Test-Optional in 2026: How to Decide Whether to Submit Your SAT or ACT Scores to Each College

August scores are landing. Somewhere on your phone right now is a number, and attached to that number is a question that has quietly become one of the most consequential decisions in the entire application: do you send it?

For four application cycles, the default advice was comfortable and vague — “submit if it helps.” That advice is now actively unhelpful, because the policy landscape has fractured. A single college list can easily contain schools that require scores, schools that ignore them entirely, schools that use them only for merit money, and schools that will read them if you send them but promise not to punish you if you don’t. Treating all of those the same way is how strong applicants leave admissions and scholarship dollars on the table.

This is a per-college decision, not a one-time decision. Here is a framework for making it school by school, with the timeline to execute it before fall deadlines.

First, stop calling it “test-optional”

The phrase covers at least four distinct policies, and the difference matters enormously:

Test-required. Scores are a mandatory part of the file. Several highly selective institutions reinstated requirements between 2022 and 2025 — MIT, Georgetown, Dartmouth, Yale, Brown, Harvard, Caltech, and the University of Texas at Austin among them. If a school on your list is in this group, the decision is made for you.
Test-optional. You choose. If you submit, scores are read as one factor among many. If you don’t, the rest of the file carries more weight.
Test-blind (or “test-free”). Scores are not considered at all for admission, even if sent. The University of California and California State University systems are the best-known examples.
Test-flexible / partial. Scores aren’t required for admission but are required or strongly used for something else — merit scholarships, honors college consideration, course placement, or specific majors and programs.

Your first task takes about ninety minutes and is entirely mechanical: open a spreadsheet, list every college you are applying to, and put each one in a bucket. Get the policy from the college’s own admissions page for the 2026–27 cycle, not from a news article, a Reddit thread, or last year’s list. Policies have been revised mid-cycle before, and several institutions announced changes on short notice. Note the date you checked, and re-verify in mid-October.

You will likely find that the “decision” only applies to a subset of your list. That subset is where the real work goes.

The threshold rule, and why the common version is wrong

The advice you’ve heard is: submit if your score is at or above the middle 50% range for admitted students. That range — usually published as a 25th and 75th percentile — is the right data to look at, but the standard interpretation of it is miscalibrated for the test-optional era.

Here’s why. When only a portion of admitted students submit scores, the published range describes submitters only. Submitters are a self-selecting group who chose to send strong numbers. So the published range has drifted upward relative to the actual academic profile of the class. A 25th percentile of 1450 does not mean a quarter of the admitted class scored below 1450 — it means a quarter of the students who chose to submit did.

That leads to two practical rules:

Rule 1: The 25th percentile is your floor, not your target. If your score is at or above the 25th percentile of admitted submitters, submitting is generally neutral-to-positive. At or above the median, submitting is a clear positive. Below the 25th percentile, submitting usually adds nothing and can subtly cap how strong the rest of your file reads.

Rule 2: Find the submission rate, not just the range. Most colleges now publish, in their Common Data Set (Section C9), what percentage of enrolled first-year students submitted SAT or ACT scores. Search “[college name] Common Data Set 2025-2026” and read that number. If 85% of the class submitted, scores are functionally close to expected, and a non-submission stands out more. If 40% submitted, non-submission is thoroughly normalized and you should hold a higher bar for sending.

Write both numbers — the 25th percentile and the submission rate — next to each school in your spreadsheet. Those two figures resolve most cases immediately.

Four context adjustments that move your threshold

The percentile comparison gives you a baseline. Then adjust:

Your school’s grading context. If you attend a high school with heavy grade inflation, or one that the college has little history with, a strong test score is one of the few externally comparable data points in your file. It does more work than it would for a student from a well-known feeder school with a rigorous, legible transcript. Lean toward submitting.

Your intended major. A quantitative score matters more where the curriculum is quantitative. If you’re applying to engineering, computer science, physics, economics, or a direct-entry business program, a strong math section is real evidence. A 780 math with a 640 verbal is a better submission for a CS applicant than a flat 1420 built from balanced sections. Conversely, if you’re applying to a studio art or performance program with a portfolio or audition, the score is far less load-bearing.

Grade-versus-score consistency. If your transcript is strong and your score is strong, they reinforce each other. If your transcript is strong and your score is mediocre, submitting introduces a contradiction the reader has to resolve — and resolving it rarely goes your way. If your transcript is uneven but your score is excellent, submitting can genuinely help, because it argues the grades understate your capability.

In-state and public-university formulas. Some public universities use scores in explicit index formulas, in automatic-admission or automatic-scholarship rules, or in out-of-state review. If a school on your list publishes any kind of index or automatic threshold, run your numbers through it before deciding. This is one of the few places where the decision is arithmetic rather than judgment.

The cost of not submitting is not zero

The most expensive mistake in this whole process is assuming “test-optional” means “scores are irrelevant everywhere at this school.” Frequently they are not. Before you withhold, check three things:

Merit scholarships. Many institutional merit awards — and most large automatic-award grids at public universities — are still keyed to test scores, or use them as one of two or three inputs. A student who withholds a 32 ACT to a test-optional school may be admitted and simultaneously excluded from a four-year award worth tens of thousands of dollars. Check the scholarship page separately from the admissions page. They are often maintained by different offices and say different things.

Honors colleges and special programs. Honors programs, accelerated degree tracks, and combined-degree programs frequently maintain their own requirements that are stricter than general admission.

Placement and credit. Some schools use SAT/ACT subscores for math or writing placement. This is minor for admission but can save you a semester of prerequisite coursework.

Also check external scholarships. Many state, corporate, and community scholarship programs still ask for scores regardless of what colleges do.

Should you test again this fall?

If your current score is below the 25th percentile at the schools you care most about, the honest question is whether one more sitting changes the answer.

Test again if all three are true: you have a specific, diagnosable reason the last score underperformed (a timing collapse in one module, a content area you’ve since fixed, an unusually bad test day); you have four to six weeks of real study capacity that will not come out of your essays; and the schools where the score would matter are test-required or heavy-submission schools where withholding genuinely costs you.

Don’t test again if you’re chasing a 20-point gain, if the study time would come directly out of application quality, or if your list is dominated by test-blind and low-submission-rate schools. In late August, a marginal score improvement is almost always worth less than a genuinely good supplemental essay. Application quality has a much steeper return curve at this point in the calendar than test prep does.

If you do register for a fall sitting, confirm the score-release date against your earliest deadline — including Early Decision and Early Action — with at least a two-week buffer. Scores that arrive after a decision is rendered help nobody.

Superscoring and score choice: know the mechanics

Two mechanical details change the math:

Superscoring. Many colleges combine your best section scores across sittings. Where superscoring applies, an additional sitting carries less downside risk — a weaker section simply doesn’t count. Where it doesn’t, you’re sending whole sittings, and a bad day is visible.

Score choice. Both the College Board and ACT let you choose which sittings to send. But some colleges ask you to send all scores, and a few require it. Check each school’s policy; where all scores are required, the calculus for taking a low-stakes “just to see” sitting is very different.

A caution on withholding selectively. You can submit to some colleges and withhold from others — that’s how the system is designed to work. But be consistent within an institution: don’t submit scores to the scholarship office while claiming test-optional status in admissions. Files get read together.

A worked example

Maya has a 1390 (720 math, 670 verbal), a 3.9 unweighted GPA from a high school with a strong track record, and plans to major in economics. Her list of nine schools sorts like this:

Two test-required schools. She submits. No decision to make.
Two University of California campuses, test-blind. She doesn’t send; sending would accomplish nothing.
Three private test-optional schools with 25th percentiles of 1440, 1400, and 1330. She withholds at the first two — she’s below the floor — and submits at the third, where she’s comfortably above it and her math section supports the economics application.
Two public universities with test-optional admission but score-based automatic merit grids. Her 1390 clears the second tier at one of them, worth $12,000 a year. She submits to both.

Same score. Five submissions, four withholdings. Every one of them defensible on its own terms.

Your timeline from here

Late August: Build the spreadsheet. Bucket every school by policy. Pull the 25th percentile and Common Data Set submission rate for each. Decide whether a fall retest is on the table — and if it is, register now, before deadlines pass.

September: Make provisional submit/withhold decisions. Check every scholarship and honors page separately. Confirm superscore and score-choice policies.

Early October: Re-verify policies on official pages — this is your last checkpoint before Early deadlines. Send scores for anything Early Decision or Early Action, respecting the college’s receipt deadline, not just the application deadline.

Late October: Finalize Regular Decision submissions. Keep the spreadsheet current; you’ll need it again in January.

The bottom line

Test-optional did not remove a decision from your application. It moved the decision from the College Board to you, and it made the decision different at every school on your list.

The good news is that this is one of the few parts of the process that rewards ordinary diligence rather than talent or luck. Ninety minutes of research, two numbers per college, and four context adjustments will get you a defensible answer for every school — and it costs you nothing but an afternoon.

Do it this week, while the rest of your list is still in flux and before essay season swallows the calendar. Then close the spreadsheet and go write.