11/9 Union

[Trotsky]

In other words: past performance is not a reliable indicator of future results.

The expression is "past performance is no guarantee..."  Inductive reasoning is not proof, but it's still highly useful.

[Rosey]

In other words: past performance is not a reliable indicator of future results.

The expression is "past performance is no guarantee..."  Inductive reasoning is not proof, but it's still highly useful.

While year-to-year returns in coaching are not independent variables, there are enough independent sub-variables that simply extrapolating past performance to future results is unreliable. That's the point of the aphorism: not that presently successful investments were uniformly distributed over the performance range in the past, but that not all successful investments continue to be successful in the future. Hence, "indicator" yet also "unreliable".

Yours is a fallacy that kills otherwise prudent investors all the time. Unfortunately for Cornell, there's no way to diversify their hockey coach holdings, so a different strategy is required.

[Trotsky]

Yours is a fallacy that kills otherwise prudent investors all the time.

I don't think either of us is saying anything strong enough to warrant a term like "fallacy."  We are arguing differences in degree only -- when has a trend gone on long enough to constitute a preponderance of evidence, and when do past results "time out" in their relevance to predicting future results?  I'm arguing there is still too much noise to call any patterns we think we're seeing clear evidence.  If our coach was Bob Gaudet and we had one great season and then played great for a few games in the next season I would be arguing the same thing: that would be too early to think he'd magically turned the corner and become competent.

The longer we play like crap the more your argument gains credence, but a couple good weekends and this may be just another in the thousand and one false highs and lows in a season.

[Rosey]

Yours is a fallacy that kills otherwise prudent investors all the time.

I don't think either of us is saying anything strong enough to warrant a term like "fallacy."  We are arguing differences in degree only -- when has a trend gone on long enough to constitute a preponderance of evidence, and when do past results "time out" in their relevance to predicting future results?  I'm arguing there is still too much noise to call any patterns we think we're seeing clear evidence.  If our coach was Bob Gaudet and we had one great season and then played great for a few games in the next season I would be arguing the same thing: that would be too early to think he'd magically turned the corner and become competent.

The longer we play like crap the more your argument gains credence, but a couple good weekends and this may be just another in the thousand and one false highs and lows in a season.

What, exactly, is my argument? I'm not calling for Schafer's head. I am, however, not being pollyanna-ish about his prospects for returning this team to the years of plenty when Cornell seemed to shit ECAC titles and fairly consistently roll over the rest of the league even in the years when they got knocked out of the playoffs by a hot Harvard team or some other one-off bad game. Those were the years when I expected Cornell to win nearly every league game, and when they didn't that disappointed me; now, I expect them to lose a lot of those games and am pleasantly surprised when they win. I vastly prefer the former, even when it means 2003's end.

I like the Parcells quote: it neatly sums up my opinion of the entire matter. How good or how bad I think you might be is irrelevant: you are as good as your record indicates. So, just win.

I don't give Schafer the benefit of the doubt. Maybe that's the key difference between the two schools of thought here.

[BearLover]

Yours is a fallacy that kills otherwise prudent investors all the time.

I don't think either of us is saying anything strong enough to warrant a term like "fallacy."  We are arguing differences in degree only -- when has a trend gone on long enough to constitute a preponderance of evidence, and when do past results "time out" in their relevance to predicting future results?  I'm arguing there is still too much noise to call any patterns we think we're seeing clear evidence.  If our coach was Bob Gaudet and we had one great season and then played great for a few games in the next season I would be arguing the same thing: that would be too early to think he'd magically turned the corner and become competent.

The longer we play like crap the more your argument gains credence, but a couple good weekends and this may be just another in the thousand and one false highs and lows in a season.

What, exactly, is my argument? I'm not calling for Schafer's head. I am, however, not being pollyanna-ish about his prospects for returning this team to the years of plenty when Cornell seemed to shit ECAC titles and fairly consistently roll over the rest of the league even in the years when they got knocked out of the playoffs by a hot Harvard team or some other one-off bad game. Those were the years when I expected Cornell to win nearly every league game, and when they didn't that disappointed me; now, I expect them to lose a lot of those games and am pleasantly surprised when they win. I vastly prefer the former, even when it means 2003's end.

I like the Parcells quote: it neatly sums up my opinion of the entire matter. How good or how bad I think you might be is irrelevant: you are as good as your record indicates. So, just win.

I don't give Schafer the benefit of the doubt. Maybe that's the key difference between the two schools of thought here.

You prefer 2003?  Then why the hell were you happy when Q and Yale made the finals last year?  That was the biggest slap in the face this program has ever gotten, and it signifies the other schools have passed us by.  More than that, though, it just makes it even tougher to return to our past dominance.  I don't know how this year is going to end up, but I think anyone here would be thrilled with a top 4 finish--and that in itself speaks wonders.

[Chris '03]

You prefer 2003?  Then why the hell were you happy when Q and Yale made the finals last year?  That was the biggest slap in the face this program has ever gotten, and it signifies the other schools have passed us by.  More than that, though, it just makes it even tougher to return to our past dominance.  I don't know how this year is going to end up, but I think anyone here would be thrilled with a top 4 finish--and that in itself speaks wonders.

I'd rather Cornell not dominate a conference that wins national titles than dominate a conference that doesn't. How many national champions dominated their conferences in the past decade or so en route to a national title? Rising tide lifts all boats, etc., etc. The ECAC (including Cornell) has a better shot at winning (national) titles when the league is very good top to bottom and 4-5 schools play in the national tournament both statistically and from the fact they've played a higher level of competition all year long.

I don't think folks on Comm Ave are pouting that Lowell "has passed them by" by being the Frozen Four last year.

We get it that you want, and expect, Cornell to dominate the ECAC and play for national titles every year. That's not realistic unless you wish the rest of the league would go back to being terrible so that Cornell can win the league's one bid every year and hope for the best. Basically no team in the country meets your criteria for success of league dominance and national contender year in and year out (BC is probably closest with 4 out of 10 HEA #1 seeds and 6 of ten HEA titles and 3 NCAA titles in ten years).

[Rosey]

You prefer 2003?  Then why the hell were you happy when Q and Yale made the finals last year?

Isn't the answer obvious? Because I want Cornell to dominate in a nationally-competitive league. (And a pony.)

My argument from March hasn't changed: Cornell can't be consistently competitive at the national level when it doesn't play nationally-competitive teams all the time. It's a necessary—but insufficient—condition.

[Trotsky]

I don't give Schafer the benefit of the doubt. Maybe that's the key difference between the two schools of thought here.

It is.

But happily we agree on the solution in either case.  Just win.

[Trotsky]

Number of games vs top 20 / 10 / 5 teams (rank on date played); and record in games vs top 10:

Code Select Expand


2002  6 3 3  1-2-0
2003 11 3 2  1-2-0
2004 10 0 0  0-0-0
2005 13 4 1  2-2-0
2006 11 4 1  2-2-0
2007  6 4 2  2-1-1
2008 10 1 0  0-0-1
2009  8 5 0  3-2-0
2010 12 4 3  1-3-0
2011 13 8 3  2-6-0
2012 11 6 1  4-1-1
2013 17 5 4  1-4-0


[Josh '99]

Number of games vs top 20 / 10 / 5 teams (rank on date played):

Code Select Expand


2002  6 [b]3 3[/b]
2003 11 [b]3 2[/b]
2004 10 0 0
2005 13 [b]4 1[/b]
2006 11 [b]4 1[/b]
2007  6 [b]4 2[/b]
2008 10 1 0
2009  8 5 0
2010 12 4 3
2011 12 7 3
2012 11 6 1
2013 17 5 4


And almost none of these (games against top 10 from 2001-02 through 2006-07) were conference games; it was largely Schafer scheduling tough nonconference opponents to get the team ready to compete in the NCAA tournament because there wasn't, by and large, a whole lot of strength in the conference beyond one-and-done Harvard.  Contrast that with the period from 2009-10 on where there are significantly more games against stronger teams, where many of those games have been tough Yale/Quinnipiac/Union/RPI teams (with 2013 skewed by the three-game playoff series against #1 Quinnipiac).  Chris is right when he says a rising tide lifts all boats.  When the WCHA was winning the national championship almost every year from 2000-06, it wasn't because any one team was dominating the conference, it was because the whole conference was so deep that playing that conference schedule prepared you for the postseason.  We may see fewer Whitelaw Trophies, but Cornell has a better shot at winning a national championship in that kind of a conference than in the ECAC we saw from, say, 1992 to 2008.

[BearLover]

You prefer 2003?  Then why the hell were you happy when Q and Yale made the finals last year?

Isn't the answer obvious? Because I want Cornell to dominate in a nationally-competitive league. (And a pony.)

My argument from March hasn't changed: Cornell can't be consistently competitive at the national level when it doesn't play nationally-competitive teams all the time. It's a necessary—but insufficient—condition.

What are you talking about?  Cornell was ~.500 in the NCAAs when the ECAC sucked.  Now they can't make the NCAAs, because the conference doesn't suck.

[Rosey]

You prefer 2003?  Then why the hell were you happy when Q and Yale made the finals last year?

Isn't the answer obvious? Because I want Cornell to dominate in a nationally-competitive league. (And a pony.)

My argument from March hasn't changed: Cornell can't be consistently competitive at the national level when it doesn't play nationally-competitive teams all the time. It's a necessary—but insufficient—condition.

What are you talking about?  Cornell was ~.500 in the NCAAs when the ECAC sucked.  Now they can't make the NCAAs, because the conference doesn't suck.

Blaming other teams for your own lack of success is loser-talk.

[Swampy]

Following some of the inductive logic on this thread, Cornell is unbeaten this season when Bardreau plays a full game. Does anyone know his situation for this coming weekend?

[BearLover]

You prefer 2003?  Then why the hell were you happy when Q and Yale made the finals last year?

Isn't the answer obvious? Because I want Cornell to dominate in a nationally-competitive league. (And a pony.)

My argument from March hasn't changed: Cornell can't be consistently competitive at the national level when it doesn't play nationally-competitive teams all the time. It's a necessary—but insufficient—condition.

What are you talking about?  Cornell was ~.500 in the NCAAs when the ECAC sucked.  Now they can't make the NCAAs, because the conference doesn't suck.

Blaming other teams for your own lack of success is loser-talk.

Oh, I didn't realize in hockey you don't actually play against other teams but just shoot pucks on unguarded nets.

[Rosey]

You prefer 2003?  Then why the hell were you happy when Q and Yale made the finals last year?

Isn't the answer obvious? Because I want Cornell to dominate in a nationally-competitive league. (And a pony.)

My argument from March hasn't changed: Cornell can't be consistently competitive at the national level when it doesn't play nationally-competitive teams all the time. It's a necessary—but insufficient—condition.

What are you talking about?  Cornell was ~.500 in the NCAAs when the ECAC sucked.  Now they can't make the NCAAs, because the conference doesn't suck.

Blaming other teams for your own lack of success is loser-talk.

Oh, I didn't realize in hockey you don't actually play against other teams but just shoot pucks on unguarded nets.

::rolleyes::