They disclose enrollment by city/county in Virginia. http://avillage.web.virginia.edu/iaas/instreports/studat/dd/enrl_city.htm
You’ll note that (2014 numbers) some 2634 undergrads are from Fairfax County alone. 763 are from Loudoun County, 433 from Prince William County, 304 from Arlington, 232 from Alexandria, 74 from Fairfax City, 91 from Falls Church City, 57 from Manassas, and 10 from Manassas Park, out of 15,405 total - that’s 4598/15405 (29.8%) from NoVa. There are only 11 other cities and counties that even have more than 100 students represented.
According to the US Census Bureau, those some locations have 2,419,504 population out of the total state population of 8,326,289 (29.1%).
http://quickfacts.census.gov/qfd/states/51000.html
Nearly the same, right? 29.8% of the state population, 29.1% of the UVA students.
You can combine the admissions data (9156 completed IS applications in 2015)
http://avillage.web.virginia.edu/iaas/instreports/studat/hist/admission/first_by_residency.htm
with SCHEV data on numbers of applicants
http://research.schev.edu/localities/LD01_Admissions.asp
to see that 4385 applications came from the above locations, 4385/9156 = 47.9% of the total IS applications were from NoVa!
29.8% of UVA population, 47.9% of IS applications, means NoVa applicants are WAY over-represented in the application pool.
I don’t have the yield (accepted admissions offers) breakdown by county. If NoVa students are admitted at the same rate as everywhere else (and I have no way to know whether this is the case), they would need to have a much lower yield to end up with that portion of the population after submitting so many more of the applications. On the other hand, if they are held to the same standard as the rest of the state (for example, top X% of your graduating class is a likely admission offer), then the applicants below that standard would be less likely to get admission offers, which would help explain why with so many applications they result in no more than proportional representation.