SAT Score Calculator

How SAT scores are calculated

Your SAT score isn't a direct percentage of questions answered correctly — it's a scaled score, converted from your raw score (the count of questions you got right) through an equating process designed to make scores comparable across different test dates and difficulty levels.

Why scaled scores instead of raw percentages? Different SAT administrations use different sets of questions with slightly different difficulty levels. A raw score of 40/44 on an easier test date isn’t necessarily equivalent to 40/44 on a harder one. The College Board’s equating process adjusts for these small difficulty differences so that a given scaled score represents a consistent level of ability regardless of which specific test date a student sat for. This is also precisely why the raw-to-scaled conversion isn’t a simple straight-line percentage — the relationship curves somewhat, especially at the very top and bottom of the scale.

No penalty for wrong answers. Unlike the older paper SAT, which historically included point deductions for incorrect guesses on certain formats, the current digital SAT counts only questions answered correctly toward the raw score. There’s no mathematical reason to leave a question blank — guessing carries no additional penalty beyond simply not gaining the point for a correct answer.

The distinction between raw score and scaled score matters most at the edges of preparation planning. A student improving from 30 to 35 correct answers on the Math section won’t necessarily see the same scaled-score jump as a student improving from 15 to 20 correct answers, because the equating curve isn’t uniform across its entire range — some regions of the raw score scale compress into smaller scaled-score movements while others expand into larger ones. This is a normal feature of how equating works, not an inconsistency, but it does mean that estimating a target scaled score from a raw score goal benefits from checking the actual conversion table rather than assuming a simple proportional relationship.

Digital SAT vs. paper SAT

The digital SAT’s defining structural change is its adaptive design — each of the two sections is split into two modules, and performance on the first module determines whether a test-taker receives a harder or easier second module. This adaptive structure is part of why the raw-to-scaled conversion isn’t perfectly linear: a strong first-module performance unlocks a harder second module worth proportionally more toward the top of the scaled score range, meaning the same raw count of correct answers can map to a different scaled score depending on which module difficulty a student was routed into. The conversion table used by this calculator reflects an approximation of the published average curve, not the exact curve for any single specific test administration, since actual curves vary slightly test to test as part of the equating process itself.

Understanding percentiles

A percentile tells you what proportion of test-takers scored at or below a given score — a 1200 at the 77th percentile means that score outperforms roughly 77% of test-takers nationally, not that the student answered 77% of questions correctly. Percentile and raw performance are related but distinct concepts, and confusing the two is one of the more common misunderstandings students have about their own results.

Percentiles are particularly useful for contextualizing a score’s actual competitiveness in a way a raw number alone can’t communicate. A 1400 might sound like “not quite perfect,” but sitting near the 94th percentile makes clear that it substantially outperforms the vast majority of test-takers — a distinction that matters when interpreting how a specific score is likely to be received in a competitive admissions context.

Percentiles published by different organizations can also differ slightly depending on which population they’re measured against — a “nationally representative” percentile (based on all students who took the PSAT, a proxy for the broader eligible population) and a “test-taker” percentile (based only on students who actually sat for the SAT, a group that already skews somewhat college-bound and academically motivated) can report different percentile figures for the identical scaled score. Neither figure is wrong; they simply describe performance relative to two different reference groups, and it’s worth noting which population a specific percentile figure is measured against before treating it as a precise, singular fact.

How college benchmarks work

The “middle 50%” figures shown for each school represent the range in which half of admitted (and typically enrolled) students’ SAT scores fall — 25% scored below the lower bound, 25% scored above the upper bound, and the middle 50% falls between them. This is a standard way schools report score data because it’s more informative than a single average, showing the actual spread among admitted students rather than collapsing everyone into one number.

“Reach,” “Match,” and “Safety” are informal, rule-of-thumb categories, not official admissions terminology, and they reflect test scores alone — actual admissions decisions weigh many other factors including GPA, coursework rigor, extracurriculars, essays, and (at many schools) an increasingly test-optional or test-flexible policy landscape. A score within a school’s reported middle-50% range doesn’t guarantee admission, and a score below that range doesn’t rule it out — these categories are a rough calibration tool, not a prediction.

Test-optional and test-flexible admissions policies add an additional layer of nuance worth understanding. Many schools that previously required SAT scores now allow applicants to choose whether to submit them at all, and at genuinely test-optional schools, students often submit scores only when doing so would strengthen rather than weaken their application relative to the rest of their profile. This means the published middle-50% range at a test-optional school reflects only the scores of students who chose to submit — a self-selected subset that may not represent the full admitted class, since students who didn’t submit scores are, by definition, excluded from that particular statistic.

Superscoring explained

Superscoring is a policy, used by many but not all colleges, where a school combines a student’s highest section scores across multiple test dates rather than requiring every section to come from the same sitting. A student who scored higher on Math in March but higher on Reading & Writing in May would have those two best section scores combined into a new, higher composite under a superscoring policy — even though neither individual test date produced that exact composite on its own.

This has a direct practical implication for retake strategy: under a superscoring policy, a student doesn’t need to improve on every section simultaneously to benefit from retaking the test — improving just one weaker section on a subsequent attempt can raise the superscored composite even if the other section’s score stays flat or even dips slightly. Policies vary by institution, though, so checking each target school’s specific superscoring policy directly (rather than assuming it applies universally) is an important step before relying on this strategy.

Some schools superscore only within the same test format — combining section scores across multiple digital SAT dates, for instance, but not mixing a digital SAT section score with an older paper SAT section score from before the 2024 transition. A student who took the paper SAT once and the digital SAT once should specifically confirm whether a target school’s superscoring policy bridges the two formats, since this detail isn’t always obvious from a school’s general admissions messaging and can meaningfully affect which retake strategy makes the most sense.

Real-world applications

Setting a realistic prep target benefits directly from understanding the raw-to-scaled relationship — knowing that a handful of additional correct answers on the weaker section can shift a scaled score meaningfully makes a prep plan feel more concrete and achievable than an abstract “improve my score” goal.

Deciding whether to retake the test is a decision that benefits from percentile context alongside the raw score change — a modest raw-score improvement near the steeper part of the scaling curve can produce a more meaningful scaled-score and percentile jump than the same raw improvement near a flatter part of the curve, which is worth understanding before committing the time and cost of an additional test date.

Building a balanced college list benefits from checking a target score against each school’s middle-50% range across a spread of schools — rather than aiming only at reach schools, comparing a current or projected score against a range of institutions’ benchmarks supports building a list with a healthy mix of reach, match, and safety options.

Evaluating whether a superscoring strategy makes sense for a specific student is another practical use — a student whose Math score is already strong but whose Reading & Writing score lags might reasonably choose to focus retake preparation exclusively on the weaker section, knowing that under a superscoring policy the stronger Math score from an earlier date will still count toward the final composite. This kind of targeted, section-specific prep is a more efficient use of limited study time than broadly reviewing both sections equally on every retake.

Common mistakes to avoid

• Treating percentile and raw percentage as the same thing. Percentile compares a score to other test-takers; it says nothing about the percentage of questions answered correctly, and the two numbers are rarely close to each other.
• Assuming every school superscores. Superscoring policies vary by institution — always verify a specific school’s policy directly rather than assuming it applies.
• Treating a middle-50% range as a hard admissions cutoff. These ranges describe past admitted students’ scores, not a guarantee of future admission decisions, which weigh many additional factors.
• Leaving questions blank on the digital SAT. With no penalty for incorrect answers, there’s no scoring benefit to skipping a question — an educated guess is always at least as good as leaving it blank.
• Assuming the conversion curve is perfectly linear. Because of the digital SAT’s adaptive module structure and equating process, raw-to-scaled conversion isn’t a straight percentage relationship — the actual curve for any specific test date may differ slightly from the approximation shown here.
• Comparing scores across the old paper SAT and the current digital SAT as if they measure identically. While the overall 400–1600 scale carries over, the digital format’s adaptive structure and shorter length mean the two versions aren’t perfectly interchangeable, even though College Board designed the transition to preserve comparability as closely as possible.
• Treating a school’s published middle-50% range as reflecting its entire admitted class at a test-optional institution. At schools where score submission is optional, the published range reflects only students who chose to submit — a self-selected group that can skew the reported figures relative to the class as a whole.