 8 games why not just "award" them 2 losses which makes them 4-4-2 and would then give them the TUC point which means they win the comparison 2-1.

In other words, if one team was 0-10 against TUC and the other one was 9-0 then no point gets awarded ? That is ridiculous.

Assume the team that is short games loses the remaining number to get to 10. If they still have a better TUC record then award them the point.

There are so many holes in the pairwise that it is just not a creditable system.

Posted
The whole pairwise system needs to be scraped or drastically remade.

It is an absurd system where it may be better to lose a game than win due to the TUC cliff.

Let's look at the current UND/Yale comparison.

RPI: UND 0.5541 to Yale 0.5460 UND wins

COP: UND 1-2-1 to Yale 2-0-1 Yale wins

TUC: UND 9-10-4 to Yale 4-2-2 No point awarded since Yale has less than 10 games against TUC.

Since it is 1-1 the tie is broken by RPI so UND wins the comparison.

So it is OK to use an extremely small sample to award an entire point for COP but TUC needs at least 10 games ? Where is the logic in that ?

Since Yale is at 8 games why not just "award" them 2 losses which makes them 4-4-2 and would then give them the TUC point which means they win the comparison 2-1.

In other words, if one team was 0-10 against TUC and the other one was 9-0 then no point gets awarded ? That is ridiculous.

Assume the team that is short games loses the remaining number to get to 10. If they still have a better TUC record then award them the point.

There are so many holes in the pairwise that it is just not a creditable system.

they changed this a couple of years a go for reasons like this. as there were comparisons being lost because team a would play 20 TUC games and say have a winning perctg. of .680 and team b woould play one game and be 1-0 and get that point. they wanted to make it a little fairer. even if UND were to sweep this weekend odds are minnesota would still remain a TUC becaue they have a decent enough rpi and would take that huge of a hit.

Posted
At the X: DU vs RIT

SCSU vs Yale

In Worcester: BC vs Ferris

UND vs MI State

In Albany: WI vs N. Michigan

BSU vs Cornell

Midwest: Miami vs UMD

Alaska vs NH

ummm, since BC has ruined my last 3 frozen 4's as a sioux fan can we play michigan please so we can skate around red's boys again? :) jk, we will smoke them in a regional, just no frozen 4's vs BC anymore ;)

Posted
At the X: DU vs RIT

SCSU vs Yale

In Worcester: BC vs Ferris

UND vs MI State

In Albany: WI vs N. Michigan

BSU vs Cornell

Midwest: Miami vs UMD

Alaska vs NH

That West Regional would be an attendance disaster at the X! They need to find a way to get UMD and/or UND there.

Posted
That West Regional would be an attendance disaster at the X! They need to find a way to get UMD and/or UND there.

Damned straight there Shawn-O. Unless the Goofs ride a wave to the auto-bid, the Sioux will be at the X come hell or high water. Apologies to the guy who doesn't like sandbagging or high water references on the board this time of year.

What really bites is that the CHA could get two teams in the NCAAs if BSU falters in the tournament. If Robert Morris beats BSU in the tourney, that should be all she wrote for the Beavers.

Posted
Damned straight there Shawn-O. Unless the Goofs ride a wave to the auto-bid, the Sioux will be at the X come hell or high water. Apologies to the guy who doesn't like sandbagging or high water references on the board this time of year.

What really bites is that the CHA could get two teams there if BSU falters in the tournament. If Robert Morris beats BSU in the tourney, that should be all she wrote for the Beavers.

yes the beavers are in a position that we havent seen really. they have a autobid locked up and there are many bubble teams out there that are huge bsu fans this weekend because if they have a hic up someone gets left out that would have been in

Posted

Finishing the WCHA tournament strong can put the Sioux into the NCAAs as a one seed. The one seed is coveted for the ability to make the last line change.

In close, playoff hockey, the ability to get the matchups you want can be the difference maker.

Ending the Gophers' season would be fun, and as a Final 5 regular I want the Sioux in St Paul, but we have bigger goals.

One game at a time, play our game, and stay healthy.

Interestingly, I like our odds playing DU again. It is really, really hard to beat any team five times in one year...

Posted
That West Regional would be an attendance disaster at the X! They need to find a way to get UMD and/or UND there.

I already "tin hat" explained how that could happen:

The WCHA needs Minnesota at The X for the Final Five for the revenue so UND "loses" (with league/officiating help) WCHA opening round.

UND falls in PWR to somewhere 7-10 and is seeded 2 or 3 at the NCAA West at The X the next week.

WCHA makes money.

NCAA makes money.

UND goes on four game winning streak. :)

Posted
By request...

I didn't bother with the in-between scenarios:

What's good to see in the UMTC graphic is that even if they get swept, they more than likely will stay in the top 25 RPI ranking and remain a TUC. Our 2 wins vs. a TUC will help us climb the PWR ladder to possibly a #1 seed.

Just keep winning.

Posted
I suck at figuring the pairwise. did we gain any ground after last nights butt kicking we put on the gophers.

thanks

Your RPI went up obviously, and you got an extra TUC win. Losing the comparison against #4 BC though in TUC and CoP. You can't take the CoP comparison against them because at most you have 1 potential game against Denver, and they have 1 potential game against Merrimack. To get that # 4 you would want Maine to drop out as a Tuc (BC is 2-0-1 against), and you want Minnie to stay up in the top 25 even if you stomp them again tonight. It'd also help if they dropped games 2 and 3 against MassAmherst. Either way, if you're in that 4/5 band NoDak will be heading out to the Northeast and playing a defacto home game against BC.

Posted
Your RPI went up obviously, and you got an extra TUC win. Losing the comparison against #4 BC though in TUC and CoP. You can't take the CoP comparison against them because at most you have 1 potential game against Denver, and they have 1 potential game against Merrimack. To get that # 4 you would want Maine to drop out as a Tuc (BC is 2-0-1 against), and you want Minnie to stay up in the top 25 even if you stomp them again tonight. It'd also help if they dropped games 2 and 3 against MassAmherst. Either way, if you're in that 4/5 band NoDak will be heading out to the Northeast and playing a defacto home game against BC.

That worked pretty well for the Sioux when they played in a NCAA verison of the Bean Pot a few years back.

Posted
That worked pretty well for the Sioux when they played in a NCAA verison of the Bean Pot a few years back.

BC is a easy win in regionals for the sioux :) now frozen 4's on the otherhand :) bad memories :D:)

Posted
BC is a easy win in regionals for the sioux :) now frozen 4's on the otherhand :) bad memories :D:)

I was thinking the same thing. Also, Sioux might want to get some revenge on the Eagles. Nothing would make me more happy than to watch Muse get lit up for about 6 goals.

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...