I don't know what "cut from the same cloth" means but if you're referring to our common humanity, sure. If you're claiming that the level of academic achievement (at least before matriculation) is the same across the ACC regardless of school for those that ultimately graduate, I beg to differ. Also, I would not use anecdotal criteria to form an opinion in this matter. That you personally knew of no NSF grant winners does not mean there were not any at Duke. More importantly, there are many criteria beyond "# of NSF grant winners" to judge the quality of work done by matriculants at a school. You could look at Rhodes scholars, for one; published research among undergraduates for another. And many more - it would be very myopic to limit such a measure to one or two such criteria. Of course, you would have to adjust for size of school. Duke is not a large school - perhaps 2nd smallest in ACC, behind Wake, unless some of these "post-8 ACC" are smaller.
Certainly it sometimes happens that excellent students do not bother to apply because they have an excellent in-state choice. This is probably most pronounced in NC and VA, where the flagship public universities are of great selectivity. But if this happened to a great extent, Duke would not have the kind of entering class profile that it does.
I am not a fan of the USN&WR ranking system, which has a substantial subjective component that is subject to manipulation by the schools and editors. However, I am not opposed to rankings per se. I think each applicant should have their own criteria in mind and use them to rank the desirability of schools. To get back to Mudge's OP, there are some objective criteria why which the ACC schools may be ranked and there's nothing inherently wrong with considering such information.
For my source, I use the 2011 Princeton Review "Best 373 Colleges" guide which I have at hand. Acceptance rates show significant differences:
School Acceptance
Duke 0.22
BC 0.30
UNC 0.32
UVA 0.32
WFU 0.38
MD 0.42
Miami 0.44
NCSU 0.55
GT 0.59
Pitt 0.59
Syr 0.60
FSU 0.61
Clem 0.63
VT 0.67
By incorporating yield, you really get some separation. Divide yield by acceptance rate; one could infer that the higher ratios reflect selectivity
and desirability:
School Yield/Accept
Duke 1.95
UNC 1.69
UVA 1.50
WFU 0.92
MD 0.83
BC 0.83
NCSU 0.82
FSU 0.69
GT 0.68
VT 0.54
Clem 0.52
Pitt 0.49
Miami 0.48
Syr 0.43
You could rank based on median test scores, the % of students who were in the top 25% of their class, etc. IMO there are significant differences, though mostly among the median performing students at each school. The top students at each school would probably be tops at most if not all of the other schools as well.
There was one very interesting ranking study done in 2005 by Avery et al (
http://papers.ssrn.com/sol3/papers.c...ct_id=601105##) that appears to be no longer freely available on the web unless you subscribe to certain journals. It constructs a revealed preference ranking by looking at many pairwise comparisons between schools for which a given student gained acceptance, and which school the applicant chose. By doing so, the authors hoped to take into account the fact that some students chose a less selective school because of any number of factors - such as cost after scholarships/financial aid, location, the existence of particular programs of study, legacy ties, etc. It purports to "reveal" the preferences of students based on actual matriculation statistics and came up with a weighted average rank based on matriculation patterns when there were such head-to-head choices made by students. Note that this is NOT by design a selectivity thing, except to the extent that applicants are making the selection. It is also hard for the schools to manipulate the rankings.
19. Duke
20. UVA
25. GA Tech
32. UNC-CH
49. U-MD
52. WFU
57. BC
58. Miami
73. FSU
108. Syracuse
Apparently they did not get data from NCSU, Pitt, Clemson, or Va Tech. It's pretty interesting that Notre Dame comes in at #13, well above Duke, but when you realize how the selection effects work, it's not so surprising. A student applying to ND (or Georgetown, #16) may be particularly interested in attending a Catholic school, so in head-to-head, they may win out more than you would think based strictly on test scores, admission rates, etc. Similarly, Wellesley at #11 comes as a bit of a shock until you consider who might apply there and how they might consider head to head choices between schools. That said, these few are the exceptions - the ranked list begins mostly as you might expect, with Harvard, Caltech, Yale, MIT, Stanford, Princeton...