TBRW Predictions: 2002

This is the overview.  For the details, see: Never Apologize, Never Explain.

Last March, they were so close.  In their fourteenth ECAC championship game, Cornell entered the third period trailing by two goals, and proceeded to outshoot SLU 17-5 -- yet still lost by two.

But the 2000-01 season was a great success, featuring a return to the Final Four, to say nothing of the Final Itself, and a solid fourth place regular season campaign.  A few wonderful weeks in January even saw the Big Red at the top of the ECAC and in the Top 10 in the nation.  Not bad for a team which barely managed two goals per game.

Now, after an insufferably long summer, Our Heroes enter the fray again, with the usual suspects (Clarkson, Harvard, RPI, SLU) and a rising star (Dartmouth) as the early favorites to challenge them for league honors.

Others pre-season polls have put Cornell anywhere from first to sixth.  To these we add our own rating, the TBRW? Predictions!

The Bottom Line

1   Clarkson
2   RPI
3   St. Lawrence
4   Cornell
5   Harvard
6   Dartmouth
7   Princeton
8   Yale
9   Colgate
10   Union
11   Vermont
12   Brown

What's Out (and In) in '02

We make changes each year to our predictive indicators.  Either we're learning, or we're always re-fighting the last war.

This year, the watchwords are "less is more."  What's Out: the attempt (albeit tongue-in-cheek) to peg a team's exact winning percentage (one which met with results varying all the way from amazingly prophetic to embarrassingly problematic in previous years).  What's In: a simple concept: how things will go is a combination of how things have gone and how things have changed.

So, we dispense with the higher math, with normalizing modifiers, and with complex weighting.  Instead we simply rank teams as better (+1), same, or worse (-1) than the league average for each modifier.

We have also diminish the regression to the mean that was obtained by averaging in the teams' historical performance.  From 50% of the total package, this has been pushed back to 25%, and the historical interval has shifted, from the whole history of the league to just the last ten seasons.

Another change is the redefinition of the measures of returning strength.  Only "Team Leaders" are now considered.  A team's leaders are last season's top nine forwards in points, top four defensemen in points, and top two goaltenders in wins.  Only these fifteen performers are considered from each squad.  We include the team-by-team breakdown of these modifiers.

Early Departures

This year, we know of just one early departure among the Leading Forwards, Defensemen, or Goalies:

In all other cases, every player is assumed to be returning unless he was listed as "Sr" or "Gr" on last year's roster.