What Is a Good SAT Score in 2026? Percentiles That Matter

Quick answer: a good SAT score is one at or above the middle range of admitted students at your target colleges. As national reference points, roughly 1050 sits near the median of all test takers, 1200 beats about three-quarters, 1350 enters the top ten percent, and 1500+ reaches the top one to two percent. “Good” is relative to your list, not to the internet.

The scoring system in five facts:

• Scale: 400 to 1600, combining Reading & Writing (200-800) and Math (200-800)
• Format: fully digital and adaptive: your second module’s difficulty depends on your first
• Length: about 2 hours 14 minutes
• No guessing penalty: never leave a blank
• Averages: the national mean composite has hovered a bit above 1020 in recent years

SAT percentiles: the reference table

Percentiles shift a point or two between years, so treat the table as a reliable map rather than a legal document. The SAT score calculator converts practice-test raw scores into this scale so your prep sessions speak the same language as the table.

What “good” means for your actual list

Colleges publish the middle 50% of admitted students’ scores, and that range is the only definition of “good” that matters:

• At or above the 75th percentile of a school’s range: your score is an asset there
• Inside the middle 50%: your score is normal for admits; other factors decide
• Below the 25th percentile: the rest of the application has to argue harder

Typical ranges by tier, painted broadly: open-enrollment and many regional publics center in the 900s to 1100s; state flagships commonly run 1150 to 1350; selective privates 1300 to 1480; the most selective handful 1450 to 1570. A 1280 is simultaneously a triumph for one list and a liability for another, which is why the school-range lookup beats every generic “is X good” answer on the internet.

Section scores matter more than students think

• Balance reads well: a 640/640 and a 720/560 share a composite; programs skim sections, and engineering admissions notice math
• College-readiness benchmarks: 480 on Reading & Writing and 530 on Math mark the levels associated with success in first-year courses; below them, expect placement testing
• Scholarship grids are often section-aware: automatic-merit tables at many publics key on composite plus section minimums
• Superscoring is common: many colleges combine your best sections across test dates, changing retake strategy entirely: a retake only needs to beat one section

Score goals that actually pay cash

The quietest use of the percentile table is scholarship math. Many public universities publish automatic merit tiers where crossing a line, often 1200, 1300, or 1400 paired with a GPA threshold, is worth thousands per year, making a summer of prep one of the highest-ROI projects available to a seventeen-year-old. Pair the score target with the GPA side of the grid, per the GPA improvement guide and the weighting explainer, because merit tables read both numbers, and the cheaper point to gain is usually whichever one you have neglected.

Retakes, timing, and when to stop

• Typical gains: a focused retake after real prep commonly adds 50 to 100 points; wild jumps happen but should not be the plan
• Best calendar: first sitting in spring of junior year, leaving fall retakes with summer prep between
• Diminishing returns are real: once you sit at or above your list’s 75th percentiles, further points buy little; the essays and grades now carry the marginal value
• Score choice: you control which dates most colleges see, so a rough first sitting costs anxiety, not admissions

Playing the adaptive format instead of fearing it

The digital SAT’s adaptive structure sounds ominous and mostly is not. Each section runs in two modules, and your first-module performance sets the second module’s difficulty and score ceiling. The strategic consequences are simple. Module one accuracy is precious: careless early errors can route you into the lower second module, capping the section, so the first module deserves your most disciplined check-your-work energy. A second module that feels harder is good news, not a crisis; difficulty is the reward for a strong start, and mid-test difficulty diagnosis (“this feels easy, I must be in the low module, panic”) is a losing game students should refuse to play. The no-penalty rule still governs everything: an answer on every question, flagged guesses included, is the floor strategy on both modules. Practice on the official digital platform so the interface, the timer, and the built-in Desmos calculator are muscle memory rather than novelties, because familiarity is worth real points on an adaptive clock.

Setting three numbers instead of one

A single target score creates only two outcomes, triumph or failure, which is terrible engineering for motivation. Set three numbers from your actual college list instead. The floor: the 25th percentile of your likeliest school, the score below which retakes are mandatory. The target: the 50th-to-60th percentile of your main cluster, where your score stops being a question. The reach: the 75th percentile of your most selective school, past which prep hours buy almost nothing. A student aiming at a 1150-1350 flagship plus one 1300-1480 selective private might set 1150 / 1300 / 1450, then let superscoring shape the retake plan: if the first sitting delivers a 680 math and 590 reading, the entire second sitting is a reading campaign. Three numbers convert an anxious abstraction into a project plan with a finish line, and the finish line is the point: past your reach number, the marginal hour belongs to essays and grades.

Where 100 points actually come from

Score improvement is unglamorous and highly repeatable. The reliable sequence: take a full timed digital practice test, build an error log that tags every miss by domain and by cause (content gap, misread, time pressure, careless), and let the log allocate the next two weeks, because errors cluster far more than students expect: three algebra topics and one reading question type typically account for most of the bleeding. Retime the weak sections in isolation, then take the next full test and repeat. Fifty to one hundred points over a summer is the normal harvest of this loop, and at many public universities that harvest crosses an automatic-merit line worth thousands of dollars a year, which makes the error log, hour for hour, one of the best-paying documents a teenager will ever maintain. What does not move scores: passive video-watching, untimed practice, and vocabulary lists divorced from passages. The loop is the whole secret, and the calculator turns each practice test into the percentile language your college list speaks.

Test-optional does not mean score-useless

The test-optional era added a strategic layer rather than removing one. At schools where submission is optional, the working rule is simple: submit when your score sits at or above the school’s published middle, withhold when it sits below, and remember that unsubmitted scores still work for you elsewhere: automatic merit grids at many publics, honors-college thresholds, and some scholarship applications read scores even when admissions offices do not require them. A strong score also quietly derisks a transcript with one rough semester, giving readers a second data point that says the rough patch was an episode rather than a trajectory, the same rehabilitation argument laid out in the GPA guide. The decision is school-by-school, made in October of senior year with real numbers in hand, which is one more reason the three-number goal framework above beats a single anxious target: it tells you in advance which schools get the score and which get silence.

Frequently asked questions

Is 1200 a good SAT score?

Nationally, clearly yes: roughly 75th percentile. For selective admissions it is context-dependent, comfortable for many flagships and below range for the most selective schools.

What is the average SAT score?

The national mean composite has recently sat a little above 1020, with the median in the same neighborhood.

Is 1400 good enough for Ivy League schools?

It is within the conversation but below most of their published middles, which cluster near 1500. At that tier every factor is a partial answer.

Do colleges see all my attempts?

Under score choice you generally select which dates to send, and superscoring schools assemble your best sections regardless.

What raw score gets an 800 on Math?

Usually at or within a question or two of perfect, varying slightly by form; the calculator shows the raw-to-scaled mapping by section.

Should I take the SAT or the ACT?

Whichever your practice tests say you score higher on after conversion, the exact decision framework in the comparison guide.

Convert every practice test in the SAT score calculator, check the other test with the ACT calculator, and the rest of the education tools keep the whole admissions math in one place.