What is a good SAT score

There is no cutoff that separates a good SAT score from a bad one. The two things that actually answer the question are where a score sits in the national distribution, and what the specific colleges on your list report about the students they admit.

Start with the distribution. Among the 2,004,965 students in the class of 2025, the mean total score was 1029, with a standard deviation of 235 points (2025 SAT Suite Annual Report). The section means were 521 on Evidence-Based Reading and Writing and 508 on Math. That is the current national picture as of the 2025 graduating cohort, which is the newest full-year data available in August 2026.

Where each score sits in the distribution

College Board publishes two percentile tables side by side. The SAT User Group column is based on the actual scores of students who graduated in the past three school years. The Nationally Representative Sample column comes from a research study weighted to represent all US 11th and 12th graders, including those who never sit the test (College Board Research).

The two columns answer different questions, so a score can look meaningfully different depending on which one you read.

Total score	Nationally Representative	SAT User Group
1600	99+	99+
1550	99+	99
1500	99	97
1450	99	95
1400	97	93
1350	94	89
1300	91	85
1250	86	80
1200	81	75
1150	74	69
1100	67	62
1050	58	55
1000	48	47
950	38	39
900	29	32
850	21	24
800	14	17
750	8	11
700	4	6
650	1	3
600	1-	2

Percentiles here are inclusive, meaning “at or below”. A 1200 at the 75th percentile means 75 percent of the comparison group scored 1200 or lower. The values printed as “99+” and “1-” are censored bounds as College Board prints them, not computed numbers. You can look up any score in ten point steps with the SAT percentile calculator or the full SAT percentile tables.

One important caveat: neither of those columns is the percentile printed on your own score report. That number is the All Tester Percentile, and it has no published lookup table. The difference is explained in SAT percentiles explained.

How the class of 2025 was actually spread out

Percentiles compress the shape of the distribution. The band table shows it more directly.

Total score band	Share of class of 2025
1400-1600	7%
1200-1390	18%
1000-1190	28%
800-990	30%
600-790	14%
400-590	2%

Roughly 58 percent of the cohort scored somewhere between 800 and 1190. The tails are thin in both directions, which is why small point differences near the middle of the scale move percentiles quickly and small differences near 1500 barely move them at all.

The two benchmarks College Board does publish

College Board sets College and Career Readiness Benchmarks at 480 on Reading and Writing and 530 on Math. In its own words, these “predict a 75% likelihood of achieving a C or higher in related first semester, credit-bearing college courses.”

In the class of 2025, 64 percent met the Reading and Writing benchmark, 41 percent met the Math benchmark, 39 percent met both, and 34 percent met neither. These are the only published thresholds attached to any outcome claim. They are not admissions cutoffs and no college uses them that way.

Section scores tell a different story from the total

A 1200 built from 650 Reading and Writing plus 550 Math reads differently to an engineering program than a 1200 built the other way around. Section percentiles for the class of 2025 comparison groups:

Section score	RW (User)	Math (User)
800	99+	99+
750	98	95
700	92	91
650	84	84
600	72	75
530	53	59
500	43	50
480	37	45
400	16	21

If you want to see how a change in one section moves the total, the Math section calculator and the Reading and Writing calculator work on each half separately.

“Good” is defined by your college list, not by the national table

The national distribution tells you how common a score is. It does not tell you whether a score is workable for the schools you care about, because selectivity varies enormously across the roughly 4,000 degree-granting institutions in the US.

The number that matters for any individual college is its published middle 50 percent range, which appears in its Common Data Set and usually on its admissions statistics page. That range describes the middle half of enrolled students at one school in one admissions year. A score at the 25th percentile of one college’s range can sit above the 75th percentile of another’s.

Two details make those ranges easier to read. They describe enrolled students, not applicants, so they reflect who accepted an offer as well as who received one. And at test-optional colleges they cover only the subset of students who chose to submit a score, which tends to skew the published range upward relative to the full class.

Testing policy is also part of the picture and it changes yearly. As of August 2026, Harvard, Yale, Brown, Dartmouth, Cornell, Penn, MIT, Caltech, Stanford, and UT Austin require test scores. Princeton and Columbia are test optional for 2026-27, with Princeton requiring scores from fall 2028 entry and Columbia from the 2027-28 cycle. Confirm any policy directly with the college before relying on it.

To work backwards from a target instead of forwards from a result, the score needed calculator takes a target total and shows the section combinations that reach it.

If you are comparing against an ACT score

Students often want to know whether an SAT result is in the same territory as an ACT result. The only official crosswalk is the 2018 ACT/SAT concordance, and ACT states that “the current ACT-SAT concordance can continue to be used.”

Read it with its own limits attached. That concordance was built between a retired four-section ACT and the pre-digital paper SAT, and it is published with standard errors of plus or minus 2 ACT points and plus or minus 80 SAT points. An 80 point band on a 400-1600 scale is wide enough to cross several percentile steps, so a concorded figure is an approximation rather than an equivalent score. Both directions are handled by the SAT to ACT and ACT to SAT tools, which show the error bands alongside the result.

Before you compare scores across attempts

Two further details change how a number should be read.

Superscoring means many colleges take your highest Reading and Writing score and your highest Math score across separate sittings. Policies differ by school, so check each one, then use the superscore calculator to see the combined figure a superscoring college would consider.

Practice test scores are not directly comparable to real ones. Paper practice tests use a raw score conversion table with a published lower and upper bound, and the resulting number is a range rather than a point. That mechanic is covered in the raw score conversion reference.

The distribution above is the honest answer to the question in the title. What counts as sufficient is a decision that belongs to you and the list of colleges in front of you, not to a national average.