2005-2006 PWR - playoffs edition!

[Robb]

[quote Ken '70][quote jmh30]I noticed that after tonight's result, Clarkson's RPI is down to .5005.  This may have already been asked, but, if Cornell wins tomorrow night, will they drop below .5 and out of TUC status?  (At that point it'd be good for them to remain a TUC, because a 3-1-0 record against them helps Cornell's TUC record.)[/quote]

It is very, very tight and depends on outcomes of other games that effect Clarkson's Op and OpOp %s.  They need Niagara, Umass, SLU, MIami and Bemidji to almost all win tonight and then to have continued success.  It also will help Clarkson stay TUC if Cornell continues to win since Clarkson has 4 instances of Cornell in its Op column.[/quote]
This is one of those stupid PWR things where it might actually end up better for Cornell to beat Clarkson 2-1 in the series rather than sweeping, if the extra win for Clarkson keeps them as a TUC AND it improves Cornell's TUC record enough to flip a comparison.  Of course, we won't know that until the end of the conference tournaments when all the games are played and we can flop the result in the DIY script, but it's possible.

Stupid PWR.

[Jim Hyla]

[quote Ken '70]Cornell is in pretty good shape if they just win, which they've had trouble doing when needed from a PWR perspective.[/quote]I beg to differ, Cornell is in great shape if they just win.:-D

[jy3]

updated the first post with the new rankings. didnt have access til this AM to a computer

[nyc94]

It's worth noting that Niagara is a TUC with the 33-22-11 bonus but not with 3-2-1.  It's that close.

[Tub(a)]

Quick question...

Haven't been following this closely, but when/why did 33-22-11 replace 3-2-1 as the likely bonus?

[Chris \'03]

[quote jy3]updated the first post with the new rankings. didnt have access til this AM to a computer [/quote]

You spelled Cornell wrong...

[Ack]

I believe the song goes, "C-O-R-N-E-DOUBLE L...."

[DeltaOne81]

[quote Tub(a)]Quick question...

Haven't been following this closely, but when/why did 33-22-11 replace 3-2-1 as the likely bonus?[/quote]

After each selection show, people try to 'reverse engineer' what the bonus was based on who ended up where. Apparently some people determined that what they were using was more like 33-22-11 than 3-2-1.

[Ken '70]

[quote Robb]This is one of those stupid PWR things where it might actually end up better for Cornell to beat Clarkson 2-1 in the series rather than sweeping, if the extra win for Clarkson keeps them as a TUC AND it improves Cornell's TUC record enough to flip a comparison.  Of course, we won't know that until the end of the conference tournaments when all the games are played and we can flop the result in the DIY script, but it's possible.

Stupid PWR.[/quote]

The worst scenario for Cornell is the one described above.  Their TUC record is 10-4-1, or .7000.  Beating Clarkson 2-1 makes their TUC 11-5-1, or .6765.  If Cornell just wins tonight and Clarkson drops for good, then their TUC is 9-3-1, or .7308.  In the case of losing to Clarkson tonight but winning tomorrow, not only would it hurt their TUC % but they'd take a RPI hit, which is far more important than TUC right now.

PWR is a pretty good way of sorting teams except that I'd make the tie-breaker H2H, when available, before RPI.  It's always better to have an actual result on the ice decide things rather than a mathematical abstraction like RPI or KRACH.

[Josh '99]

[quote TCHL8842]Well I like the spot we are in now, 2 weekends in Albany wont be that bad[/quote]While I might still prefer going to Lake Placid over going to Albany, I MUCH prefer Albany to Grand Forks.

[DeltaOne81]

[quote Ken '70]
PWR is a pretty good way of sorting teams except that I'd make the tie-breaker H2H, when available, before RPI.[/quote]

I was actually thinking this last night. It would have eliminated that DU-CC controversy last year anyway.

[jtwcornell91]

[quote Ken '70]PWR is a pretty good way of sorting teams except that I'd make the tie-breaker H2H, when available, before RPI.  It's always better to have an actual result on the ice decide things rather than a mathematical abstraction like RPI or KRACH.[/quote]

AAAARGH! :-( RPI, KRACH, and the arbitrarily chosen criteria of PWR all depend only on the "actual result on the ice".  Don't perpetuate the ridiculous fallacy that rating systems which deal with the unavoidable complications of an unbalanced schedule are somehow not determined by game results.

[abmarks]

[quote jtwcornell91][quote Ken '70]PWR is a pretty good way of sorting teams except that I'd make the tie-breaker H2H, when available, before RPI.  It's always better to have an actual result on the ice decide things rather than a mathematical abstraction like RPI or KRACH.[/quote]

AAAARGH! :-( RPI, KRACH, and the arbitrarily chosen criteria of PWR all depend only on the "actual result on the ice".  Don't perpetuate the ridiculous fallacy that rating systems which deal with the unavoidable complications of an unbalanced schedule are somehow not determined by game results.[/quote]

I think the main point Ken was trying to make was that Head to Head was a preferable tiebreaker vs. how teams did against the schedule as a whole.

[Ken '70]

[quote jtwcornell91][quote Ken '70]PWR is a pretty good way of sorting teams except that I'd make the tie-breaker H2H, when available, before RPI.  It's always better to have an actual result on the ice decide things rather than a mathematical abstraction like RPI or KRACH.[/quote]

AAAARGH! :-( RPI, KRACH, and the arbitrarily chosen criteria of PWR all depend only on the "actual result on the ice".  Don't perpetuate the ridiculous fallacy that rating systems which deal with the unavoidable complications of an unbalanced schedule are somehow not determined by game results.[/quote]

The result of how two teams actually did against each other as a means of breaking ties in an otherwise more complex system can hardly be called arbitrary.  RPI and KRACH are mathematical abstractions drawn from real results, but not the results themselves.

But this is a religious argument for you, so I'll bow out at this point.  Reasoning against an Islamic jihadist or KRACH devotee is equally futile.

[Robb]

[quote Ken '70]
The worst scenario for Cornell is the one described above.  Their TUC record is 10-4-1, or .7000.  Beating Clarkson 2-1 makes their TUC 11-5-1, or .6765.  If Cornell just wins tonight and Clarkson drops for good, then their TUC is 9-3-1, or .7308.  In the case of losing to Clarkson tonight but winning tomorrow, not only would it hurt their TUC % but they'd take a RPI hit, which is far more important than TUC right now.[/quote]
I think your point is correct, but we're already 2-1 vs. Clarkson, so if they drop out of TUC, then wouldn't our TUC record go to 8-3-1 (.7083) instead of .7308?

Best case scenario is that we sweep them AND they remain a TUC, which would get us to 11-4-1 (.7188).  Of course, our "current" 10-4-1 is dependent upon Niagara "remaining" a TUC, which they are only with at least a 33/22/11 bonus (exactly at .500 RPI right now) or winning their tourney. If they tack on two more wins but lose their conference championship, then perhaps a 3/2/1 bonus will still get them to TUC status, but it'll be close.

Go Purple Eagles!!!!