Bracketology 2016-17 Style

[Beeeej]

sweep next weekend and it takes care of itself.

I don't believe even sweeping next weekend is a guarantee. There are still some conference tournament upset possibilities, more so because we're at #11 and can't afford to lose a game to Clarkson.

[Dafatone]

I still wish Lake Placid had a consolation game.  Rationally, I know such a thing is just as likely to doom a team as help them, but irrationally, I can't shake "win one in LP and we're in" from my head.

Also, I know we aren't there yet.

[upprdeck]

I know that, its also a known that several teams right around us have to loss 2 games before its all done

nd/prov and vt/bc st and cloud/ndak 3 of them will lose twice next week

osu/wisc someone is losing 1-2 times next week and several of b10 teams have to lose again as well the following week

with  a sweep there is only so far we can fall with 1 loss in placid.

vt/bc hurts us if they run the table but one will be out by next week and the same for nd/prov us 2-1 and them 0-2 or 1-2 they wont catch us.

st cloud would win 2-1 over NDak and then loss the following week and not be .500 and hurt ndak chances too.

[Trotsky]

I still wish Lake Placid had a consolation game.

Me too, but IIRC the coaches were unanimously in favor of killing it.

As for the league's ostensible reason of improving our bids, the best way to do that would be to cut the ECAC tourney back to 8 teams and bar the PWR-killer Cinderellas from taking a run at our bubble teams.

[Trotsky]

Back to 10 in PWR, with the cut at 14.

[adamw]

Gosh, I'm sorry I missed the earlier conversation ... but just to point some things that maybe already were, I don't know ...

This is our Pairwise Probability Matrix ... The methodology is explained right on the page. 20,000 Monte Carlo style simulations based on KRACH. Whether KRACH overweights the possibility of winning one game, I don't know. It's the basis of a very good discussion that I'd love to be a part of. It's a good question. But using scores differentials also has its flaws.  So, you are free to take it for what it's worth.

http://www.collegehockeynews.com/ratings/probabilityMatrix.php

This is my article today based off of the latest Matrix, about projecting a final bracket...

http://www.collegehockeynews.com/news/2017/03/05_ncaa_bracket_abcs;_early_march.php

I've needled Jim, and others, before about bothering to discuss a "Bracketology" article that lays out a bracket "if the season ends today" - because it doesn't end today, which makes it rather pointless - to me. But to each their own I suppose. I usually stick to just laying out possible caveats and interesting potential matchups. But with the Matrix, and us being close to selection day, I used the Matrix projection to project a bracket.

Whether the Pairwise will fluctuate a lot between now and March 19th depends upon your definition of a lot. I have found that things really haven't changed much since December, relatively speaking. I think there is less fluctuation than there was in years past because nowadays the Pairwise is almost exclusively based upon the RPI. The other criteria come into play very little, and if you look at the Grid, you see very few cases where teams below others in the overall list, beat them in individual comparisons. The Grid used to look like an eyechart. Now it's fairly clean.

As for the tools to see what would happen "if only we won this game" or "lost that game" - etc... Our Customizer allows you to do that. Click the Tab from the Pairwise page.  This was inspired by John Whalen's old scripts.

[Jim Hyla]

Gosh, I'm sorry I missed the earlier conversation ... but just to point some things that maybe already were, I don't know ...

This is our Pairwise Probability Matrix ... The methodology is explained right on the page. 20,000 Monte Carlo style simulations based on KRACH. Whether KRACH overweights the possibility of winning one game, I don't know. It's the basis of a very good discussion that I'd love to be a part of. It's a good question. But using scores differentials also has its flaws.  So, you are free to take it for what it's worth.

http://www.collegehockeynews.com/ratings/probabilityMatrix.php

This is my article today based off of the latest Matrix, about projecting a final bracket...

http://www.collegehockeynews.com/news/2017/03/05_ncaa_bracket_abcs;_early_march.php

I've needled Jim, and others, before about bothering to discuss a "Bracketology" article that lays out a bracket "if the season ends today" - because it doesn't end today, which makes it rather pointless - to me. But to each their own I suppose. I usually stick to just laying out possible caveats and interesting potential matchups. But with the Matrix, and us being close to selection day, I used the Matrix projection to project a bracket.

Whether the Pairwise will fluctuate a lot between now and March 19th depends upon your definition of a lot. I have found that things really haven't changed much since December, relatively speaking. I think there is less fluctuation than there was in years past because nowadays the Pairwise is almost exclusively based upon the RPI. The other criteria come into play very little, and if you look at the Grid, you see very few cases where teams below others in the overall list, beat them in individual comparisons. The Grid used to look like an eyechart. Now it's fairly clean.

As for the tools to see what would happen "if only we won this game" or "lost that game" - etc... Our Customizer allows you to do that. Click the Tab from the Pairwise page.  This was inspired by John Whalen's old scripts.

Adam, it's okay to look at what it would have been like "if only we won this game", but not "if the season ends today"?

After all you do say "I have found that things really haven't changed much since December". So it's kind of fun to follow the change in tournament possibilities over time. No one here, I hope, thinks it's reality.

Come on, all of this is discussion, which is the purpose of this, and all other forums. It's generally harmless, so why do you care?

[adamw]

Adam, it's okay to look at what it would have been like "if only we won this game", but not "if the season ends today"?

After all you do say "I have found that things really haven't changed much since December". So it's kind of fun to follow the change in tournament possibilities over time. No one here, I hope, thinks it's reality.

Come on, all of this is discussion, which is the purpose of this, and all other forums. It's generally harmless, so why do you care?

I did say I was just needling you. And to each their own. I'm not exactly losing sleep over it. For a fun discussion, sure. Some people take the Bracketology articles way too seriously though, as if they have any meaning, which they don't really. So I like to remind people of that. Not you, necessarily. I just personally find the exercise pointless, and void of any educational value.

[Trotsky]

I just personally find the exercise pointless, and void of any educational value.

I used to torture Anne with "what ifs."  She hated thinking about them, I loved speculating.  This is one of those "there are two types of people" things.

We all know that the only thing that matters is the performance on the ice.  I see no harm in speculating, and in fact I think the trial run picking of brackets does teach more casual fans how the field is chosen and seeded.

[Trotsky]

Latest odds from PlayoffStatus.com:

ECAC Frozen Four: 75%
ECAC Title Game: 38%
ECAC Champions: 18% (1 in 6)

Make the NCAA tourney: 91%
Round 2: 43%
Frozen Four: 19% (1 in 5)
Title Game: 9% (1 in 11)
NCAA Champions: 4% (1 in 25)

I'll take those odds.  After 36 seasons as a fan they don't seem daunting.  

[Beeeej]

Latest odds from PlayoffStatus.com:

ECAC Frozen Four: 75%
ECAC Title Game: 38%
ECAC Champions: 18% (1 in 6)

Make the NCAA tourney: 91%
Round 2: 43%
Frozen Four: 19% (1 in 5)
Title Game: 9% (1 in 11)
NCAA Champions: 4% (1 in 25)

I'll take those odds.  After 36 seasons as a fan they don't seem daunting.  

But Greg, those odds don't take into account the bad bounces that frequently occur in hockey games, or the possibility that a giant space squid will crash down through the roof of Lynah Rink and land on Mitch Gillam. Why do you hate America?

[BearLover]

Latest odds from PlayoffStatus.com:

ECAC Frozen Four: 75%
ECAC Title Game: 38%
ECAC Champions: 18% (1 in 6)

Make the NCAA tourney: 91%
Round 2: 43%
Frozen Four: 19% (1 in 5)
Title Game: 9% (1 in 11)
NCAA Champions: 4% (1 in 25)

I'll take those odds.  After 36 seasons as a fan they don't seem daunting.  

But Greg, those odds don't take into account the bad bounces that frequently occur in hockey games, or the possibility that a giant space squid will crash down through the roof of Lynah Rink and land on Mitch Gillam. Why do you hate America?

You joke, but those odds look far more realistic than the probability matrix ones.  75% chance of making Lake Placed is considerably more reasonable than 95% or whatever the other model gave us.

[Beeeej]

Latest odds from PlayoffStatus.com:

ECAC Frozen Four: 75%
ECAC Title Game: 38%
ECAC Champions: 18% (1 in 6)

Make the NCAA tourney: 91%
Round 2: 43%
Frozen Four: 19% (1 in 5)
Title Game: 9% (1 in 11)
NCAA Champions: 4% (1 in 25)

I'll take those odds.  After 36 seasons as a fan they don't seem daunting.  

But Greg, those odds don't take into account the bad bounces that frequently occur in hockey games, or the possibility that a giant space squid will crash down through the roof of Lynah Rink and land on Mitch Gillam. Why do you hate America?

You joke, but those odds look far more realistic than the probability matrix ones.  75% chance of making Lake Placed is considerably more reasonable than 95% or whatever the other model gave us.

I don't recall a probability matrix that gave us such great odds of making the ECAC semis. I recall the CHN probability matrix that gave us - and still gives us - a 98% chance of making the NCAA field. Is that what you're thinking of?

[adamw]

You joke, but those odds look far more realistic than the probability matrix ones.  75% chance of making Lake Placed is considerably more reasonable than 95% or whatever the other model gave us.

What Beeeej said.  In fact, if you look at CHN's Probability Matrix, our odds of Cornell winning the ECAC Tournament are pretty much exactly the same as the other site that's mentioned:

http://www.collegehockeynews.com/ratings/probabilityMatrix.php

(I actually did not re-run this after last night's games, FWIW)

We don't do all of the other breakdowns, but I assume we use the same methodology. Nowhere did anyone say, or suggest, that Cornell has a 95% chance of making the next round.

[Al DeFlorio]

Finally, according to this, our probabilities for tourney advance are currently:

.85 Tourney
.42 QF
.19 SF
.09 F
.04 That Which Shall Be Nameless

Is there a way to check what a loss to RPI would do to us in the PWR?  Because unless such a loss would knock us down just a few spots, that 85% number feels way too high.

As I recall, I think it was this 85% number that caused the controversy in the first place.  BearLover's last sentence is key.  The 85% number didn't assume we had to beat RPI in order to make the NCAAs, as a loss to RPI still left a certain, not at all insignificant, % possibility of making the tournament even with a loss.