What Counts as a Good SAT Score

The number everything else is measured from

For the class of 2025, the mean total SAT score was 1029 — 521 on Reading and Writing and 508 on Math — across just over two million test takers. That is the anchor. Every claim about a score being good or bad is a claim about its distance from it.

The standard deviation on the total was 235, which is worth holding onto because it converts point differences into meaningful ones. A 60-point improvement sounds substantial and is about a quarter of a standard deviation. A move from 1029 to 1400 is a move of more than a standard and a half, which is why it takes a year rather than a month.

What the distribution actually looks like

Percentiles get quoted loosely, so it is worth seeing the raw bands the College Board publishes. These are the class of 2025, and they settle most arguments about whether a given score is good.

Read down that list and two things stand out. The first is that 1200 is a genuinely uncommon score — roughly one test taker in four. The second is that the gap between 1400 and 1600 contains only 7% of everybody, which is why competition at the top of the scale behaves so differently from competition in the middle.

• 1400-1600 — 149,767 students, 7% of the cohort
• 1200-1390 — 357,574 students, 18%
• 1000-1190 — 554,819 students, 28%
• 800-990 — 609,159 students, 30%
• 600-790 — 281,174 students, 14%
• 400-590 — 52,472 students, about 3%

Good is a comparison, so name the comparison

The useful version of this question is not what a good score is but what a good score is for the eight or ten universities a student will actually apply to. That number is published. Every US university that completes the Common Data Set reports, in section C9, the 25th and 75th percentile scores of its enrolled first-year students.

Build the list first, look up the range for each, and the abstract question dissolves. A score at or above the 75th percentile figure is strong for that university. A score below the 25th percentile figure is not fatal, but it means the rest of the application has to carry the file.

Do this before choosing a target score rather than after. Students routinely spend six months chasing a number that was never required by anywhere on their list, and would have been better spent on the essays and the transcript.

A different definition of good: college readiness

The College Board also publishes benchmarks, which answer a different question — not whether a score is competitive but whether it predicts coping with first-year coursework. They are set at 480 for Reading and Writing and 530 for Math, and they correspond to a 75% likelihood of achieving a C or higher in related first-semester credit-bearing college courses.

In the class of 2025, 64% met the Reading and Writing benchmark and 41% met the Math one. The gap between those two figures is the single most useful fact on this page for a student planning preparation: Math is where the majority of test takers fall short, and it is also the section most likely to be read separately by the courses that a strong maths student wants.

Reading all this as an applicant from Pakistan

Percentiles are computed against US test takers, and that is not the pool a Pakistani applicant competes in. International applicants are usually considered within a separate and much smaller set of places, where admission rates run well below the headline figure a university publishes.

The practical consequence is that a score which would be comfortable for a domestic applicant is closer to a baseline for an international one. We are direct with families about that: a strong SAT is necessary at selective universities and nowhere near sufficient, and a student whose plan consists only of raising the score has an incomplete plan.