The proposed "inner ring/outer ring" admissions solution is interesting. The inner ring of admission slots (say 20%) are filled purely on academic merit metrics (NMS, GPA, SAT score, etc), then 80% of slots are filled based on dumping everybody else who meets some much lower cut-off on those academic merits into a lottery for the remaining admissions slots.
I'm not entirely sure the author appreciates the likelihood of various strange outcomes using a purely random lottery for admissions, for example "outlier" admissions classes that heavily over-represent one racial/gender group by chance.
Another rather damning point of the article is that university admission officers may be spectactularly unqualified to actually make admissions decisions. But I have been privy to the admissions decision process at the graduate level and bucketing applicants into the "will succeed" and "will not succeed" and "who knows" bucket usually winds up with a rather sizable number in the "who knows" bucket. So maybe the critque on the admission officers should be less worrisome than presented. I think PG has made statements that one of the biggest problems with YC applicants is that there just isn't an easy way to look at many applicants and make any reasonable determination about whether or not they could create successful startups.
Anyway, this is an excellently thought-provoking article and thanks for bringing it here.
I'm not entirely sure the author appreciates the likelihood of various strange outcomes using a purely random lottery for admissions, for example "outlier" admissions classes that heavily over-represent one racial/gender group by chance.
Another rather damning point of the article is that university admission officers may be spectactularly unqualified to actually make admissions decisions. But I have been privy to the admissions decision process at the graduate level and bucketing applicants into the "will succeed" and "will not succeed" and "who knows" bucket usually winds up with a rather sizable number in the "who knows" bucket. So maybe the critque on the admission officers should be less worrisome than presented. I think PG has made statements that one of the biggest problems with YC applicants is that there just isn't an easy way to look at many applicants and make any reasonable determination about whether or not they could create successful startups.
Anyway, this is an excellently thought-provoking article and thanks for bringing it here.