Methodology
The complete methodology behind every score AdmitPath generates — including which sources we use, how we weight each dimension, and what the score deliberately does NOT claim to predict.
The 0–100 composite score is a weighted average of seven sub-scores. Weights are not equal — they reflect what selective US admissions committees actually weigh per the Common Data Set Section C7. The weights below are the AdmitPath defaults; per-school overlay tunes these for all 102 schools where we have school-specific CDS C7 data.
Course-load strength relative to what your school offers. The rigor scorer detects and weights AP courses, IB (Higher/Standard Level), dual enrollment (DE) college courses, and honors sections separately. Calibrated against your school's `apsOffered` value (default 12 if unknown) so 5 APs at a 5-AP school scores like 12 APs at a 30-AP school. Junior-year courseload is weighted 1.4x freshman-year because admissions reads recent rigor more heavily. IB HL courses are treated as AP-equivalent; DE courses weighted 0.8x AP.
Sources
Roles held + scope + duration. President of a 30-member club for 3 years > captain of a varsity team for one season. Self-started initiatives (founded the X club, launched the Y program) score higher than inherited roles. Family responsibility (caring for siblings, supporting a family business) counts as leadership.
Sources
Tiered: international > national > regional > state > school. Selectivity matters more than count: 1 international award beats 5 school awards. Research awards (Regeneron STS, ISEF) score very high. Participation 'awards' (NHS membership) score zero.
Sources
Time + commitment + verticality across the activities list. Five years of one violin teacher + state-level performances reads as depth. Eight clubs joined freshman year reads as breadth. Calibrated to favor longer commitments and increasing scope of action year-over-year.
Sources
The single deepest vertical in the profile, scored on (1) clarity of focus, (2) artifact strength (research paper, app, novel, founded org, etc.), (3) trajectory (increasing depth year-over-year). Profiles without a clear spike score lower at T20 schools where spike is preferred over breadth.
Sources
6-dimension sub-score (authenticity, insight, specificity, storytelling, impact, voice). Authenticity weighted highest (0.20) because admissions readers can detect heavily edited essays. Voice weighted 0.10 and evaluated via a 4-axis voice rubric based on the College Essay Guy framework: place (grounding in a physical location with proper nouns and sensory details), detail (sentence-level craft, specific objects, dialogue, sentence-length variance), vulnerability (honest self-disclosure, admission of doubt, willingness to sit in discomfort), and surprise (unexpected connections, intellectual tension, position-taking). These 4 voice axes plus 2 additional axes map into the full 6-dimension essay score. Place and detail are the strongest signals of a genuine essay.
Sources
Weighted lower because students have less control over rec quality once submitted. Score reflects (1) breadth of teacher contact, (2) brag-sheet quality (a proxy for rec quality), (3) counselor-relationship signals. The dimension exists to remind students to invest in teacher relationships early.
Sources
The default weights above are reasonable for the average T50 US college. For all 102 schools in AdmitPath's database with publicly available Common Data Set Section C7 data, AdmitPath overlays the school-specific weights on top of the default. Example: Yale weights "character/personal qualities" as very important, so the leadership + spike + essay-quality sub-scores are weighted higher in Yale-specific composite calculations. Caltech weights "test scores" as not considered after the 2025 pivot, so the rigor sub-score absorbs that weight in Caltech-specific calculations.
All 102 schools in AdmitPath have school-specific C7 weights. Schools added in the future without published CDS data would use the defaults until their CDS is sourced. We refresh weights at the start of each application cycle (September–October) when schools publish their updated CDS sections.
AdmitPath's per-school predictor returns one of four bands — Very Likely, Possible, Long Shot, Hail Mary — instead of a precise percentage like "63%." Internally, the admit-rate engine uses 4-tier probability curves that map composite scores to conditional admit-rate ranges for each school.
This is deliberate. Selective US admissions has too much hidden state (institutional priorities, hooks, year-to-year yield management, individual reader subjectivity) for any model to claim two-decimal accuracy. False precision misleads users into treating a 63% as "I'll probably get in" when the actual underlying uncertainty is much wider.
The 4-tier probability curves are calibrated against published per-school admit rates (from CDS Section C) plus the applicant's composite score relative to admitted-class medians. The bands map roughly:
Honesty about the model's limits is part of the methodology. The score deliberately excludes:
These factors can swing individual decisions significantly. Use the score to identify gaps and prioritize work — not as an oracle.
Every profile runs through the same deterministic pipeline. No black boxes, no hidden variables — you can trace every number back to a source.
Profile Ingestion
GPA, courses, activities, awards, essays, recommendations entered by student
School-Context Normalization
Rigor scaled against school's AP offering count. 5 APs at a 5-AP school = 12 APs at a 30-AP school
7-Dimension Sub-Scoring
Each dimension scored 0-100 independently: Academic Rigor (20%) + Leadership (15%) + Awards (10%) + Activity Depth (20%) + Spike (15%) + Essay Quality (15%) + Recommendations (5%)
CDS C7 School-Specific Overlay
For all 102 schools, weights are adjusted to match what that specific school values per CDS Section C7 reports
Weighted Composite Score (0-100)
Final composite = weighted average of 7 sub-scores, anti-inflated to correlate with real admit outcomes
Per-School Band Prediction
Composite mapped to 4 honest bands: Very Likely / Possible / Long Shot / Hail Mary. No false-precision percentages
Personalized Action Plan
Gap analysis generates a prioritized 30/60/90-day plan targeting the dimensions that will move your composite the most
Every number in AdmitPath traces back to a verifiable source. We don't manufacture data — we ingest it from institutions that publish it.
CDS Section C7
102 schools with school-specific weights
The Common Data Set Section C7 is where each university self-reports how much weight it places on academics, activities, character, and other factors. We pull these annually to calibrate per-school scoring overlays.
College Scorecard API
U.S. Department of Education
Net price by income band, median earnings 10 years post-graduation, student-loan repayment rates, and completion rates. Powers our /net-price and /roi calculators.
IPEDS / NCES
Federal enrollment & outcomes data
Integrated Postsecondary Education Data System — enrollment figures, retention rates, graduation rates, and demographic breakdowns. The same dataset behind College Navigator.
Common App Statistics
Application volume & activity norms
Aggregate application trends, activity categorization patterns, and essay submission statistics that help us benchmark what 'normal' looks like across the applicant pool.
All data is refreshed at the start of each admissions cycle (September–October) when schools publish updated CDS sections and the Department of Education releases new Scorecard data.
Most admissions tools inflate scores to make users feel good. A student with a 3.5 GPA and mid-tier extracurriculars gets told they have a "78% chance" at Stanford. This is dishonest — and it leads to heartbreak.
AdmitPath does the opposite. Our scoring is deliberately calibrated to correlate with actual admit outcomes, not to flatter users. This means:
Score compression at the top
A composite of 90+ requires genuine national-level achievements. You can't get there with good-but-not-exceptional credentials.
School-context normalization
5 APs at a school that offers 5 means something very different than 5 APs at a school that offers 30. We normalize for opportunity, not raw count.
Participation ≠ achievement
NHS membership scores zero in awards. 'Volunteering' without measurable impact scores low in activity depth. We reward substance, not resume padding.
No score inflation over time
If your profile hasn't meaningfully improved, your score shouldn't go up. We resist the temptation to gamify progress with artificial score bumps.
We're not the only tool that claims to help with admissions. But we're the only one that does all of this.
| AdmitPath | CollegeVine | Niche | Private Counselor | |
|---|---|---|---|---|
| Price | Free / $19.99/mo | Free (ad-supported) | Free (ad-supported) | $5K–$50K |
| Scoring transparency | Full methodology published | Proprietary algorithm | Letter grades, no methodology | Subjective opinion |
| Data sources cited | CDS C7, IPEDS, Scorecard | Not disclosed | Not disclosed | Varies by counselor |
| Per-school calibration | 102 schools w/ CDS overlay | Unknown methodology | No per-school scoring | Anecdotal |
| Essay feedback | 6-axis AI + line-by-line | Peer review marketplace | None | Yes (expensive) |
| Anti-inflation scoring | Yes — deliberately honest | No — inflated % | No — letter grades | Depends on counselor |
| Action plan generation | 30/60/90-day personalized | Generic checklist | None | Yes (manual) |
| Available 24/7 | Yes | Yes | Yes | By appointment |
Run a free analysis and see which of the 7 dimensions are pulling your composite up -- and which are dragging it down.