Never Apologize, Never Explain 2015-16

The predictions use these metrics, described below:

The Metrics

RS

RS is each team's points in the prior ECAC regular season:

	RS
Brown	13
Clarkson	19
Colgate	26
Cornell	22
Dartmouth	26
Harvard	25
Princeton	6
Quinnipiac	35
RPI	18
SLU	29
Union	17
Yale	28

PS

PS is the number of upset (lower seed) advances (+1) or eliminations (-1) in the previous season's ECAC tournament.  The consolation game is ignored.

The upsets in the last season's ECAC tournament were:

• + Union upset Cornell in First Round
• + RPI upsets Clarkson in First Round
• + Harvard upset Yale in QF
• + Harvard upset Quinnipiac in SF
• + Colgate upset St. Lawrence in SF
• + Harvard upset Colgate in F

	PS
Brown	0
Clarkson	-1
Colgate	0
Cornell	-1
Dartmouth	0
Harvard	+3
Princeton	0
Quinnipiac	-1
RPI	+1
SLU	-1
Union	+1
Yale	-1

Imp

Imp is the difference in points gained in the second half (games 12-22) minus the first half (games 1-11) of the prior ECAC regular season, divided by 2 (and rounded down).  A positive Imp indicates the team had a better second half:

	Pts	2nd ½	1st ½	Diff	Imp
Brown	13	11	2	+9	+4
Clarkson	19	7	12	-5	-2
Colgate	26	12	14	-2	-1
Cornell	22	10	12	-2	-1
Dartmouth	26	18	8	+10	+5
Harvard	25	9	16	-7	-3
Princeton	6	4	2	+2	+1
Quinnipiac	35	17	18	-1	0
RPI	18	8	10	-2	-1
SLU	29	15	14	+1	0
Union	17	8	9	-1	0
Yale	28	15	13	+2	+1

WRet

WRet captures how much of last year's players' contribution returns (or has been lost to graduation or early departure).

First the returning strength of three subcomponents is calculated for the team's "significant" players, as defined here:

• RetF: returning % of points of top 6 scoring forwards
• RetD: returning % of points of top 4 scoring defensemen
• RetG: returning % of minutes of all goalies

We then weight these numbers by their position's contribution to the team at full strength (3 forwards, 2 defensemen, and 1 goaltender), and multiply the result by 3 to approximate the dispersion from prior Returning Strength metrics:

WRet = (3*RetF + 2*RetD + RetG)/6

The team breakdowns of  Returning Players are shown below.

All of last year's significant seniors are assumed non-returning, unless otherwise noted.  The seniors from last season returning are:

• (None)

The following significant players with eligibility remaining are not returning:

• + Kyle Baun, Colgate, Jr, F #2

  + Ryan Johnston, Colgate, Jr, D #2

Backup Data:

Returning (Y) and Non-Returning (N) players by team by position.  Early departures in bold red:

	RetF		RetD		RetG		WRet
Brown	0.60		0.58		1.00		0.66
Ret	69		18		1869
Total	115		31		1869
1	23	N	13	N	1058	Y
2	23	Y	7	Y	773	Y
3	21	Y	6	Y	38	Y
4	12	Y	5	Y
5	11	N
6	7	Y
7	6	N
8	6	Y
9	6	N
Clarkson	0.76		0.81		1.00		0.82
Ret	103		34		2223
Total	135		42		2223
1	21	Y	15	Y	998	Y
2	20	N	10	Y	795	Y
3	17	Y	9	Y	430	Y
4	15	Y	8	N
5	13	Y
6	13	Y
7	12	Y
8	12	Y
9	12	N
Colgate	0.63		0.45		1.00		0.63
Ret	117		24		2297
Total	187		53		2297
1	31	Y	15	Y	2095	Y
2	29	N	15	N	202	Y
3	26	Y	14	N
4	24	N	9	Y
5	20	Y
6	20	Y
7	17	N
8	11	Y
9	9	Y
Cornell	0.58		0.28		1.00		0.55
Ret	57		9		1870
Total	99		32		1870
1	22	N	14	N	1476	Y
2	14	Y	9	N	394	Y
3	14	Y	5	Y
4	12	N	4	Y
5	11	Y
6	8	N
7	8	Y
8	5	Y
9	5	Y
Dartmouth	0.37		0.53		1.00		0.53
Ret	60		23		1992
Total	162		43		1992
1	30	N	14	Y	1577	Y
2	24	N	13	N	415	Y
3	23	N	9	Y
4	22	Y	7	N
5	19	Y
6	15	N
7	10	Y
8	10	N
9	9	Y
Harvard	1.00		0.57		0.02		0.69
Ret	229		28		56
Total	229		49		2288
1	58	Y	21	N	2232	N
2	48	Y	10	Y	43	Y
3	30	Y	9	Y	13	Y
4	27	Y	9	Y
5	18	Y
6	17	Y
7	17	Y
8	7	Y
9	7	Y
Princeton	0.79		0.40		1.00		0.70
Ret	54		8		1785
Total	68		20		1792
1	14	Y	8	N	1676	Y
2	10	Y	5	Y	109	Y
3	10	Y	4	N	7	N
4	8	N	3	Y
5	7	Y
6	6	N
7	5	Y
8	4	Y
9	4	Y
Quinnipiac	0.82		0.50		1.00		0.75
Ret	169		33		2393
Total	205		66		2393
1	39	Y	22	N	2153	Y
2	36	N	20	Y	240	Y
3	33	Y	13	Y	0	Y
4	31	Y	11	N
5	17	Y
6	15	Y
7	13	Y
8	12	Y
9	9	Y
RPI	0.65		0.71		0.74		0.68
Ret	84		25		1816
Total	130		35		2450
1	20	Y	10	N	1816	Y
2	19	N	9	Y	634	N
3	16	N	8	Y
4	15	Y	8	Y
5	14	Y
6	13	Y
7	12	Y
8	11	N
9	10	Y
SLU	0.72		0.78		1.00		0.79
Ret	118		53		2245
Total	164		68		2245
1	26	Y	20	Y	2186	Y
2	24	N	17	Y	59	Y
3	23	Y	16	Y
4	22	N	15	N
5	17	Y
6	14	Y
7	13	Y
8	13	Y
9	12	Y
Union	0.64		1.00		0.27		0.70
Ret	152		66		644
Total	239		66		2390
1	50	N	31	Y	1742	N
2	50	Y	13	Y	644	Y
3	28	Y	11	Y	4	N
4	28	N	11	Y
5	25	Y
6	19	Y
7	22	Y
8	9	N
9	8	Y
Yale	0.93		0.57		1.00		0.82
Ret	132		29		2044
Total	142		51		2044
1	23	Y	21	Y	1925	Y
2	21	Y	14	N	119	Y
3	19	Y	8	N
4	19	Y	8	Y
5	18	Y
6	15	Y
7	10	N
8	10	Y
9	7	Y

Summary:

	RetF	RetD	RetG	WRet
Brown	.60	.58	1.00	.66
Clarkson	.76	.81	1.00	.82
Colgate	.63	.45	1.00	.63
Cornell	.58	.28	1.00	.55
Dartmouth	.37	.53	1.00	.53
Harvard	1.00	.57	.02	.69
Princeton	.79	.40	1.00	.70
Quinnipiac	.82	.50	1.00	.75
RPI	.65	.71	.74	.68
SLU	.72	.78	1.00	.79
Union	.64	1.00	.27	.70
Yale	.93	.57	1.00	.82

Avg10

Avg10 measures traditional program strength, by taking the team's mean number of Points over the prior 10 seasons.  This will be used to compute the expected Incoming factor.

Backup Data:

	Brn	Clk	Col	Cor	Drt	Hvd	Prn	Qpc	RPI	SLU	Uni	Yal
2006	11	20	30	29	30	27	17	17	22	25	22	14
2007	16	30	17	24	27	22	22	24	17	33	15	17
2008	15	33	21	25	15	27	28	22	15	16	25	22
2009	10	20	17	29	24	24	28	21	13	26	20	32
2010	16	11	26	31	17	17	18	22	23	23	28	32
2011	18	19	11	24	26	15	24	19	24	13	36	35
2012	14	22	23	30	19	25	16	23	17	21	32	22
2013	20	19	15	19	22	14	20	37	27	22	24	25
2014	17	24	29	24	16	16	8	28	21	18	37	24
2015	13	19	26	22	26	25	6	35	18	29	17	28
Pts	150	217	215	257	222	212	187	248	197	226	256	251
Avg	15.0	21.7	21.5	25.7	22.2	21.2	18.7	24.8	19.7	22.6	25.6	25.1

Summary:

	Avg10
Brown	15.00
Clarkson	21.70
Colgate	21.50
Cornell	25.70
Dartmouth	22.20
Harvard	21.20
Princeton	18.70
Quinnipiac	24.80
RPI	19.70
SLU	22.60
Union	25.60
Yale	25.10

Putting it All Together

From these independent metrics, we now generate several calculated metrics:

All but the last of these calculated metrics are just convenient conceptual groupings -- the final metric, Nieu, could be calculated directly from the independent metrics.  But we'll sacrifice brevity for clarity of expression (and check our work) by spelling them out in the table below: