So, here's how we do it. Follow the links for detailed computations.
Regular season performance in previous season. |
|
Post-season performance in previous season. |
Factored into these base values is a function of the following modifiers:
Standing in total conference record in the three previous seasons. |
|
Second-half improvement, previous regular season. |
|
Returning % of Leading Forwards points. |
|
Returning % of Leading Defensemen points. |
|
Returning % of Goaltender wins. |
Pred is derived by adding all of the base values and modifiers together:
FPred, the basis for our final predictor, is derived by regressing Pred 25% back towards the team's aggregate conference standing over the previous ten seasons:Pred = (RS + PS) + (Prv3 + 1/2 + RetF + RetD + RetG)
FPred = 3 * Pred + Prv10
Sum of all base values and modifiers. |
|
Standing in total conference record in the ten previous seasons. |
|
Final predictor and basis for the TBRW? Prediction. |
Our final prediction, TBRW, is the rank order of FPred.
Predicted ranking. |
The modifiers are computed for the present year, below.
RS = (12 - Standing)
i.e., a reversal (returning the largest number for first place) of the prior season's ECAC regular season standings.
|
|
Standing | RS |
|
Brown |
12 | 0 |
| Clarkson | 8 | 4 |
| Colgate | 2 | 10 |
| Cornell | 3 | 9 |
| Dartmouth | 1 | 11 |
| Harvard | 4 | 8 |
| Princeton | 9 | 3 |
| Quinnipiac | 10 | 2 |
| RPI | 7 | 5 |
| St. Lawrence | 5 | 7 |
| Union | 6 | 6 |
| Yale | 11 | 1 |
The number of post-season upset series wins (+1) and losses (-1) in the previous season's ECAC playoffs. The consolation game is ignored.
|
|
PS |
|
Brown |
0 |
| Clarkson | 0 |
| Colgate | -1 |
| Cornell | 0 |
| Dartmouth | -1 |
| Harvard | +2 |
| Princeton | 0 |
| Quinnipiac | +1 |
| RPI | -1 |
| St. Lawrence | 0 |
| Union | -1 |
| Yale | +1 |
The third (+1 top 33%, 0 middle 33%, -1 bottom 33%) in conference winning percentage (Pct) for the previous three years, i.e., those seasons in which the current returning classes have played.
|
|
Pct | Rank | Prv3 |
|
Brown |
.447 | 7 | 0 |
| Clarkson | .409 | 9 | -1 |
| Colgate | .689 | 2 | +1 |
| Cornell | .727 | 1 | +1 |
| Dartmouth | .644 | 3 | +1 |
| Harvard | .614 | 4 | +1 |
| Princeton | .326 | 11 | -1 |
| Quinnipiac | .386 | 10 | -1 |
| RPI | .470 | 5 | 0 |
| St. Lawrence | .462 | 6 | 0 |
| Union | .439 | 8 | 0 |
| Yale | .311 | 12 | -1 |
The third (+1 top 33%, 0 middle 33%, -1 bottom 33%) in improvement in points gained in the second half (games 12-22) over the first half (games 1-11) of the previous ECAC regular season.
|
|
Pts Total | Pts 12-22 | Pts 1-11 | Pts Diff | 1/2 |
|
Brown |
11 | 5 | 6 | -1 | 0 |
| Clarkson | 20 | 9 | 11 | -1 | 0 |
| Colgate | 30 | 13 | 17 | -4 | -1 |
| Cornell | 29 | 14 | 15 | -1 | 0 |
| Dartmouth | 30 | 18 | 12 | +6 | +1 |
| Harvard | 27 | 13 | 14 | -1 | 0 |
| Princeton | 17 | 13 | 4 | +9 | +1 |
| Quinnipiac | 17 | 11 | 6 | +5 | +1 |
| RPI | 22 | 9 | 13 | -4 | -1 |
| St. Lawrence | 25 | 9 | 16 | -7 | -1 |
| Union | 22 | 12 | 10 | +2 | +1 |
| Yale | 14 | 5 | 9 | -4 | -1 |
The third (+1 top 33%, 0 middle 33%, -1 bottom 33%) in returning % of Leading Forwards points. Team-by-team breakdown here.
|
|
RetF% | RetF |
|
Brown |
1.00 | +1 |
| Clarkson | .88 | +1 |
| Colgate | .59 | -1 |
| Cornell | .55 | -1 |
| Dartmouth | .60 | 0 |
| Harvard | .77 | 0 |
| Princeton | .86 | +1 |
| Quinnipiac | .89 | +1 |
| RPI | .73 | 0 |
| St. Lawrence | .47 | -1 |
| Union | .69 | 0 |
| Yale | .52 | -1 |
The third (+1 top 33%, 0 middle 33%, -1 bottom 33%) in returning % of Leading Defensemen points. Team-by-team breakdown here.
|
|
RetD% | RetD |
|
Brown |
.85 | +1 |
| Clarkson | .83 | 0 |
| Colgate | .87 | +1 |
| Cornell | .17 | -1 |
| Dartmouth | .68 | 0 |
| Harvard | .59 | -1 |
| Princeton | .38 | -1 |
| Quinnipiac | 1.00 | +1 |
| RPI | .16 | -1 |
| St. Lawrence | .63 | 0 |
| Union | 1.00 | +1 |
| Yale | .77 | 0 |
The half (roughly) (+1 top, -1 bottom) in returning % of Leading Goaltenders wins. Team-by-team breakdown here.
|
|
RetG% | RetG |
|
Brown |
.05 | -1 |
| Clarkson | 1.00 | +1 |
| Colgate | 1.00 | +1 |
| Cornell | .00 | -1 |
| Dartmouth | 1.00 | +1 |
| Harvard | .05 | -1 |
| Princeton | .30 | -1 |
| Quinnipiac | .95 | +1 |
| RPI | 1.00 | +1 |
| St. Lawrence | .95 | +1 |
| Union | .00 | -1 |
| Yale | .90 | +1 |
Pred = (RS + PS) + (Prv3 + 1/2 + RetF + RetD + RetG)
|
|
RS | PS | Prv3 | 1/2 | RetF | RetD | RetG | Pred |
|
Brown |
0 | 0 | 0 | 0 | +1 | +1 | -1 | 1 |
| Clarkson | 4 | 0 | -1 | 0 | +1 | 0 | +1 | 5 |
| Colgate | 10 | -1 | +1 | -1 | -1 | +1 | +1 | 10 |
| Cornell | 9 | 0 | +1 | 0 | -1 | -1 | -1 | 7 |
| Dartmouth | 11 | -1 | +1 | +1 | 0 | 0 | +1 | 13 |
| Harvard | 8 | +2 | +1 | 0 | 0 | -1 | -1 | 9 |
| Princeton | 3 | 0 | -1 | +1 | +1 | -1 | -1 | 2 |
| Quinnipiac | 2 | +1 | -1 | +1 | +1 | +1 | +1 | 6 |
| RPI | 5 | -1 | 0 | -1 | 0 | -1 | +1 | 3 |
| St. Lawrence | 7 | 0 | 0 | -1 | -1 | 0 | +1 | 6 |
| Union | 6 | -1 | 0 | +1 | 0 | +1 | -1 | 6 |
| Yale | 1 | +1 | -1 | -1 | -1 | 0 | +1 | 0 |
Prv10 is a reversal (returning the largest number for first place) of the regular season Standing of the previous ten seasons, 1994-95 through 2003-04. Prv10 = (12 - Standing).
|
|
Pct | Standing | Prv10 |
|
Brown |
.393 | 10 | 2 |
| Clarkson | .594 | 2 | 10 |
| Colgate | .576 | 3 | 9 |
| Cornell | .658 | 1 | 11 |
| Dartmouth | .502 | 7 | 5 |
| Harvard | .557 | 4 | 8 |
| Princeton | .422 | 9 | 3 |
| Quinnipiac | .386 | 12 | 0 |
| RPI | .511 | 6 | 6 |
| St. Lawrence | .518 | 5 | 7 |
| Union | .388 | 11 | 1 |
| Yale | .454 | 8 | 4 |
Finally, TBRW is the rank order of FPred, and is the basis for our predicted ranking.
|
|
Pred | 3*Pred | Prv10 | FPred | TBRW |
|
Brown |
1 | 3 | 2 | 5 | 11 |
| Clarkson | 5 | 15 | 10 | 25 | 5(t) |
| Colgate | 10 | 30 | 9 | 39 | 2 |
| Cornell | 7 | 21 | 11 | 32 | 4 |
| Dartmouth | 13 | 39 | 5 | 43 | 1 |
| Harvard | 9 | 27 | 8 | 35 | 3 |
| Princeton | 2 | 6 | 3 | 9 | 10 |
| Quinnipiac | 6 | 18 | 0 | 18 | 8 |
| RPI | 3 | 9 | 6 | 15 | 9 |
| St. Lawrence | 6 | 18 | 7 | 25 | 5(t) |
| Union | 6 | 18 | 1 | 19 | 7 |
| Yale | 0 | 0 | 4 | 4 | 12 |