mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
01-22-18 05:57 PM - Post#244407
In response to SRP
Yeah, it's very interesting. I do think there's mass underestimation of what being a Top 40-50 KenPom team takes.
The reason why Harvard could sustain dominance is that it stacked recruiting classes together. Harvard's 2009 entry class has the fourth-highest predicted win shares in the model (and finished 6th of all classes), while its 2011 class has the second-highest predicted win shares in the model (and finished 3rd). Along with those, Harvard had okay 2008 and 2010 classes as well (that happened to over-achieve).
Where Harvard went off the rails recently is having three okay classes in a row (2013, 2014, 2015), having injuries impact much of the potential playing time from that group and having other players remaining underachieve. You don't need to hit every class (you can get to Pomeroy Top 50 just by hitting on 2 in 4 years), but you can't miss on 3 in a row without feeling it.
Of the 130 predicted Ivy classes going back to 2003, here's where Harvard's scored in terms of expected win shares (rank overall is in parentheses first, then hyphen projected win shares then parentheses actual win shares):
Harvard 2016 (1) - 62.9 (16.8 actual through 1.5 years)
Harvard 2011 (2) - 44.9 (44.4 actual)
Harvard 2018 (4) - 32.7 (TBD)
Harvard 2009 (5) - 30.9 (36.5)
Harvard 2002 (25) - 20.2 (19.4)
Harvard 2013 (26) - 19.6 (6.9)
Harvard 2005 (T32) - 18.7 (17.1)
Harvard 2010 (T32) - 18.7 (34.0)
Harvard 2012 (T32) - 18.7 (25.9)
Harvard 2015 (T32) - 18.7 (7.2 through 2.5 years)
Harvard 2006 (59) - 16.1 (18.4)
Harvard 2008 (70) - 14.7 (25.3)
Harvard 2003 (T103) - 10.6 (3.3)
Harvard 2007 (T103) - 10.6 (0.3)
Harvard 2014 (T103) - 10.6 (1.9 through 3.5 years)
Harvard 2017 (T103) - 10.6 (0.1 through 0.5 years)
Harvard 2004 (130) - 4.0 (2.5)
Edited by mrjames on 01-23-18 06:25 PM. Reason for edit: Didn't have Mason Forbes in Harvard 2018 class, projections changed a bit with him added
|
