Looking Ahead

[LGR14]

What are the PWR implications of the Harvard v. BU game?

A Harvard win helps a little bit but probably not enough to move to 12th. A BU win knocks CU back to 14th.

Are you certain of that? It's the BC/Providence comparison that BC upsets to tie us.  We beat them with a decent chunk of RPI edge.  Not sure that'd flip.

Harvard win is good for a .5496 RPI. A BU win would drop us to a .5482 RPI. But you're right, a BU win doesn't knock us back to 14th anymore after the loss from earlier. For what it's worth, BC would sit at .5432 with a BU victory.

[abmarks]

Rk   Team   PCWs   RPI   Rk   QWB-†   W-L-T   Win %   Wgtd Win % - ‡
1   Minnesota-Duluth   59   .6078   1   .0137   18-5-5   .7321   .7133
2   Denver   58   .5995   2   .0107   20-6-4   .7333   .7260
3   Harvard   57   .5816   3   .0055   18-5-2   .7600   .7797
4   Minnesota   56   .5691   4   .0044   18-8-2   .6786   .6889
5   Boston University   55   .5673   5   .0071   19-9-2   .6667   .6783
6   Penn State   54   .5654   6   .0009   18-6-2   .7308   .7190
7   Western Michigan   53   .5653   7   .0061   16-8-4   .6429   .6439
8   Providence   51   .5581   8   .0055   18-8-4   .6667   .6599
9   Mass.-Lowell   51   .5575   9   .0039   19-9-3   .6613   .6667
10   North Dakota   50   .5550   10   .0074   15-11-3   .5690   .5479
11   Union   49   .5533   11   .0025   20-8-2   .7000   .7152
12   St. Cloud State   48   .5507   12   .0094   15-14-1   .5167   .5200
13   Cornell   47   .5496   13   .0014   16-6-3   .7000   .6953
14   Boston College   47   .5422   14   .0037   18-12-2   .5938   .6031
15   Ohio State   45   .5421   15   .0085   14-8-6   .6071   .6088
16   Vermont   44   .5408   16   .0033   17-10-3   .6167   .6069
17   Nebraska-Omaha   43   .5400   17   .0034   15-12-5   .5469   .5422
18   Notre Dame   42   .5397   18   .0046   17-9-4   .6333   .6096
19   St. Lawrence   41   .5382   19   .0037   15-9-6   .6000   .6143
20   Air Force   40   .5375   20   .0012   19-8-5   .6719   .6844

[Trotsky]

Code Select Expand


[u]Rk Team       PCWs RPI Rk QWB   W- L-T  Win %  Wgtd Win %[/u]
 9 Mass.-Lowell 51 .5575 9 .0039 19- 9-3  .6613    .6667
10 North Dakota 50 .5550 10 .0074 15-11-3  .5690    .5479
11 Union        49 .5533 11 .0025 20- 8-2  .7000     .7152
12 St. Cloud State 48 .5507 12 .0094 15-14-1  .5167    .5200
13 Cornell        47 .5496 13 .0014 16- 6-3  .7000    .6953
14 Boston College 47 .5422 14 .0037 18-12-2  .5938    .6031


Nearly as close to 9 as to 14.

[abmarks]

Nearly as close to 9 as to 14.

How so? I Don't see it

[andyw2100]

Nearly as close to 9 as to 14.

How so? I Don't see it

.0074 above 14.
.0079 below 9.

[ursusminor]

Nearly as close to 9 as to 14.

How so? I Don't see it

.0074 above 14.
.0079 below 9.

It actually could be slightly closer (or further apart) because when you subtract numbers with four-figure accuracy, the difference calculated could be off by one in the fourth place.

[Jim Hyla]

Nearly as close to 9 as to 14.

How so? I Don't see it

.0074 above 14.
.0079 below 9.

It actually could be slightly closer (or further apart) because when you subtract numbers with four-figure accuracy, the difference calculated could be off by one in the fourth place.

picky, picky, picky::burnout::

[Swampy]

Nearly as close to 9 as to 14.

How so? I Don't see it

.0074 above 14.
.0079 below 9.

It actually could be slightly closer (or further apart) because when you subtract numbers with four-figure accuracy, the difference calculated could be off by one in the fourth place.

picky, picky, picky::burnout::

But the numbers are only measured to 4-digits precision. So if we round off to 3 digits, using standard conventions, we get Cornell with .550, UML with .558, and BC with .542. So we're .008 away from both #9 & #14.

[jkahn]

Code Select Expand


[u]Rk Team       PCWs RPI Rk QWB   W- L-T  Win %  Wgtd Win %[/u]
 9 Mass.-Lowell 51 .5575 9 .0039 19- 9-3  .6613    .6667
10 North Dakota 50 .5550 10 .0074 15-11-3  .5690    .5479
11 Union        49 .5533 11 .0025 20- 8-2  .7000     .7152
12 St. Cloud State 48 .5507 12 .0094 15-14-1  .5167    .5200
13 Cornell        47 .5496 13 .0014 16- 6-3  .7000    .6953
14 Boston College 47 .5422 14 .0037 18-12-2  .5938    .6031


Nearly as close to 9 as to 14.

It should be noted that we are in a more vulnerable position than the teams around us.  25% of RPI is based upon a team's winning percentage, and since we're at .700, a loss knocks more points off our percentage than it does for the teams with lesser win percentages (and more than Union, also at .700, since Union's played more games).  We go 5 and 3 the rest of the way and our RPI goes down, while BC's goes up with the same 5 and 3 for them.  I'm making the assumption here that the other RPI factors don't change all that much the rest of the way, but the most significant shifts at this point in the season will be come from teams' own wins and losses.  I'm also ignoring the home/road factors for simplification though I'm aware of them.

[wakester2468]

With two weeks remaining in the regular season and all teams having four games left, and with the top four teams all within two point in the ECAC standings, it seems quite possible that at the end, some teams will be tied  with the same total points. I'm looking for some clarity on tie breaking procedures. Obviously head to head is the first
deciding factor and I suspect total wins might be next ( one win being better than two ties} but after that I'm pretty clueless on the protocol. Anyone know the list in terms of importance right up to a flip of a coin?

[marty]

With two weeks remaining in the regular season and all teams having four games left, and with the top four teams all within two point in the ECAC standings, it seems quite possible that at the end, some teams will be tied  with the same total points. I'm looking for some clarity on tie breaking procedures. Obviously head to head is the first
deciding factor and I suspect total wins might be next ( one win being better than two ties} but after that I'm pretty clueless on the protocol. Anyone know the list in terms of importance right up to a flip of a coin?

::bugeye::Look!::bugeye::

[Trotsky]

Presumably it is still possible to have a 4-way tie for first.

[Tom Lento]

Presumably it is still possible to have a 4-way tie for first.

Yes, there are several paths, although of course at least one tie is required. Here's one that involves the current top 4 winning all but one game against teams 5-12 (the one exception being a Union draw, potentially vs. QC):

Cornell and SLU tie, Cornell beats Clarkson, Union, and RPI.
SLU beats Harvard, Brown, and Colgate.
Harvard beats Clarkson, Yale, and Brown
Union goes 2-0-1 against Colgate, QC, and Princeton

Here's one with just one tie, but it requires either Brown or Colgate to beat SLU and Cornell plus one other (QC most likely) to beat Union:

SLU beats Cornell and Harvard and splits with Brown and Colgate
Cornell beats Union, Clarkson, and RPI
Harvard goes 2-0-1 against Clarkson, Yale, and Brown
Union goes 2-1-0 against Colgate, QC, and Princeton

And with that, I have wasted enough time at work.

[Trotsky]

In a game to potentially watch tonight, PWR 8 Lowell is at BC.

[Jim Hyla]

In a game to potentially watch tonight, PWR 8 Lowell is at BC.

Also Miami at UMD.