ACT to SAT Conversion

Every figure on this page is from the Guide to the 2018 ACT/SAT Concordance, published by ACT and the College Board, jointly. Produced from 589,753 students who were graduating seniors in 2017 and took both tests between February 2016 and June 2017, linked using equipercentile methods. Nothing here is interpolated, and where the official table stops — SAT 590 and ACT 9 — this page stops with it rather than extending the trend.

That matters more than it sounds. Several conversion charts online publish rows for ACT 1 through 8 that exist in no official source, and at least one well-known site has an off-by-one error in the thirties. The publishers’ own page is the thing to check us against.

Yes, and it is worth being precise about why, because both tests have changed since the table was built. The SAT went digital and adaptive in 2024, and the Enhanced ACT dropped Science from the composite in 2025–26. No updated concordance has been published for either.

ACT addressed it directly in its 2025 guidance for schools: “In addition, the current ACT-SAT concordance can continue to be used.” The same document adds that formal progress on updating the concordance has not begun, and that an update would need data from students who have taken both the Enhanced ACT and the digital SAT.

The reasoning holds at both ends. ACT equated the Enhanced ACT to the legacy test and reports that three-section and four-section composites concord point for point above a 13, with a correlation of 0.99. The College Board kept the digital SAT on the same 400–1600 scale and linked it statistically to the paper test. So both ends of the 2018 table are anchored to scales that are still in use.

ACT’s own phrasing is that it believes the concordance will hold and will validate that over time. That is a hedge, and this page keeps it as one. ACT also anticipates average composites falling by roughly 0.1 to 0.4 points under the three-section format — its own published estimate — which is small, real, and not in the table.

This is the question people actually arrive with, and the honest answer is that a concordance cannot answer it. The table says the two scores are equivalent. It does not say which test suits you.

The useful number is the one almost nobody quotes. The publishers put the standard error of an SAT→ACT conversion at ±2.26 ACT points, and an ACT→SAT conversion at ±79.57 SAT points. On a 36-point scale, two points is enormous relative to what students are deciding. A meaningful number of people really do score better on one test than the chart predicts — in both directions — and no published research tells you which group you are in.

There is one structural difference you can check against yourself rather than take on faith, and it is arithmetic from the published formats. The digital SAT gives about 82 seconds per question across its two sections. The Enhanced ACT gives about 57. The widest gap is verbal: ACT English allows 42 seconds a question against the SAT’s 71. If running out of time is your failure mode rather than not knowing the material, that difference is real and it points at the SAT.

What we will not tell you is that one test is easier. ACT’s own FAQ says questions on the two tests have very similar average difficulty and it comes down to preference — and when the publisher declines to claim an advantage, nobody else should. A timed diagnostic answers in three minutes what a conversion chart cannot answer at all.

Automatic merit scholarships are awarded off published grids, and those grids are written in bands. Cross a boundary and the award steps up; miss it by one question and it does not. The boundaries sit exactly where the concordance puts them, because the schools use the same table.

The University of Alabama’s published out-of-state grid is the clearest example. Its SAT columns are the official concordance bands, unchanged. An SAT of 1290 falls in the band paying $8,000 a year. An SAT of 1300 falls in the next band, paying $10,000 a year. Ten SAT points — one question — is $8,000 over four years, and the break is at the same place the table puts the 27/28 boundary.

That is why a converter is worth more than trivia to an SAT student. Of the 501 schools whose grids we have read and verified, 211 key their awards to an ACT score, so reading yours means converting first. Compare the grids or work out what one more point is worth.

One warning that almost nobody passes on: the concordance guide tells institutions not to superscore across the two tests, and calls combining ACT and SAT scores into one superscore strongly discouraged. Superscore within a test, never across.

All 28 rows of the official table, with the published band and the single point the publishers designate for a one-to-one comparison. Band widths are not uniform — they run three to five SAT steps, because that is how the published table is built.

The percentile column is ACT’s own, from the 2026-2027 national ranks — ACT-tested high school graduates from 2024-2026. Most conversion charts that show a percentile do not say whose, or for which cohort, and they disagree with each other by as much as ten points at the same score. A percentile without a named population is not a figure.

Verbatim from the source: “Concordance tables for the ACT Composite were derived from concordances of the ACT sum score.” Last checked against the publishers’ file on 2026-10-22.

1600 SAT is a 36 ACT, 1500 SAT is a 34 ACT, 1400 SAT is a 31 ACT, 1300 SAT is a 28 ACT, 1200 SAT is a 25 ACT, 1100 SAT is a 22 ACT, 1000 SAT is a 19 ACT.

Going the other way: a 36 ACT is about a 1590, a 34 ACT is about a 1500, a 32 ACT is about a 1430, a 30 ACT is about a 1370, a 28 ACT is about a 1310, a 26 ACT is about a 1240, a 24 ACT is about a 1180, a 20 ACT is about a 1040.

• Merit aid by ACT score — what each score is worth at 501 schools.
• ACT percentile calculator — where a composite sits nationally.
• ACT superscore calculator — within one test, which is the only way it is allowed.
• The Enhanced ACT — what changed in 2025–26 and what it means for scores.