SAT prep that tells you the truth.
Full-length adaptive practice tests. A score range that states its own confidence. And a repair plan for every point you leak — not another 10,000-question pile to wander through.
Free to start · No card required · The full core loop is free forever
A school district piloted a later start time for one semester. Attendance improved, and first-period tardiness fell across all four high schools. ______, the district voted to expand the schedule change the following year.
Which choice completes the text with the most logical and precise transition?
- AFor example
- BHowever
- CAs a result
- DMeanwhile
This trap usually comes from labeling every transition as a tone word instead of first asking how sentence two relates to sentence one.
The full loop: Today, full-length Tests, Practice, Review, Progress, Bank, Lessons, Vocabulary.
Calculus AB, Microeconomics, Statistics — labelled beta on every AP screen inside the product too, not just here.
Schools, essays, timeline, and packet. Local-first with best-effort account sync.
Question threads, four rooms, friends compared on effort — never on a score.
AI coach note and chat, built only from your own recorded evidence.
One loop. Run it until the range moves.
Most prep is a content library with a scoreboard. algₐ is a loop: test under pressure, read the result honestly, repair the exact leak, prove it held. Then again.
Sit the real thing
Timed, Bluebook-style modules — mark for review, answer eliminator, the tools you get on test day.
Get the damage read honestly
A range with its confidence, the route the test took, and every point traced to where it went.
Repair before you retest
Every miss becomes a repair target, then comes back later at higher difficulty to prove it held.
Watch the evidence move
Accuracy by skill, so the next block is picked from what you did — not from what you feel like doing.
Four questions. No score at the end.
Answer them the way you would on test day. You’ll get the read after each one — the specific trap, named. What you won’t get is a score, because four questions can’t support one and we’d rather show you that than fake it. Nothing is saved until you make a free account.
A school district piloted a later start time for one semester. Attendance improved and first-period tardiness fell across all four high schools. ______ the district voted to expand the change the following year.
The read unlocks after you commit.
Just like the real product: no answer leaks, no spoilers. Submit and algₐ shows why the answer works and the trap it would flag.
Warm-up only — four questions can’t produce an official score. Nothing here writes to an account or the algₐ Signal. Based on algₐ practice only, not an official SAT prediction.
If you tap through it, it doesn’t count.
An answer submitted faster than a human could physically read the question is not evidence of anything. Most prep tools count it anyway — which is how a score estimate ends up describing a student who was barely looking.
978 of 2,545 recorded answers — including 243 Reading answers on passages over 400 characters long.
Across 1,345 answers. That is chance on a four-option question — a coin, not a score.
Every student who was actually reading clears chance. The two populations separate at two seconds.
- Different sections, different floors
Five seconds on a Math item, twelve on Reading. The bar is what's physically impossible, not merely what's fast — an engaged student loses nothing.
- Rushed answers still need repair
They're cut from the Signal, mastery and weakness detection only. A question missed in one second still enters the review queue — arguably it belongs there most.
- Unknown time counts as considered
If we don't know how long you took, you get the benefit of the doubt. algₐ doesn't accuse a student on absent evidence.
- Excluded evidence is always disclosed
Where a number is built on fewer answers than you gave, it says so on the spot. A filtered figure that hides its filter is a lie with a decimal point.
The daily loop is free. The Test Center is the unlock.
The daily loop
FreeThe Test Center
ProTimed modules with the real higher/lower Module 2 routing
Grid-in questions, verified server-side like the real section
Section ranges, route explanation, pacing diagnosis, point-leak map
3–7 days of targeted blocks after every full test
Also in beta: AP practice (Calculus AB, Microeconomics, Statistics) and Apply — a local-first admissions workspace for schools, essays, and deadlines.
Explore admissionsYour worst unit isn’t the one to fix. The expensive one is.
Every AP prep tool ranks units by your accuracy. College Board publishes what each unit is actually worth on the paper — so algₐ multiplies your miss rate by the published weight and sends you where the points are. In Microeconomics, Unit 3 is 22–25% of the exam and Unit 6 is 8–13%. At equal accuracy, accuracy-only ranking sends you to the cheaper one.
Calculus AB, Microeconomics and Statistics. The question bank is being re-authored, so what’s sold here is the exam model — every AP screen inside the product carries the same beta label this one does.
No success stories yet. Here’s what there is.
Two modules per section, and Module 2 really does get harder or easier based on how Module 1 went.
The real shape: 54 Reading and Writing, 44 Math, including student-produced responses.
Every one carries its skill, its difficulty band, and an explanation written for the miss, not the answer.
Each weighted by College Board's own published unit weights, with the source and the date it was read.
Nothing here is a score gain or a student count — algₐ is new, and those are numbers we don’t have yet. When students agree to be quoted, their words go here, with their permission.
Free to train. Pay for pressure.
No bundles, no yearly lock-in, no upsell maze. Two plans, billed monthly, cancel any time.
The core training loop. No card, no trial clock.
Use the existing pricing flow to unlock full-length adaptive SAT tests and the full post-test report.
Asked before you ask.
Is the algₐ Signal official?
No. It's a practice-only estimate — a range with stated confidence, built from your completed algₐ work. Based on algₐ practice only — not an official SAT prediction.
What stays free?
The entire daily loop: Today, Practice, Review, Progress, Bank, and Lessons, plus a free diagnostic. Pro unlocks the full-length Test Center and the deep post-test analytics.
Do I need a card to start?
No. Create an account and train — Today, Practice, Review, Progress, Bank, and Lessons are free forever. Pro is $2.49/month (75% off launch) if you want the Test Center, and only when checkout is open.
Can I cancel?
Yes. Paid plans are billed monthly through Paddle and can be cancelled before any renewal from the billing portal.
What does “it doesn't count” actually mean?
Answers submitted faster than a human could read the question are excluded from your Signal, your mastery, and your weakness detection — five seconds on Math, twelve on Reading. They still enter your review queue, because a question missed in one second still needs repairing. Where a number leaves evidence out, it tells you so.
Is the AI unlimited on Pro?
No, and we'd rather say so than let you find out. Pro budgets are per ten minutes: 40 deeper explanations, 10 coach plans, and 20 coach chat messages. Free includes 3 deeper explanations per ten minutes and no AI coach. Every question has a written explanation on both plans.
Is the AP track finished?
No. AP is in beta and labelled that way inside the product on every AP screen. Buy Pro for the SAT Test Center and analytics; treat AP as a bonus that is still being built, not as the reason to subscribe.
Train on evidence, not vibes.
Run the first full test, or try one question right now. Either way you leave with the same thing: an honest read and a next move.
Based on algₐ practice only — not an official SAT prediction.