Gpakit
Band: · combined z-score:

What this calculator does and does not do

College admissions at competitive US schools are holistic. Admissions officers read essays, weigh the strength of your course selection, talk to your recommenders, consider your extracurricular depth, and factor in everything from state of residence to intended major to first-generation status. No calculator can model that. What a calculator can do is translate a few published numbers — your GPA, your SAT, and the school's middle-50 range and overall admit rate — into a rough statistical baseline. That is the purpose of this tool, and it is the limit of the tool.

Treat the output as the question "Where do I stand numerically relative to the typical admitted student?" and nothing more. A 45% baseline does not mean you have a 45% chance of admission. It means that on the two axes we can measure, you are slightly below the typical admitted profile for a school with that acceptance rate. The other 55 percentage points of your application — the parts that matter most at selective schools — are invisible to this calculator.

The math, written out

The calculator computes two standardized scores. The first is your SAT expressed in standard-deviation-like units relative to the school's middle-50 range: z_sat = (your SAT − midpoint) ÷ half-spread, where the midpoint is the average of the 25th and 75th percentile SAT scores. The second is your GPA relative to the school's admitted average: z_gpa = (your GPA − average) ÷ 0.25, using 0.25 as a rough one-sigma spread for GPA at competitive schools. These combine with a 60/40 weighting (test slightly heavier because published SAT data is usually more precise than GPA averages): z_combined = 0.6 × z_sat + 0.4 × z_gpa.

That combined z-score is added to the logit (log-odds) of the school's overall acceptance rate, with a scaling coefficient of 1.1 on the z-term — a conservative choice that prevents the estimate from shooting to 99% for applicants above the 75th percentile. A logistic function squashes the total into a 0–100% baseline: p = 1 / (1 + e^(−(logit + 1.1 × z_combined))). It is transparent, reproducible, and deliberately modest.

Reach, target, safety — the usual buckets

College counselors organize applicants' school lists into reaches, targets, and safeties. The calculator applies the convention mechanically: under 30% baseline = reach, 30–60% = target, 60%+ = safety. One important caveat: any school admitting under 15% of applicants is functionally a reach for every applicant, regardless of stats. At a 6% admit rate, high-stat applicants are still rejected more often than not — the base rate simply dominates everything else. Use numeric buckets as a rough sort, not as a guarantee.

A balanced application list generally includes at least two safeties where you would be genuinely happy to enroll, several targets, and one to three reaches. If your safety list is empty or consists of schools you would resent attending, the list is not actually balanced — rebuild it.

What the calculator deliberately ignores

Every one of these matters at selective schools, and none are in the model: your essays, the quality and specificity of your recommendation letters, your intended major and its competitiveness (computer science and business are often the hardest admits even within a given school), course rigor beyond raw GPA, class rank, the high school you attended (admissions offices know the feeder schools), legacy status, first-generation status, racial and ethnic background (post-SFFA, still factored informally via essays), state of residence for public schools, ability to pay for schools that are not need-blind, interviews and demonstrated interest at schools that track it, and athletic or arts recruitment.

An applicant with average stats and an extraordinary application (national-level achievement, a distinctive and well-told story, strong rec letters, match with institutional priorities) will beat a higher-stat but thinner application at most selective schools. The statistical baseline is only the entry fee — it gets you read seriously. Everything past that is qualitative.

Where to find the input numbers

Every US college publishes a Common Data Set (CDS), a standardized annual report of admissions statistics. Search "{school name} Common Data Set" to find the most recent year — usually published as a PDF on the institutional-research or admissions office's site. Section C1 lists overall acceptance rate. Section C9 lists SAT and ACT middle-50 ranges for enrolled freshmen. Section C11 lists admitted-student GPA distributions (when reported). For schools that do not publish a CDS, The College Board's BigFuture, IPEDS via College Navigator, or The Princeton Review's college profiles are workable alternatives.

Using the output responsibly

The right way to use a baseline estimate: run it for every school on your list, identify whether your list is numerically balanced, and then put the number away and focus on the parts of your application you can actually change — essays, the strength of your course selection in senior year, the specificity of your activity descriptions. The wrong way: comparing your 32% baseline to a classmate's 41% and concluding their odds are meaningfully better. Both numbers are estimates of a small piece of the picture. If a college counselor or admissions officer tells you your stats are in range, believe them over any calculator output.

Frequently asked questions

Is this a real prediction of my admission chances?
No. This is a statistical baseline that compares your GPA and SAT against a school's published middle-50 range and overall acceptance rate. It deliberately ignores essays, recommendation letters, extracurriculars, legacy status, first-generation status, demographics, interviews, portfolios, and institutional priorities — all of which weigh heavily in holistic review at selective colleges. Treat the output as a rough numeric starting point for conversations with a counselor, never as a forecast.
How is the probability calculated?
The calculator computes a standardized score (z-score) for your SAT relative to the school's middle-50 midpoint, and a z-score for your GPA relative to the school's admitted average. These are combined with a 60/40 weighting (test slightly heavier because the range data is usually more precise) and then added to the logit of the school's overall acceptance rate. A logistic (sigmoid) function squashes the result into a 0–100% estimate.
What do "reach", "target", and "safety" mean?
These are conventional shorthand. A reach is a school where your numbers put you below the typical admitted profile (estimated probability under 30%). A target is where you are in the ballpark (30–60%). A safety is where your numbers are above the typical admitted student and the school has a meaningful acceptance rate (60%+). Selective sub-15% acceptance schools (Ivies, MIT, Stanford) should generally be considered reaches for everyone regardless of stats — holistic review dominates.
My SAT is above the 75th percentile — why isn't my probability 90%+?
Because the school's overall acceptance rate anchors the estimate. At a school admitting 8% of applicants, even a strong test score does not push the mathematical baseline into near-certainty territory. That is actually realistic: at highly selective schools, most rejected applicants also have strong test scores, so stats alone do not guarantee admission.
Where do I find middle-50 SAT and acceptance-rate data?
Every US college publishes a Common Data Set (CDS) with these numbers — search "{school name} Common Data Set" to find the most recent year. Section C9 lists test-score ranges. Acceptance rate is section C1. For international schools, admissions statistics pages or The Princeton Review profiles are workable substitutes.
What about ACT instead of SAT?
Convert your ACT to an equivalent SAT first using the College Board concordance table (or our ACT-to-SAT converter), then enter the SAT-equivalent number. Most colleges publish both ACT and SAT middle-50 ranges in their CDS, and the concordance is generally accurate within a narrow band.
Does this work for test-optional schools?
Partially. If the school is test-optional, published middle-50 ranges reflect only admitted students who submitted scores — usually a self-selected higher-scoring group. Your baseline estimate using that data will be slightly pessimistic if you plan to submit, and completely uncalibrated if you plan not to submit. For test-optional applications, GPA, transcript rigor, and non-numeric factors matter disproportionately.
From the blog

Learn more