Step Pass-Likelihood Estimator

Methodology — everything on the table

This tool is a transparent benchmark-based calculator, not a statistical model. It starts from national benchmarks and adjusts them with fixed, published mapping tables shown below. It uses no regression coefficients, no proprietary score-conversion tables, and no commercial predictor algorithms. The adjustments reflect two robust published findings:

Preadmission scores explain little on their own. Gullo et al. (2015, Medical Science Educator) found MCAT totals and math-science GPA together explained only about 11.7% of the variance in Step 1 performance (adjusted R²) — so MCAT moves the estimate modestly.
In-school performance is the dominant predictor. In the same study, adding the first medical-school exam scores raised prediction to ~27.9%, and adding second-year exam data pushed it past 50%. Preclinical performance moves the estimate more than the MCAT does.

Supporting evidence: the AAMC's MCAT Validity Research Summary reports large median correlations between MCAT scores and preclerkship, Step 1, clerkship, and Step 2 CK performance across schools; Gauer & Jackson (2016, Medical Education Online) found MCAT components explained 17.7% of Step 1 and 12.0% of Step 2 CK variance, with top/bottom MCAT ranges predicting top/bottom USMLE ranges. These associations justify direction and relative size of the adjustments below — they were not used to derive precise coefficients.

The mapping tables (exactly what the tool does)

Step 1 pass likelihood = 89% (national U.S. MD first-time pass rate, 2024) + MCAT adjustment + preclinical adjustment, clamped to 72–98%, displayed as ±4 points:

Input	Adjustment
MCAT 472–500	−8
MCAT 501–507	−4
MCAT 508–512	0
MCAT 513–517	+2
MCAT 518–521	+3
MCAT 522–528	+4
MCAT not entered	0
Preclinical: struggled	−14
Preclinical: below average	−7
Preclinical: about average	0
Preclinical: above average	+4
Preclinical: excelled	+6

Projected Step 2 CK range = 249 (national mean, 2023–24) + MCAT offset + preclinical offset, clamped to 220–270, displayed as a ±10-point range rounded to the nearest 5:

Input	Offset
MCAT 472–500	−15
MCAT 501–507	−8
MCAT 508–512	0
MCAT 513–517	+5
MCAT 518–521	+8
MCAT 522–528	+10
MCAT not entered	0
Preclinical: struggled	−12
Preclinical: below average	−6
Preclinical: about average	0
Preclinical: above average	+6
Preclinical: excelled	+10

Limitations — read before acting on this

These are illustrative adjustments, not model coefficients. The numbers above are expert-constructed from benchmark levels and published effect sizes — they did not come out of a regression fitted to student data.
Group statistics don't predict individuals. A correlation across thousands of students says almost nothing about one student. Your study habits, dedicated-prep quality, test-day performance, and your school's curriculum matter enormously — none of which this tool sees.
The literature is mostly from the scored era. Most MCAT–Step 1 prediction research was published when Step 1 had a three-digit score; the pass/fail era is still young.
Institutional findings aren't individual ones. One cited study (PLOS ONE, 2019, 89 schools) found strong MCAT–USMLE correlations at the school level — that's about schools, not about you, and it's presented here as context, not as part of the math.
Single-study figures need care. The CBSE ≥53 → 96.4% pass figure comes from one school's 172 students (conference presentation, not a peer-reviewed paper). Directionally useful; not a national standard.
This is not advising. Only your school's academic support team — which sees your actual exam data — can assess your readiness. If your estimate is below the national average, that team is the right next stop.

Sources

USMLE Performance Data — NBME (usmle.org): Step 1 first-time pass rates 2021–2024 (U.S. MD 95/91/90/89%; U.S. DO 94/91/86/86%); Step 2 CK mean 249, first-time pass 98% (MD) / 96% (DO).
AAMC MCAT Validity Research Summary: MCAT scores strongly predict preclerkship, Step 1, clerkship-exam, clerkship-GPA, and Step 2 CK performance (large median correlations across schools).
Gullo CA, McCarthy MJ, Shapiro JI, Miller BL. "Predicting Medical Student Success on Licensure Exams." Medical Science Educator 2015;25:447–453. (abstract) MCAT + math-science GPA: adjusted R² 11.7% for Step 1; +first med-school exam → 27.9%; +year-2 exams → >50%; preadmission data relatively poor predictors.
Gauer JL, Jackson JB. "Do MCAT scores predict USMLE scores? An analysis on 5 years of medical student data." Medical Education Online 2016;21:31795. (abstract) MCAT components explained 17.7% of Step 1 and 12.0% of Step 2 CK variance; weak-to-moderate associations.
Purkiss J et al. "Examination of Medical College Admission Test Scores and US Medical Licensing Examination Step 1 and Step 2 Clinical Knowledge Scores." JAMA Network Open 2021. (full text) MCAT positively associated with Step 1 and Step 2 CK scores for all students (old MCAT scale — cited as supporting evidence only).
Institutional-level context: PLOS ONE 2019, cross-sectional study of 89 U.S. allopathic medical schools — institutional MCAT correlated r=.84 with Step 1, r=.62 with Step 2 CK. (article) Not used in the tool's math (school-level, not individual-level).
Colón Ortiz, IAMSE 2024 conference presentation (n=172, single school): CBSE ≥53 → 96.4% probability of passing Step 1. Older correlations: CBSE–Step 1 r ≈ 0.69–0.71.