Most families think the process is meritocratic. The SFFA v. Harvard lawsuit forced six years of admissions data into open court — and it told a different story. Recruited athletes are admitted at 86%. Legacies at 33%. Unhooked applicants at six.
Four letters, four advantages: Athletes, Legacies, Dean's-list children of donors, and Children of faculty and staff. Inside Harvard's admissions office, these applicants are read on a separate track.
The numbers, pried loose by the Arcidiacono–Kinsler–Ransom analysis of six admission cycles, are not subtle. Recruited athletes were admitted at 86% — fifteen times the baseline rate for everyone else. Children of faculty: 47%. Donor-list candidates: 42%. Legacies: 33%.
For an unhooked applicant, the rate hovers around five or six percent. Same applications, same office, same year. The hook is the variable.
A higher rate is one thing. A larger share of the class is another.
Apply by November 1, agree in writing to enroll if accepted, and watch the math change. At Dartmouth, the published Early Decision rate is 19.1% against an overall rate of 5.4% — a 3.5× multiplier. Columbia: 13.2% versus 3.9%. Northwestern: 23% versus 7.7%. Across the elite private universities, the pattern holds school after school.
Some of that gap is self-selection — ED applicants tend to be stronger and more committed. But pool quality alone cannot explain a 3× swing. Binding ED guarantees yield, the single number that drives U.S. News rankings hardest. Every ED admit is a seat locked, a yield-rate calculator that will read 100%.
The catch: ED commits before the financial-aid letter arrives. Families who need to compare aid offers across schools cannot use it. The early-round bonus becomes a privilege of the families that don't have to do the math.
If ED locks yield in, what happens to applicants who didn't play that card?
Imagine a school in the 15-to-40 range — selective enough to care about yield, not selective enough to assume it. A 1580 SAT, 4.0 GPA applicant arrives with no campus visit, no “Why Us?” essay, no early-decision flag. The committee suspects she is using them as a safety. The applicant is waitlisted.
That, in a sentence, is yield protection — sometimes called Tufts Syndrome after the school it became associated with in the 1990s. No admissions office has ever officially confirmed the practice. But Naviance scattergrams and counselor observations consistently show an “inverted-U” admit pattern: rates rise with stats, then dip at the very top.
The schools most often named: Tufts, Tulane, Northeastern, Case Western, Boston University, the University of Chicago, the University of Richmond. Applying ED neutralizes the penalty entirely — a binding commitment is, by definition, a guaranteed yield.
If timing and signaling matter this much, what about the schools where stats do not?
Counselors have a phrase for this: “Every T20 is a reach for everyone.” Once a school's overall acceptance rate falls below ten percent, even a 1600 SAT, national-tier extracurricular profile cannot push admit probability above roughly 15-20%. The variance starts to come from elsewhere — reader assignment, that year's institutional needs, randomness.
The chart on the right takes the standard counselor taxonomy — safety, target, reach, high-reach, lottery — and plots how each maps to actual admit probability for an unhooked applicant. Notice how the “target” band collapses entirely once acceptance rates drop below 10%.
For an applicant with median-of-admits stats applying unhooked to a sub-10% school, the structural ceiling is roughly the school's overall rate times 1.5. Above that, the dice take over.
Some schools watch every click. Others ignore them completely.
Tulane tracks email opens. Tufts tracks campus visits. Wake Forest reads “Why Us?” essays as load-bearing evidence. Harvard, Princeton, Stanford, and the University of California system don't track demonstrated interest at all — their pools are self-selecting.
We pulled Common Data Set Section C7 filings for 55 simulation colleges. Of those, 35 mark demonstrated interest as “not considered.” Seventeen rate it “considered.” Three call it “important” — Tulane, Northeastern, and Wake Forest. None call it “very important.”
At schools that do track it, the boost is real but modest: a 1.1-1.3× multiplier on base probability. The penalty for ignoring it at a tracking school can be larger. At schools that don't track it, the campus visit is a research trip, not a tactic.
Inside the office, the numbers feed a six-axis rating sheet.
SFFA's findings of fact described the rubric in detail. Each application receives a rating from 1 (best) to 6 (worst), with plus and minus modifiers, across six dimensions: academic, extracurricular, athletic, personal, recommendations, and the alumni interview.
A “1” in any single category is rare — top one percent of applicants nationally — and almost always advances to admit. A “3” is the median. A “4” is described in court filings as “bland, somewhat negative, or immature.”
The personal rating drew the most fire in the trial. Reviewers scored applicants on humor, sensitivity, grit, leadership, integrity, helpfulness, courage, kindness. Asian American applicants received systematically lower personal ratings than comparable peers — a finding the plaintiffs argued was discriminatory and the school argued was an artifact of teacher recommendations and interviews.
Hooks. Rounds. Demonstrated interest. Ratings. They all multiply.
Pull every documented multiplier into one chart and the architecture of admissions becomes legible. Recruited athlete: 3.5×. Donor list: 3.0×. Legacy: 2.5×. Faculty child: 2.0×. First-generation: 1.4×. Underrepresented geography: 1.3×. Early Decision: 1.8×. Demonstrated interest, where tracked: 1.2×. Yield-protection penalty: 0.7×.
The multipliers compound. A recruited athlete who is also a legacy and applied ED starts the process near a ceiling. An unhooked applicant from California with no ED commitment, no campus visit, and a competitive major starts well below the published rate.
None of these are secret. They're in lawsuit exhibits, NACAC surveys, Common Data Set filings, and admissions-office FAQs. Together they explain why “the same college” feels like five different colleges depending on who is applying.