Bracketology 2019

Started by Swampy, January 24, 2019, 11:12:22 PM

Previous topic - Next topic

Trotsky

Quote from: adamwif anyone is still on the fence as to the intelligence level of

The answer to this question is always no.

Dafatone

To sum up the numbers-crunching I've done so far (keep in mind that Pairwise is just RPI except in weird exceptional cases):

If we beat Brown, we're almost definitely in. There is a scenario where we aren't, though: If Western Michigan wins the NCHC, BU or BC wins Hockey East, Bowling Green wins the WCHA, AND Sucks beats us in the ECAC finals, we wind up in 14th, with at large bids only going to the top 13. I didn't crunch this for every iteration of Atlantic Hockey, but it's unlikely that changes there would get us the .0017 we need to catch Clarkson in RPI.

If we lose to Brown, we're still in better shape than not, but things start getting hairy. If everything shakes out just wrong, we fall below Providence in RPI. Bowling Green passes us if they win, and maybe still if they lose depending on how other things go. Western Michigan passes us if they win their conference, or if they lose their semi-final game and beat St. Cloud in the third-place game, but not if they beat Denver in a semi-final (though this could change with various other details). If we lose and Harvard loses, it gets very tight between the two of us. The lowest we could drop is 16th, which would have us out. How many teams pass us and how many spots are available is the question.

Root for Clarkson over Harvard, Western Michigan to lose twice, and the favorites to win in Hockey East (NE and UMass) and WCHA (Minnesota State). CHN has us at 95% to get in. That honestly sounds about right, but keep in mind their methods don't calculate how hot a team is at the moment, so they probably underrate Brown.

That being said, Brown's up to 26th in the Pairwise. What a world.

adamw

Quote from: DafatoneTo sum up the numbers-crunching I've done so far (keep in mind that Pairwise is just RPI except in weird exceptional cases):

If we beat Brown, we're almost definitely in. There is a scenario where we aren't, though: If Western Michigan wins the NCHC, BU or BC wins Hockey East, Bowling Green wins the WCHA, AND Sucks beats us in the ECAC finals, we wind up in 14th, with at large bids only going to the top 13. I didn't crunch this for every iteration of Atlantic Hockey, but it's unlikely that changes there would get us the .0017 we need to catch Clarkson in RPI.

If we lose to Brown, we're still in better shape than not, but things start getting hairy. If everything shakes out just wrong, we fall below Providence in RPI. Bowling Green passes us if they win, and maybe still if they lose depending on how other things go. Western Michigan passes us if they win their conference, or if they lose their semi-final game and beat St. Cloud in the third-place game, but not if they beat Denver in a semi-final (though this could change with various other details). If we lose and Harvard loses, it gets very tight between the two of us. The lowest we could drop is 16th, which would have us out. How many teams pass us and how many spots are available is the question.

Root for Clarkson over Harvard, Western Michigan to lose twice, and the favorites to win in Hockey East (NE and UMass) and WCHA (Minnesota State). CHN has us at 95% to get in. That honestly sounds about right, but keep in mind their methods don't calculate how hot a team is at the moment, so they probably underrate Brown.

That being said, Brown's up to 26th in the Pairwise. What a world.

my goal is to take recent play into account in future iterations of the Probability Matrix. Again, this will require someone better at math than I am to help come up with a formula.

FYI - I don't think you noted that WMU plays a Game 3 tonight - and so just that one loss would help as well.
College Hockey News: http://www.collegehockeynews.com

Dafatone

Quote from: adamw
Quote from: DafatoneTo sum up the numbers-crunching I've done so far (keep in mind that Pairwise is just RPI except in weird exceptional cases):

If we beat Brown, we're almost definitely in. There is a scenario where we aren't, though: If Western Michigan wins the NCHC, BU or BC wins Hockey East, Bowling Green wins the WCHA, AND Sucks beats us in the ECAC finals, we wind up in 14th, with at large bids only going to the top 13. I didn't crunch this for every iteration of Atlantic Hockey, but it's unlikely that changes there would get us the .0017 we need to catch Clarkson in RPI.

If we lose to Brown, we're still in better shape than not, but things start getting hairy. If everything shakes out just wrong, we fall below Providence in RPI. Bowling Green passes us if they win, and maybe still if they lose depending on how other things go. Western Michigan passes us if they win their conference, or if they lose their semi-final game and beat St. Cloud in the third-place game, but not if they beat Denver in a semi-final (though this could change with various other details). If we lose and Harvard loses, it gets very tight between the two of us. The lowest we could drop is 16th, which would have us out. How many teams pass us and how many spots are available is the question.

Root for Clarkson over Harvard, Western Michigan to lose twice, and the favorites to win in Hockey East (NE and UMass) and WCHA (Minnesota State). CHN has us at 95% to get in. That honestly sounds about right, but keep in mind their methods don't calculate how hot a team is at the moment, so they probably underrate Brown.

That being said, Brown's up to 26th in the Pairwise. What a world.

my goal is to take recent play into account in future iterations of the Probability Matrix. Again, this will require someone better at math than I am to help come up with a formula.

FYI - I don't think you noted that WMU plays a Game 3 tonight - and so just that one loss would help as well.

Thanks! I forgot about that makeup game. So, if CC beats WMU, then WMU is out of our way and we only have to worry about CC winning the NCHC. If WMU beats CC, they get a leg up in the pairwise, and we probably need them to lose both of their remaining games for us to stay ahead of them if we lose to Brown.

upprdeck

if CC beats west mich and minn duluth and st cloud then they deserve to get in.

bowling green has pretty good shot at beating Minn st

we can drop out, we can get a 2 seed..

lets just beat brown and the worry on sat..

BearLover

Quote from: Dafatone
Quote from: adamw
Quote from: DafatoneTo sum up the numbers-crunching I've done so far (keep in mind that Pairwise is just RPI except in weird exceptional cases):

If we beat Brown, we're almost definitely in. There is a scenario where we aren't, though: If Western Michigan wins the NCHC, BU or BC wins Hockey East, Bowling Green wins the WCHA, AND Sucks beats us in the ECAC finals, we wind up in 14th, with at large bids only going to the top 13. I didn't crunch this for every iteration of Atlantic Hockey, but it's unlikely that changes there would get us the .0017 we need to catch Clarkson in RPI.

If we lose to Brown, we're still in better shape than not, but things start getting hairy. If everything shakes out just wrong, we fall below Providence in RPI. Bowling Green passes us if they win, and maybe still if they lose depending on how other things go. Western Michigan passes us if they win their conference, or if they lose their semi-final game and beat St. Cloud in the third-place game, but not if they beat Denver in a semi-final (though this could change with various other details). If we lose and Harvard loses, it gets very tight between the two of us. The lowest we could drop is 16th, which would have us out. How many teams pass us and how many spots are available is the question.

Root for Clarkson over Harvard, Western Michigan to lose twice, and the favorites to win in Hockey East (NE and UMass) and WCHA (Minnesota State). CHN has us at 95% to get in. That honestly sounds about right, but keep in mind their methods don't calculate how hot a team is at the moment, so they probably underrate Brown.

That being said, Brown's up to 26th in the Pairwise. What a world.

my goal is to take recent play into account in future iterations of the Probability Matrix. Again, this will require someone better at math than I am to help come up with a formula.

FYI - I don't think you noted that WMU plays a Game 3 tonight - and so just that one loss would help as well.

Thanks! I forgot about that makeup game. So, if CC beats WMU, then WMU is out of our way and we only have to worry about CC winning the NCHC. If WMU beats CC, they get a leg up in the pairwise, and we probably need them to lose both of their remaining games for us to stay ahead of them if we lose to Brown.
WMU just lost to CC. Does that mean we're in barring CC winning the NCHC (which would require two major upsets)? WMU looked to be in fairly safe territory to make the NCAAs a month ago. Then they gave up two goals to #1 SCSU in the final three minutes and things went downhill from there. The NCHC is a brutally difficult conference.

Dafatone

Colorado College has beaten Western Michigan. This is good.

It looks like the worst we could do is 15th.

Dafatone

Quote from: BearLover
Quote from: Dafatone
Quote from: adamw
Quote from: DafatoneTo sum up the numbers-crunching I've done so far (keep in mind that Pairwise is just RPI except in weird exceptional cases):

If we beat Brown, we're almost definitely in. There is a scenario where we aren't, though: If Western Michigan wins the NCHC, BU or BC wins Hockey East, Bowling Green wins the WCHA, AND Sucks beats us in the ECAC finals, we wind up in 14th, with at large bids only going to the top 13. I didn't crunch this for every iteration of Atlantic Hockey, but it's unlikely that changes there would get us the .0017 we need to catch Clarkson in RPI.

If we lose to Brown, we're still in better shape than not, but things start getting hairy. If everything shakes out just wrong, we fall below Providence in RPI. Bowling Green passes us if they win, and maybe still if they lose depending on how other things go. Western Michigan passes us if they win their conference, or if they lose their semi-final game and beat St. Cloud in the third-place game, but not if they beat Denver in a semi-final (though this could change with various other details). If we lose and Harvard loses, it gets very tight between the two of us. The lowest we could drop is 16th, which would have us out. How many teams pass us and how many spots are available is the question.

Root for Clarkson over Harvard, Western Michigan to lose twice, and the favorites to win in Hockey East (NE and UMass) and WCHA (Minnesota State). CHN has us at 95% to get in. That honestly sounds about right, but keep in mind their methods don't calculate how hot a team is at the moment, so they probably underrate Brown.

That being said, Brown's up to 26th in the Pairwise. What a world.

my goal is to take recent play into account in future iterations of the Probability Matrix. Again, this will require someone better at math than I am to help come up with a formula.

FYI - I don't think you noted that WMU plays a Game 3 tonight - and so just that one loss would help as well.

Thanks! I forgot about that makeup game. So, if CC beats WMU, then WMU is out of our way and we only have to worry about CC winning the NCHC. If WMU beats CC, they get a leg up in the pairwise, and we probably need them to lose both of their remaining games for us to stay ahead of them if we lose to Brown.
WMU just lost to CC. Does that mean we're in barring CC winning the NCHC (which would require two major upsets)? WMU looked to be in fairly safe territory to make the NCAAs a month ago. Then they gave up two goals to #1 SCSU in the final three minutes and things went downhill from there. The NCHC is a brutally difficult conference.

I don't think it's quite that neat. Brown winning the ECAC or BU/BC winning Hockey East could threaten us as well.

BearLover

Quote from: Dafatone
Quote from: BearLover
Quote from: Dafatone
Quote from: adamw
Quote from: DafatoneTo sum up the numbers-crunching I've done so far (keep in mind that Pairwise is just RPI except in weird exceptional cases):

If we beat Brown, we're almost definitely in. There is a scenario where we aren't, though: If Western Michigan wins the NCHC, BU or BC wins Hockey East, Bowling Green wins the WCHA, AND Sucks beats us in the ECAC finals, we wind up in 14th, with at large bids only going to the top 13. I didn't crunch this for every iteration of Atlantic Hockey, but it's unlikely that changes there would get us the .0017 we need to catch Clarkson in RPI.

If we lose to Brown, we're still in better shape than not, but things start getting hairy. If everything shakes out just wrong, we fall below Providence in RPI. Bowling Green passes us if they win, and maybe still if they lose depending on how other things go. Western Michigan passes us if they win their conference, or if they lose their semi-final game and beat St. Cloud in the third-place game, but not if they beat Denver in a semi-final (though this could change with various other details). If we lose and Harvard loses, it gets very tight between the two of us. The lowest we could drop is 16th, which would have us out. How many teams pass us and how many spots are available is the question.

Root for Clarkson over Harvard, Western Michigan to lose twice, and the favorites to win in Hockey East (NE and UMass) and WCHA (Minnesota State). CHN has us at 95% to get in. That honestly sounds about right, but keep in mind their methods don't calculate how hot a team is at the moment, so they probably underrate Brown.

That being said, Brown's up to 26th in the Pairwise. What a world.

my goal is to take recent play into account in future iterations of the Probability Matrix. Again, this will require someone better at math than I am to help come up with a formula.

FYI - I don't think you noted that WMU plays a Game 3 tonight - and so just that one loss would help as well.

Thanks! I forgot about that makeup game. So, if CC beats WMU, then WMU is out of our way and we only have to worry about CC winning the NCHC. If WMU beats CC, they get a leg up in the pairwise, and we probably need them to lose both of their remaining games for us to stay ahead of them if we lose to Brown.
WMU just lost to CC. Does that mean we're in barring CC winning the NCHC (which would require two major upsets)? WMU looked to be in fairly safe territory to make the NCAAs a month ago. Then they gave up two goals to #1 SCSU in the final three minutes and things went downhill from there. The NCHC is a brutally difficult conference.

I don't think it's quite that neat. Brown winning the ECAC or BU/BC winning Hockey East could threaten us as well.
Ah, thanks. We could really drop below Prov? Even still, I didn't think this through enough--Brown+BU/BC winning would mean we'd need to be top-13.

EDIT: Yeah, I played around with the CHN "You are the Committee" tool, and it looks like there are still quite a few permutations that would lead to us missing the NCAAs. Seems like the key is Brown not winning the ECAC. Brown winning would not only would take away an at-large spot, but it would boost Providence's ranking because they played each other twice.


JasonN95

Quote from: adamwYou Are The Committee

https://www.collegehockeynews.com/ratings/yatc.php

This is slick. Works well on my iPhone. Thanks, Adam.

andyw2100

Quote from: DafatoneRoot for Clarkson over Harvard,

I realize that the post I am quoting is talking about what it takes to make sure we get into the tourney.

But if we prefer to root for the things that get is as high as possible in the pairwise, I believe we need to root for Harvard (ouch) over Clarkson.

Using Adam's "You Are The Committee" site, and selecting all the favorites, including Clarkson over Harvard, we wind up 8th in pairwise. Switch the Harvard and Clarkson result, so that we beat Harvard in the Championship game (always fun) and we wind up 7th in pairwise.

Dafatone

Quote from: andyw2100
Quote from: DafatoneRoot for Clarkson over Harvard,

I realize that the post I am quoting is talking about what it takes to make sure we get into the tourney.

But if we prefer to root for the things that get is as high as possible in the pairwise, I believe we need to root for Harvard (ouch) over Clarkson.

Using Adam's "You Are The Committee" site, and selecting all the favorites, including Clarkson over Harvard, we wind up 8th in pairwise. Switch the Harvard and Clarkson result, so that we beat Harvard in the Championship game (always fun) and we wind up 7th in pairwise.

Absolute best finish is, somehow, 6th. There's a bunch that goes into it, but ND over Penn State and UMass not doing as well in Hockey East helps us against Ohio State. Northeastern losing in the semis or finals helps us pass Northeastern, and Denver would have to lose in both the semis and the consolation game in the NCHC. We pass Clarkson and ASU if we win out.

ugarte

Quote from: Dafatone
Quote from: andyw2100
Quote from: DafatoneRoot for Clarkson over Harvard,

I realize that the post I am quoting is talking about what it takes to make sure we get into the tourney.

But if we prefer to root for the things that get is as high as possible in the pairwise, I believe we need to root for Harvard (ouch) over Clarkson.

Using Adam's "You Are The Committee" site, and selecting all the favorites, including Clarkson over Harvard, we wind up 8th in pairwise. Switch the Harvard and Clarkson result, so that we beat Harvard in the Championship game (always fun) and we wind up 7th in pairwise.

Absolute best finish is, somehow, 6th. There's a bunch that goes into it, but ND over Penn State and UMass not doing as well in Hockey East helps us against Ohio State. Northeastern losing in the semis or finals helps us pass Northeastern, and Denver would have to lose in both the semis and the consolation game in the NCHC. We pass Clarkson and ASU if we win out.
my interests are more morbid and we can definitely beat brown and still fall to 13th and out

upprdeck

we beat brown and lots of bad has to happen to not get in

we lose to brown and there is still some good that can happen to get in.

basically the higher we want to finish the more we want bad teams to win and our margin of error gets slimmer.

if we win friday it will become pretty clear by nights end just how much chaos we can root for