They go to a question that quietly took four minutes. To a blank on a test that subtracts nothing for a wrong answer. To the choice that was built, deliberately, for the exact mistake you were about to make.
Superscore is 15,097 practice questions and seven drills for the part that isn’t knowing the answer.
Which is why another thousand practice questions does nothing about them.
They run on the same bank and change what you have to do with an item, because what is being trained is a decision.
Ten questions from one shared bank of time at the real per-question budget. Charts where the minutes went — and what share of your clock you spent on questions you got wrong.
10 questions · one clockTen seconds to call each question: do it now, flag it, or guess and go. Then live with the call. Scored on the judgment, not the answers.
12 questions · ten-second callsRule out two choices before the answer unlocks. Tells you how often you were narrowing correctly even on the ones you missed.
10 questions · no clockThe answer is given. Match each wrong choice to the specific slip that produces it — the architecture behind every distractor on the test.
8 questions · matchingNumeric choices and an unknown. Test the choices instead of solving forward, on a clock too tight to do both.
2,308 eligible itemsAlgebraic choices. Substitute rather than manipulate, and see which expression keeps up.
1,041 eligible itemsA whole section under test conditions: two modules, the real clock, flags, an eliminator, a question map, and a second module routed by how the first one went.
27 or 22 questions · timedThe bank was built so that each distractor comes from a named student mistake, recorded alongside it. That is what makes the drills possible, and it is why the app can tell you what your particular wrong answer came from rather than just showing you the right one.
Every passage is written for this app — no reproduced text, nothing scraped. Difficulty is calibrated per item on one 0–100 scale, so a rung in geometry means the same thing as a rung in vocabulary. And the score estimate is item response theory, the way the real test scores you, rather than a raw-score chart that doesn’t exist for the digital SAT.
The most common complaint about every paid app in this category is not the questions. It is the billing. So there is no subscription here, and the price inside the app is the same as the price on this page.
For comparison: a single hour with a branded SAT tutor runs $175–$200, and a full prep engagement runs $1,500–$6,000.
The web app works today, on any device, offline, with nothing to install. Leave an address and I’ll write once when the native app ships — and not otherwise.
One email, when there is something to say. No list, no sequence, no sharing.
No. Superscore is independent. SAT® is a registered trademark of the College Board and ACT® of ACT, Inc.; neither is affiliated with this app and neither endorses it. Every question and passage here is original — nothing is reproduced from a real test.
For official practice questions, you shouldn’t — Khan Academy plus Bluebook is the best free content there is, it comes from the people who write the test, and you should use it.
What it does not do is train the decisions. There is no drill there for pacing against the real budget, no scored triage, no elimination practice, and nothing that makes you name the mistake behind a distractor. That gap is the whole reason this exists.
Yes. The entire question bank ships inside the page — about four and a half megabytes, compressed — and nothing is fetched while you use it. Once the page has loaded, you can turn off the network and keep going.
Nothing, unless you type it in. There is no account, no sign-up and no tracking. Your progress is saved in your own browser. If you give an address on this page, it is used for one launch email and nothing else.
It uses the same method the real test uses — item response theory, fitted to which questions you get right rather than how many — and it reports a range because that is what the arithmetic supports.
Two honest limits: the difficulty scale is calibrated from the questions themselves rather than from how real students perform on them, and the estimate knows nothing about test-day nerves. The app says both of these on the screen where the number appears, rather than in a footnote here.
The content and the drills are built around the digital SAT, and the overlap with the ACT is large — the same algebra, the same grammar, the same distractor architecture. The pacing drills use SAT budgets. A dedicated ACT track, with its own timing and its own science section, is the next thing after the native app.