John t whelan ranking simulator

[Patman]

Re: John t whelan ranking simulator

I would take that regional in a heart beat.

[FlagDUDE08]

Re: John t whelan ranking simulator

As of games completed 23 November 2013, 11:59 Aleutian Time:

5.00 Minnesota
4.75 Michigan
4.50 Ferris State
4.25 Providence
4.00 St. Cloud State
3.75 Quinnipiac
3.50 Boston College
3.25 Cornell
3.00 LSSU
2.75 Minnesota State Mankato
2.50 Clarkson
2.25 UMASS Lowell
2.00 Yale
1.75 Bowling Green
1.50 Wisconsin
1.25 Notre Dame
1.00 New Hampshire
0.75 Miami
0.50 Union
0.25 Northeastern

And the tournament field:

Minnesota
Michigan
Providence
Ferris State

Quinnipiac
St. Cloud State
Boston College
Cornell

LSSU
UMASS Lowell
Clarkson
Bowling Green

New Hampshire
Wisconsin
Yale
AHA Champ (37 - Air Force)

NOTE: Minnesota State Mankato was removed from 10th overall due to their sub-500 record.

[FlagDUDE08]

Re: John t whelan ranking simulator

As of games completed 22 November 2013, 11:59 Aleutian Time:

5.00 Quinnipiac
4.75 Minnesota
4.50 Michigan
4.25 Ferris State
4.00 St. Cloud State
3.75 Boston College
3.50 LSSU
3.25 Providence
3.00 Yale
2.75 Cornell
2.50 Wisconsin
2.25 Bowling Green
2.00 Notre Dame
1.75 New Hampshire
1.50 Clarkson
1.25 UMASS Lowell
1.00 Miami
0.75 Northeastern
0.50 Northern Michigan
0.25 Union

And the tournament field:

Minnesota
Quinnipiac
Michigan
Ferris State

Boston College
St. Cloud State
LSSU
Providence

Wisconsin
Yale
New Hampshire
Cornell

Notre Dame
Bowling Green
UMASS Lowell
AHA Champ (36 - Mercyhurst)

[Numbers]

Re: John t whelan ranking simulator

exp(ki) is the KRACH rating, ie, ki for team i is a number which needs to exponentiated to be in Whelan's form. You don't need to exponentiate again. h, the home ice advantage, also needs to be exponentiated to make patman's formula pr(home win) = ca/(ca+b) correct. It's generally easier to estimate ki instead of Ki=exp(ki) because the constraint that Ki>0 is a pain. when working with ki, ie ln(Ki) the coefficient is unconstrained.

Note all of the Ks are still only determined up to a multiplicative constant, so you can define the absolute value of the KRACH ratings any way you want. Some common methods are to make the best team 100, to make the sum 1, or, in an understandable if somewhat idiosyncratic view at siouxsports.com, to make North Dakota = 100.

(Neither h nor theta can be scaled. They have to be takes as is.)

Thanks Go,

That's what I meant. I appreciate all the explaining.

[goblue78]

Re: John t whelan ranking simulator

exp(ki) is the KRACH rating, ie, ki for team i is a number which needs to exponentiated to be in Whelan's form. You don't need to exponentiate again. h, the home ice advantage, also needs to be exponentiated to make patman's formula pr(home win) = ca/(ca+b) correct. It's generally easier to estimate ki instead of Ki=exp(ki) because the constraint that Ki>0 is a pain. when working with ki, ie ln(Ki) the coefficient is unconstrained.

Note all of the Ks are still only determined up to a multiplicative constant, so you can define the absolute value of the KRACH ratings any way you want. Some common methods are to make the best team 100, to make the sum 1, or, in an understandable if somewhat idiosyncratic view at siouxsports.com, to make North Dakota = 100.

(Neither h nor theta can be scaled. They have to be takes as is.)

[FlagDUDE08]

Re: John t whelan ranking simulator

No, it doesn't affect autobids, just at-large bids. Wisconsin earned an at-large bid in 07-08 with a record of 15-16-7. They did end up beating Denver 6-2 in their first NCAA tournament game, but lost to North Dakota in overtime the next game. The "Wisconsin rule" was implemented the year after.

If that's still the case, New Hampshire would be the replacement. It does show that it's very early in the season, though.

[The Exiled One]

Re: John t whelan ranking simulator

Does autobids have any impact on that? Alabama Huntsville made the tournament a few years ago with a sub-500 record, but they had the CHA autobid. Also, I seem to remember about 6 or 7 years ago UVM had a team that almost didn't make the Hockey East playoffs but ended making some noise in the national tournament.

No, it doesn't affect autobids, just at-large bids. Wisconsin earned an at-large bid in 07-08 with a record of 15-16-7. They did end up beating Denver 6-2 in their first NCAA tournament game, but lost to North Dakota in overtime the next game. The "Wisconsin rule" was implemented the year after.

[FlagDUDE08]

Re: John t whelan ranking simulator

Wisconsin rule. That's the one prohibiting teams with less that .500 records from an at-large place in the tournament.

Does autobids have any impact on that? Alabama Huntsville made the tournament a few years ago with a sub-500 record, but they had the CHA autobid. Also, I seem to remember about 6 or 7 years ago UVM had a team that almost didn't make the Hockey East playoffs but ended making some noise in the national tournament.

[Numbers]

Re: John t whelan ranking simulator

The what rule?

Wisconsin rule. That's the one prohibiting teams with less that .500 records from an at-large place in the tournament.

[FlagDUDE08]

Re: John t whelan ranking simulator

Did they get rid of the Wisconsin rule? If not, North Dakota doesn't currently qualify for an at-large bid.

The what rule?

[Numbers]

Re: John t whelan ranking simulator

GoBlue,

I have another question about this KRACH with ties and home/away. The mathematical formulation uses exp(k1+h) for example. I assume that is because it makes the math come out - because you can take the logs and then it's addition rather than multiplication. However, it seems to me that the k(i) that comes out of that calculation is not the k(i) that comes out of Whelan's example of how to calculate KRACH here and on CHN.
It seems to me that the Ki all need to re-exponentiated to get the numbers that are equivalent to the KRACH ratings in Whelan's formulation.

Is that correct?

Thanks again.

[The Exiled One]

Re: John t whelan ranking simulator

And the tournament field:

North Dakota

Did they get rid of the Wisconsin rule? If not, North Dakota doesn't currently qualify for an at-large bid.

[Ralph Baer]

Re: John t whelan ranking simulator

And the tournament field:

Minnesota
St. Cloud State
Providence
Quinnipiac

Boston College
Michigan
Ferris State
Wisconsin

Miami
LSSU
Notre Dame
Bowling Green

Cornell
Clarkson
North Dakota
AHA Champ (37 - Air Force)

Interesting that currently each of the five non-AHA leagues has three representatives.

[FlagDUDE08]

Re: John t whelan ranking simulator

Apologies for not previously releasing, as time was spent with FlagDUDETTE. Here's what we have, as of games ending 20 November:

5.00 Minnesota
4.75 St. Cloud State
4.50 Quinnipiac
4.25 Providence
4.00 Boston College
3.75 Michigan
3.50 Ferris State
3.25 Wisconsin
3.00 LSSU
2.75 Miami
2.50 Notre Dame
2.25 Bowling Green
2.00 Cornell
1.75 Clarkson
1.50 Northern Michigan
1.25 North Dakota
1.00 New Hampshire
0.75 Minnesota State Mankato
0.50 Union
0.25 UMASS Lowell

And the tournament field:

Minnesota
St. Cloud State
Providence
Quinnipiac

Boston College
Michigan
Ferris State
Wisconsin

Miami
LSSU
Notre Dame
Bowling Green

Cornell
Clarkson
North Dakota
AHA Champ (37 - Air Force)

[FlagDUDE08]

Re: John t whelan ranking simulator

I "finished" updates to my interpretation of the RPI. As a background, my implementation is written in PHP and primarily serves to power my exhaustive PWR predictor, which I won't be firing up until championship weekend, though it could possibly do some monte carlo a couple weekends earlier. I'm hoping to develop it further to aid prognosticators in finding corner-cases and understanding how things can shake out. Again, it's only really useful for the last weekend of league championship games.

But, might as well get the key parts done early. I've also included fairly in-depth breakdowns of how the RPI is formed for each team -- let me know if you spot any mistakes or would like to see any other components in further detail.

http://pwr.reillyhamilton.com/pwr.html

It agrees with JimDahl's RPI for all teams except Mankato (and there's only a .0002 difference there, seems to have been rounding on one of our parts when determining negative wins). Haven't taken a close look at why it differs from FlagDude's, as I'm not sure what "stage" of the calculations are listed in the GUI, ie are OWP and OOWP before or after negative impact wins have been removed?

It's also doing PWR, but I haven't scrutinized that closely, so I'm not confident it's correct. It was accurate (compared to USCHO, CHN, and JimDahl/SiouxSports) last year, so I imagine it shouldn't be far off this year, as the only changes were removal of the TUC comparison and .5000 RPI qualifier. It's also November...

Currently not updating automatically (but current through today's games); I may implement a caching layer and/or cron-job in a couple days.





By the way, I find all the different ways that we all think to be fascinating. I'm completely lost by some of the "actual math" going on here; I think much more iteratively. I do volunteer to help a scraping / data acquisition effort if it would be helpful to the greater simulator cause, as I love the idea and I'm already working on a bunch of collegehockeystats.net scraping for RPI TV's titles and graphics package.

The stuff that is listed is after the games have been removed. Take a look in the command line (assuming you run it from there and not the exe) to determine which specific games have been removed. Do we still disagree there?