I believe we will disagree on on the weight of geography in the admission process. We don’t get to see the process in each Admission Office so we’re all speculating to some degree. I believe in looking at the statistics. Over time, those will tell you exactly what a college wants its class to look like. Every Top 20 college receives applications of “qualified” students to fill its class multiple times over. (I’m not being critical of Wellesley. It is typical of nearly all.) If NESCAC colleges wanted 25% of their respective student bodies from the Midwest, they certainly receive a sufficient number of qualified applications. The proportion of applications is irrelevant to the gross number of qualified applicants and the ability of a college to distribute seats according to objective criteria. Distance from home likely impacts other metrics like yield, retention and transfer rates. I’d like to think that colleges don’t admit students to target these stats used for ratings, but…
I do agree that many small liberal arts colleges have a bias toward high schools with historic relationships. They know the schools’ demographics well. Private prep schools and wealthy suburban schools districts feed them qualified, full-pay applicants. There is definitely more uncertainty in the ability to pay from the public high school in a low-income zip code for example. (I will believe in a fully need-blind admission process when I see statistics that show a college with random, need-based financial aid expenditures.)
My point in looking in-depth at the historic admission statistics rather than scores/GPAs is to avoid being the unhappy Honors student at State U when you really wanted an LAC experience.