Incredibly Dull Boring 2022 Pairwise and Tournament Possibilites Thread

[Numbers]

Suprised I couldn't find anything on this.....But I suppose no chatter since there isn't much to discuss.

Everything pretty much goes by RPI now, which makes things much less interesting. And, this year is so Western dominated that there isn't even much fanbase chatter.

A few ideas at this point....
Denver and WMU are still at risk of losing their #1s.

Loveland is going to be all-west because the committee will have to keep eastern teams east attendance.

The last couple of at large spots are going to go right down to the last shift of the conf tourneys.


And, I was thinking of a slight to the PWR. What do people think of this:
1- Everything goes according to RPI, except if the RPIs are within .001 of each other, in which case....
2- The CoP will continue as is
3- H2H is changed to 1 single point to whichever team has the better head to head record
4- In the case where the RPI is very close, a team possessing both the CoP and H2H points, wins the compare.

I'm not sure where to call (RPI is very close) though. I chose .001 above just as a place holder, but I am thinking about a situation like this one:

Say Minnesota plays Ohio State 5 games (including B10 tourney)
Massachusetts plays BC 5 times (same idea)

And, the Minnesota v Massachusetts compare is really close.
Now, say that OSU and BC played a game in November.
Since both UMinn and UMass played those teams often, the result of the OSU/BC game is going to have a large effect on both teams RPIs, and therefore on the compare.

This is in some sense a strange result, since that game is early season, and didn't directly either Minn or Mass. So, it seems like a rating system in which this situation can't totally affect the compare would be better, and this is what I am going for.

In other words, the RPI parses too tightly. It makes differences where there isn't one. So, there should be a cushion for such things.

Anyone have comments?

 

[J.D.]

I forget what all the attendance rules were for each site last year but in a CHN podcast I thought they said the committee could have done more to maximize attendance last year especially without an official PWR to use and they didn't do it. So that will be interesting to see. Also curious to see what impact if any the day off in between games has on attendance. I've never had to stay at a hotel for a regional but for those that do I would think adding the extra day is more of a nuisance and will end up hurting attendance even more.

 

[purpleinnebraska]

I've never had to stay at a hotel for a regional but for those that do I would think adding the extra day is more of a nuisance and will end up hurting attendance even more.

I understand why they did it, but this will further hurt attendance, given how little there is to do in some places (No offense Allentown, Loveland, et. al.). Fans of the teams who lose in the first round will check out of the hotel and leave, as opposed to spending 1 or 2 more nights in a hotel.

 

[Scarlet]

I can't find a schedule - for some reason my company has NCAA.com blocked. USHCO just has the locations. Can someone help a girl out and paste into a post? Thanks!!

 

[J.D.]

Allentown:

1. Western Michigan (4)
2. Minnesota (5)
3. North Dakota (11)
4. Notre Dame (13)

Worcester:

1. Michigan (1)
2. St. Cloud State (8)
3. UMass (9)
4. AIC (16)

Albany:

1. Minnesota State (2)
2. Quinnipiac (6)
3. Ohio State (12)
4. Clarkson (15)

Loveland:

1. Denver (3)
2. Minnesota-Duluth (7)
3. Michigan Tech (10)
4. BU (14)

I initially had Michigan in Allentown but believe they are flying regardless so this made the most sense IF the committee cares about attendance.

 

[J.D.]

Suppose BU and Clarkson could be debated for that spot in Albany.

 

[paulfb]

I forget what all the attendance rules were for each site last year but in a CHN podcast I thought they said the committee could have done more to maximize attendance last year especially without an official PWR to use and they didn't do it. So that will be interesting to see. Also curious to see what impact if any the day off in between games has on attendance. I've never had to stay at a hotel for a regional but for those that do I would think adding the extra day is more of a nuisance and will end up hurting attendance even more.

Especially with costs of lodging and gas now way up. Luckily I live in NY City and can do the first night in Albany up and back before my girlfriend and I go to Allentown, where we have friends. But this will cost at the pump. Never thought I'd have to use loyalty points to pare down hotel costs in Bethlehem, Pa.

 

[paulfb]

I can't find a schedule - for some reason my company has NCAA.com blocked. USHCO just has the locations. Can someone help a girl out and paste into a post? Thanks!!

If it's the dates you're looking for, Albany and Loveland are Thursday-Saturday while Worcester and Allentown are Friday-Sunday.

 

[LordofBrewtown]

In other words, the RPI parses too tightly. It makes differences where there isn't one. So, there should be a cushion for such things.

Anyone have comments?

I'm generally against anything that trends back toward giving the committee more freedom in who makes the tournament/trends back to 'decisions in smoke filled rooms' vs 100% transparency. For that reason, I like the hockey selection system far better than football or basketball (which I view as corrupt).

That said, I'm not convinced that the 'quality wins bonus' is necessary or the formula is correct, and I'm also not sure I agree with the current weighting of the RPI calculation (the 25/21/54). I'd argue that too much weight is put on record of opponents' opponents. I suppose, if your a fan of the KRACH, you could argue that the Pairwise are so close to KRACH this year, that the Pairwise is fine; but, in that case they should just use the KRACH.

 

[paulfb]

I'm generally against anything that trends back toward giving the committee more freedom in who makes the tournament/trends back to 'decisions in smoke filled rooms' vs 100% transparency. For that reason, I like the hockey selection system far better than football or basketball (which I view as corrupt).

That said, I'm not convinced that the 'quality wins bonus' is necessary or the formula is correct, and I'm also not sure I agree with the current weighting of the RPI calculation (the 25/21/54). I'd argue that too much weight is put on record of opponents' opponents. I suppose, if your a fan of the KRACH, you could argue that the Pairwise are so close to KRACH this year, that the Pairwise is fine; but, in that case they should just use the KRACH.

Agree that we may need to tweak the formula, but not go back to the smoke-filled rooms.

 

[Darius]

A hot team on a roll factor would be nice. Far more interesting to watch a team that is playing great, but took a while to find its legs than one with a higher PWR trending down.

 

[Numbers]

A hot team on a roll factor would be nice. Far more interesting to watch a team that is playing great, but took a while to find its legs than one with a higher PWR trending down.

This was done for awhile many years ago. The PWR had a last16 component. But it was discarded because your record in 16 games is too dependent on scheduling.

 

[Numbers]

and, let me make clear...I'm not in favor of smoke filled rooms.

I'm in favor of using the RPI (although I might prefer KRACH, but that won't happen). I just know that it has limits.

In cases where RPI is nearly equal, I am suggesting that in those cases, the ComOpp and H@H factors be used to determine the way the Comparison falls. it would still be a completely objective ranking. Formula known ahead. No discussion in a room. Nothing like that. Just pure numbers.

 

[LordofBrewtown]

A hot team on a roll factor would be nice. Far more interesting to watch a team that is playing great, but took a while to find its legs than one with a higher PWR trending down.

I am not a fan of the 'hot team on a roll' factor - the equivalent being the old record in last 10 games when that was an ingredient in the pairwise. The reason: a team's schedule plays a huge role in what that record can be. If you are in the Big Ten - what if you had Bucky, Sparty or Ped St early in the season/maybe only have 2 total games against those bottom teams in your last 10 games? That puts you behind the 8 ball in comparison to someone who might have had 6 of their last 10 against those teams (or maybe even 8 since the schedule is often 'uneven'). You can come up with the same hypotheticals in every conference (even more amplified on larger conferences like Hockey East and the ECAC).

 

[Numbers]

Saturday morning.....
More or less 3 games left in the regular season for everyone except HEA and NCHC.

What stands out in the PWR is...

1) Western Michigan is being pushed by Minnesota and Quinnipiac for the final #1 seed.

2) On the #2/#3 edge, UMass, UMD, NoDak, SCSU are close enough that one tiny result from anywhere could flip the whole thing upside down. One would say, first, that "Well, those teams should just play each other" except that they can't, unless WMU falls out of the #1 spot.

3) On the bubble, OSU looks safer at #13 for their gap to 14, but that Lowell, Clarkson and BU could be covered with a blanket. And, there are always tourney upsets.

 

[Numbers]

Saturday night Feb 19.
Big 10, CCHA, ECAC, AHA one week left. NCHC, HEA 2 weeks left regular season.

In the PWR, right now it appears that Michigan, Minnesota State, and Denver are all in very good shape to be #1 seeds. In fact, by groups:

1-3: Michigan, Minnesota State, Denver. Denver, of course, would play in Loveland. Michigan and Minn State make little difference in their placement.

4-5: Western Michigan and Minnesota. There is a bit of a gap between these 2 schools and everyone else below them, and above them. So, it is very likely that the 4[SUP]th[/SUP] #1 comes from these 2. And, perhaps, they will end up playing in the same regional.

6-8: Quinnipiac, Minn-Duluth, and North Dakota. Again, a bit of a gap.

9-13: Massachusetts, St Cloud State, Notre Dame, Michigan Tech and Ohio State.

14-20: BU, Lowell, Connecticut, Clarkson, Northeastern, Providence, and Merrimack. All of these would have a chance if they put together a good streak of wins.

Finally, at 21, AIC leads the AHA. And, everyone else down here needs a tournament championship.

And, it should be mentioned that every comparison of interest currently breaks according to RPI except Providence over Denver.

 

[Fighting Sioux 23]

Saturday night Feb 19.
Big 10, CCHA, ECAC, AHA one week left. NCHC, HEA 2 weeks left regular season.

In the PWR, right now it appears that Michigan, Minnesota State, and Denver are all in very good shape to be #1 seeds. In fact, by groups:

1-3: Michigan, Minnesota State, Denver. Denver, of course, would play in Loveland. Michigan and Minn State make little difference in their placement.

Yes. These three are almost guaranteed to be #1 seeds. I would anticipate that they would move around Michigan / Minnesota State based on what quarterfinal pairing would be best attendance wise.

4-5: Western Michigan and Minnesota. There is a bit of a gap between these 2 schools and everyone else below them, and above them. So, it is very likely that the 4[SUP]th[/SUP] #1 comes from these 2. And, perhaps, they will end up playing in the same regional.

6-8: Quinnipiac, Minn-Duluth, and North Dakota. Again, a bit of a gap.

I agree that either Western or Minnesota will likely get that last spot. That being said, I would note that the gap between #6 (North Dakota) and #8 (Minnesota Duluth) right now is greater than the gap between North Dakota and Minnesota (and very similar to the gap between North Dakota and Western). Also, North Dakota plays Western next weekend. I don't expect UND to sweep Western; however, I didn't expect them to sweep this weekend. Ultimately, there is still a lot of hockey left to be played. If Minnesota has a hiccup against Wisconsin, it could mean they need to win the B1G Tournament to have a shot at that last #1 seed.

9-13: Massachusetts, St Cloud State, Notre Dame, Michigan Tech and Ohio State.

SCSU has really fallen off a cliff. They have the experience to regroup and right the ship, but they could slip a bit more in the next few weeks.

14-20: BU, Lowell, Connecticut, Clarkson, Northeastern, Providence, and Merrimack. All of these would have a chance if they put together a good streak of wins.

It's crazy how packed up Hockey East is in the PWR. I would have to imagine that one of the group in the 14-20 range makes a push out one of the 9-13 group.

Finally, at 21, AIC leads the AHA. And, everyone else down here needs a tournament championship.

And, it should be mentioned that every comparison of interest currently breaks according to RPI except Providence over Denver.

We may be looking at different PWR. North Dakota loses its comparison with Cornell. BU loses its comparison with Northern Michigan.

 

[Fighting Sioux 23]

I forget what all the attendance rules were for each site last year but in a CHN podcast I thought they said the committee could have done more to maximize attendance last year especially without an official PWR to use and they didn't do it. So that will be interesting to see. Also curious to see what impact if any the day off in between games has on attendance. I've never had to stay at a hotel for a regional but for those that do I would think adding the extra day is more of a nuisance and will end up hurting attendance even more.

Maximizing attendance last year wasn't really a thing, given the limited attendance permitted due to Covid. But they could have limited travel, which they didn't do.

 

[Numbers]

Yes. These three are almost guaranteed to be #1 seeds. I would anticipate that they would move around Michigan / Minnesota State based on what quarterfinal pairing would be best attendance wise.



I agree that either Western or Minnesota will likely get that last spot. That being said, I would note that the gap between #6 (North Dakota) and #8 (Minnesota Duluth) right now is greater than the gap between North Dakota and Minnesota (and very similar to the gap between North Dakota and Western). Also, North Dakota plays Western next weekend. I don't expect UND to sweep Western; however, I didn't expect them to sweep this weekend. Ultimately, there is still a lot of hockey left to be played. If Minnesota has a hiccup against Wisconsin, it could mean they need to win the B1G Tournament to have a shot at that last #1 seed.



SCSU has really fallen off a cliff. They have the experience to regroup and right the ship, but they could slip a bit more in the next few weeks.



It's crazy how packed up Hockey East is in the PWR. I would have to imagine that one of the group in the 14-20 range makes a push out one of the 9-13 group.



We may be looking at different PWR. North Dakota loses its comparison with Cornell. BU loses its comparison with Northern Michigan.

For looking at the matter again, it would perhaps be better to group as such:

1-3
4-5
6-7
8-13
14-20

In other words,
6-7: NoDak and Quinn
8-13: UMD, UMass, SCSU, NoDame, MTU, OSU

As for the comment about comparisons of interest going according to RPI...I was referring to comparisons in which both teams are currently top-20. And, my rational for that was that losing one comparison against RPI doesn't drop you in the list order. So, Prov over Den doesn't drop DU to 4th, for example.

 

[Numbers]

For discussion:
Who would draw fans to Albany?

Would WMU, Notre Dame, or Ohio State draw fans to Allentown?

How far will NoDak fans travel to watch the Fighting Hawks (it's hard to type that still)?