Opponents and Others, 2022-23

[scoop85]

I assume after this weekend's games are all complete, someone smarter than I am could figure out what needs to happen for us to get in IF we were to lose to Harvard?

We are currently 12th in the pairwise. Minnesota State is 13th. They're on to the CCHA finals. They'll only pass us if they win and we lose next round, in which case they get an autobid anyway. Alaska is 14th. They're done playing for the season, and we have enough of an RPI lead that we should remain ahead of them if we lose next round.

15th is Merrimack. They're in OT against BC right now, but I think they fall into the same "gotta win their tourney to pass us" territory that Minnesota State does, though maybe not. Let's go BC.

16th is Michigan State, they're losing to Minnesota. Same deal, they'd have to win two at which point they are in anyway.

Where I'm going with this is that we are likely finishing 12th, with some possible autobids finishing ahead of us to push us backwards which is a wash. If we lose next round, it all comes down to how many teams outside the top 16 get autobids. Atlantic Hockey is a given. CCHA hurts us cause it's either Northern Michigan stealing a spot or Minnesota State likely passing us and stealing a spot. So we're rooting for favorites everywhere else.

Go Denver. EDIT: Or Saint Cloud State. Root for BU (if we can stomach it). Root for Colgate to not win the ECAC. Root for Minnesota to beat MSU or, if they lose, Michigan to beat MSU. I think if two of those four things happen, we are in good shape.

Or we could just win a game or two next weekend.

MSU lost, so there will be no autobid outside the top 12 from the Big 10. Seems like at this point, for Cornell to miss, all the following would need to occur:
(1) Cornell loses to Harvard
(2) Colgate wins the ECAC
(3) BU does not win Hockey East*
(4) Denver/SCSU does not win the NCHC

*maybe this could also be satisfied by Merrimack winning the semis and then losing to BU in the finals, i.e. they pass us for an at-large bid. So we want Merrimack to lose next round to be sure.

CHN's probability matrix today gives us a 99.7% chance of making the NCAAs based on 20,000 Monte Carlo simulations

Not sure if trolling.

Don't look at the matrix.

It severely overestimates the likelihood of higher seeds winning (i.e., auto-bids outside the top 16). Our chances are high, but not that high.

No, not trolling. I'm not a stats guy but 20,000 Monte Carlo simualtions, I presume, accounts for all types of outcomes. But it does seem anecdotally to be too high.

Sorry, didn't mean to accuse you of trolling. This has been discussed in some other threads in the last week or so. The issue is that the model relies on KRACH rankings to predict the likelihood of different outcomes. So the 20,000 simulations are based on flawed probabilities.

Sure, the data coming out is only as good as the data going in.

[Trotsky]

The statement that the probabilities are too high has not been demonstrated empirically.  They may be, they may not, but "feeling" they are too high is not a mathematical argument.

Easy enough to run the numbers.  If people want to put some reality behind their gut feelings, they should.

[adamw]

The better way to look at 99.7% is that there are very few scenarios where Cornell doesn't get it. You are then free to argue about Cornell's true "probability" academically.

The alternative is to play out every permutation and combination of results, as opposed to 20,000 Monte Carlo simulations. In that case, however, you'd probably see wrongly worse "odds" because so many of those scenarios are unlikely.

But again, saying the Probability Matrix is useless and flawed, really misses the point. Lather, rinse, repeat.

[adamw]

It should be noted that - indeed the Probability Matrix is bad at things like this ... St. Cloud State has been mediocre (at best) lately, because it lost 3 defensemen, including its best one. So they are limping into the NCAAs, and could easily lose today, and/or next week -- which gives a below-15 NCHC team much better odds of winning that tournament than the Matrix is presenting.

My hope for the future of it is to weight recent results more heavily, on a sliding scale.  But that requires math knowledge that I don't have - because I don't want to just do it arbitrarily.

[KGR11]

The statement that the probabilities are too high has not been demonstrated empirically.  They may be, they may not, but "feeling" they are too high is not a mathematical argument.

Easy enough to run the numbers.  If people want to put some reality behind their gut feelings, they should.

Actually, jfeath17 demonstrated that KRACH-based win percentages over 65% were too high empirically five years ago. He looked at results for 1,129 games in 2016 and 2017. here.

IMO, applying this kind of "translation" of KRACH-based win percentages to empirically-based ones is probably the best way to improve the Pairwise Probability Matrix. The frustrating thing is that potential outcomes have changed since jfeath17 did this analysis (OT wins count different than regulation wins).

[Jeff Hopkins '82]

I assume after this weekend's games are all complete, someone smarter than I am could figure out what needs to happen for us to get in IF we were to lose to Harvard?

We are currently 12th in the pairwise. Minnesota State is 13th. They're on to the CCHA finals. They'll only pass us if they win and we lose next round, in which case they get an autobid anyway. Alaska is 14th. They're done playing for the season, and we have enough of an RPI lead that we should remain ahead of them if we lose next round.

15th is Merrimack. They're in OT against BC right now, but I think they fall into the same "gotta win their tourney to pass us" territory that Minnesota State does, though maybe not. Let's go BC.

16th is Michigan State, they're losing to Minnesota. Same deal, they'd have to win two at which point they are in anyway.

Where I'm going with this is that we are likely finishing 12th, with some possible autobids finishing ahead of us to push us backwards which is a wash. If we lose next round, it all comes down to how many teams outside the top 16 get autobids. Atlantic Hockey is a given. CCHA hurts us cause it's either Northern Michigan stealing a spot or Minnesota State likely passing us and stealing a spot. So we're rooting for favorites everywhere else.

Go Denver. EDIT: Or Saint Cloud State. Root for BU (if we can stomach it). Root for Colgate to not win the ECAC. Root for Minnesota to beat MSU or, if they lose, Michigan to beat MSU. I think if two of those four things happen, we are in good shape.

Or we could just win a game or two next weekend.

MSU lost, so there will be no autobid outside the top 12 from the Big 10. Seems like at this point, for Cornell to miss, all the following would need to occur:
(1) Cornell loses to Harvard
(2) Colgate wins the ECAC
(3) BU does not win Hockey East*
(4) Denver/SCSU does not win the NCHC

*maybe this could also be satisfied by Merrimack winning the semis and then losing to BU in the finals, i.e. they pass us for an at-large bid. So we want Merrimack to lose next round to be sure.

I don't think Merrimack would pass us in that scenario. But I'm not certain. If they would, then it's possible that UMass Lowell could beat BU, then beat Merrimack in the finals and maybe Hockey East gets three bids. But I don't think so.

Otherwise, yeah, you summed it up well.
Otherwise yeah, I think that's right.

Lowell is playing Merrimack in the semis. It would have to be Providence beating BU in the semis then Merrimack in the finals.  But I don't think Merrimack will get in that scenario.  Time for YATC.

[BearLover]

SCSU beat Duluth, so the NCHC now only has two teams outside the top 16 out of the four remaining semifinalists. On the other hand, one of those teams is NoDak, who beat Omaha tonight to move on. With NoDak's level of talent, seems quite possible for them to steal an at-large bid. None of this really matters too much as long as Colgate doesn't win the ECAC.

[upprdeck]

NDaK has to beat St cloud and then probably Denver unless they barf it up..

[BearLover]

NDaK has to beat St cloud and then probably Denver unless they barf it up..

They'd be an underdog to Denver, but I'd guess the matchup vs. St. Cloud is about a coin-flip.

By the way, You Are the Committee is up on CHN. I haven't found a way for Merrimack to pass Cornell without Merrimack winning the Hockey East tournament. I.e., from Cornell's perspective, the only thing we care about in Hockey East is whether BU wins or not. Hockey East cannot get 2 teams ahead of Cornell without a non-BU team winning the tourney. (Though, if Merrimack wins the semis and then loses to BU in the finals, i they get very close to us, so maybe someone else can figure out a way for them to pass us.)

Assuming Merrimack can't pass us without winning HE, that makes things simple. We are in if one of the following four events occurs:
1. We beat Harvard
2. BU wins HE
3. Denver or SCSU wins NCHC
4. Colgate does not win ECAC

[upprdeck]

Yea we need just like 1 good thing to happen.

You can see a path to #7-9 seed as well

Whether thats a good thing or not hard to tell. It would mean No harvard in the our Bracket I would think

[Jeff Hopkins '82]

Yea we need just like 1 good thing to happen.

You can see a path to #7-9 seed as well

Whether thats a good thing or not hard to tell. It would mean No Harvard in the our Bracket I would think

I don't want 8 or 9.  That means Fargo.

[Trotsky]

I've never been to North Dakota.  How bad can it be?  I've been to Potsdam.

[Dafatone]

I've never been to North Dakota.  How bad can it be?  I've been to Potsdam.

I really liked Fargo for the two nights I was there. Couple fun bars, surprisingly good sushi.

The drive there was pretty desolate, and that's coming from someone who lives in a pretty desolate place.

[ugarte]

I've never been to North Dakota.  How bad can it be?  I've been to Potsdam.

Never been to Fargo but had a nice steak dinner in Minot while otherwise not stopping in the state at all.

[BearLover]

I've never been to North Dakota.  How bad can it be?  I've been to Potsdam.

I really liked Fargo for the two nights I was there. Couple fun bars, surprisingly good sushi.

The drive there was pretty desolate, and that's coming from someone who lives in a pretty desolate place.

How did you end up there?