I think it would help greatly if there could be more visibility into the stages of the admission decision process.
In other words, it’s probably the case that in the first screen, academic factors dominate, whereas in the second and third/final screens, other factors rise in prominence.
Also, there’s almost certainly pooling going on ie dividing the applications into cohorts that conform to the major categories by which the university leadership evaluates the success of its admissions process. Obviously these cohorts or pools correspond to the major USNWR metrics-- SAT, grade point, rank, etc-- but equally obviously, the leadership wants to see certain ethnicity/URM thresholds maintained.
Less obvious is the elephant in the room: the admissions process has to optimize for REVENUE.
Without question the adcom’s leadership is aware of, and working toward, some threshold number for tuition revenue that the process, in the end, must attain. This skews the process in favor of those applicants who, the adcoms and their consultants can predict with high confidence, will accept an offer of admission that carries no aid at all. So another pool certainly has to be those minimally-qualified, academically-suitable applicants whose parents’ occupations, ZIP codes, and other non-FAFSA indicators suggest that they will easily pay full tuition-- and not receive aid anywhere.
In short, wealth is another hook.
The only question is at which stage in the admissions review process-- first screen, second or final-- the pooling takes place.