MCAT Score Conversion: Section Scores and Percentiles

An MCAT score goes through two steps: first your raw score, the number of questions you answer correctly in each section, then a scaled score between 118 and 132 per section, and 472 to 528 in total. The catch is that there is no single official chart that maps a raw score to a scaled score, because the AAMC adjusts the conversion for each version of the test. This guide explains the conversion, the equating behind it, and how to read the section scores and percentiles that come out of it.

The exam in numbers: sections, questions, time

The MCAT has four scored sections, 230 questions in total, with 6 hours and 15 minutes of testing time and about 7 and a half hours of total seated time including breaks. The table below gives each section’s full name, its question count, and its time limit.

There is no penalty for wrong answers, and the exam is not adaptive: every question counts the same, and a blank counts as wrong, so guessing is always better than leaving a question unanswered. The score you receive is five numbers: one for each section, from 118 to 132, and a total from 472 to 528. The sections are scored independently, which is why a strong total can hide a weak section, and why the section scores deserve their own reading.

Raw to scaled in three steps

The AAMC scoring process has three steps. First, the number of correct answers is counted separately for each section: that is the raw score, a count out of 59, 53, 59, or 59. Second, each number correct is converted to a scaled section score from 118 to 132. Third, the four scaled scores are added to produce the total, from 472 to 528. The first step is counting; the second is the conversion this guide explains; the third is addition.

The conversion in step two is where the subtlety lives. A raw score of 45 out of 59 in one section does not convert to the same scaled score as 45 out of 59 in another, because the sections differ in difficulty, and it does not convert to the same scaled score on two different exam dates, because the forms differ in difficulty. The scaled score is designed to mean the same thing regardless of which form you took, and the conversion is what makes that true. The raw count is therefore not the score, and comparing raw counts across sections or across dates is comparing different things.

Equating: why there is no fixed chart

Equating is the statistical process that adjusts for differences in difficulty between exam forms, so that the same scaled score means the same thing no matter which day you tested. Each form contains a set of anchor questions that have appeared on earlier forms, and the performance on those questions tells the AAMC how hard the new form is relative to the old ones. A form that proves harder gets a more generous conversion; a form that proves easier gets a stricter one. The adjustment is small, usually a point or two at any given raw count, but it is real, and it is the reason no fixed chart exists.

The practical consequence is that any raw-to-scaled conversion you find online, including the estimator on this page, is an approximation, not the AAMC’s conversion. The estimator is transparent about its method and labels its output an estimate, and it is useful for tracking progress across practice tests taken under similar conditions, because the same approximation applied consistently shows whether you are improving. It is not useful for predicting an exact official score, and it should never be quoted as one.

The estimator on this page: how it maps percents to scores

The estimator on this page works from percents, not from raw counts directly. Each section’s raw count is converted to a percent correct, and the percent is mapped to a band score: 0 percent maps to 118, 50 percent maps to 123, and 100 percent maps to 132, with the mapping rising smoothly between those points. The defaults of 45, 40, 47, and 50 correct give sections of 128, 128, 129, and 130, for an estimated total of 515, which sits at about the 91st percentile under the current AAMC ranks.

The mapping is a transparent, repeatable approximation, and its limits are printed on the page: real forms are equated separately, only an official AAMC score report is official, and the estimate is a progress marker, not a prediction. Used within those limits, it is a genuinely useful tool: take the same practice test conditions, run the same estimator, and watch the estimated total move. Used beyond them, as a promise of an official score, it becomes a source of false confidence or false discouragement.

Section percentiles: the same score means different things

Section scores have their own percentile ranks, and the same number means something different in each section, because the sections differ in how test takers distribute across them. The table below gives the section percentile ranks in force from May 1, 2026 to April 30, 2027, at two scores applicants most often ask about.

Read the table by column, not by row: a 129 in CARS, at the 95th percentile, is a stronger result than a 129 in Psychological, Social, and Biological Foundations of Behavior, at the 85th, even though the number is the same. The reason is that test takers score higher on average in the behavioural sciences section, so a 129 there is less distinctive than a 129 in CARS, where scores spread wider. Section balance should therefore be judged in percentiles, not in raw section points: two sections a point apart can be ten percentile points apart, and that is the difference schools notice.

Worked example: reading a practice result

Take a practice test where you answer 45 of 59 in Chemical and Physical Foundations, 40 of 53 in CARS, 47 of 59 in Biological and Biochemical Foundations, and 50 of 59 in Psychological, Social, and Biological Foundations of Behavior. As percents, those are 76%, 75%, 80%, and 85%. The estimator maps them to section scores of 128, 128, 129, and 130, for an estimated total of 515, at about the 91st percentile. The reading is: a balanced profile, slightly stronger in the behavioural sciences, with no section lagging, and a total above the 2025-26 matriculant mean of 512.1.

Now change one number: suppose CARS drops to 30 of 53, which is 57%, while the other sections stay the same. The estimated CARS score falls to around 124 or 125, and the total falls to around 511 or 512, but the more important change is the balance: the profile now has a section four or five points below the others, and that gap, visible in percentiles as well as points, is what an admissions reader would ask about. The fix is targeted CARS practice, not more total-score practice, because the total will not move much while the weak section stays weak.

What schools see: every score, every section

The AAMC reports every score. Since April 2003 it has followed a full-disclosure policy, and your scores are pulled into your AMCAS application automatically on the release date. Schools see all of your scored attempts within the recent window, and each school decides which ones it will consider; most use the most recent, or the highest, or an average, according to their own policy, which the MSAR database records. A retake that improves the score helps; a retake that lowers it is visible too, which is why retaking only makes sense when practice testing supports a higher, more likely result.

Scores do not expire in the AAMC’s records, but medical schools generally accept scores dating back two or three years, and the window is measured differently by different schools: some count back from the year you apply, others from the date you would enroll. Scores are released 30 to 35 days after the exam, by 5 p.m. Eastern on the scheduled release date, and the AAMC publishes the exact release date for each exam at the start of the testing cycle. The attempt limits are three times in a single testing year, four times across two consecutive testing years, and seven times in a lifetime, counting from April 2015, with voided exams and no-shows counting toward every limit.

Mistakes, and a routine that always works

The commonest mistake is treating a raw count as a score: 45 correct is not a score, and 45 correct in CARS is not the same achievement as 45 correct in Chemical and Physical Foundations, because the sections have different question counts and different difficulties. The second mistake is quoting an estimated total as an official score, which the estimator’s labelling warns against. The third mistake is comparing percentiles across years without noting the period: a 510 was the 82nd percentile in 2017-18 and the 79th in the current table, and the difference is the table, not the test taker.

The routine that works is short. One, take practice tests under timed conditions and record the raw counts per section. Two, convert them with the same estimator every time, and track the estimated total and the section balance, in percentiles as well as points. Three, read the weakest section in percentile terms and target it specifically. Four, plan the real exam around the score-release calendar, 30 to 35 days, and the attempt limits, three in a year, four in two, seven in a lifetime. Five, rely only on an official AAMC score report as an official score, and on the MSAR database for what each school expects.

Frequently asked questions