Maths SolverThe digital SAT switched from a single long maths section to a two-module adaptive format in 2024, and the new scoring model is considerably more interesting than the College Board’s official explanation suggests. The marketing version is “your performance on Module 1 determines the difficulty of Module 2.” The technically more accurate version is more useful to know if you’re sitting the exam.
This post is the technically accurate version, written for students who want to understand what they are actually being scored on.
The structure, briefly
The Math section is delivered in two modules. Each module is $35$ minutes long and contains $22$ questions, for a total of $44$ questions and $70$ minutes of Math. Most questions are multiple-choice with four options; about a quarter are “student-produced response” (you type a number into a box).
Module 1 is the same for everyone. It is a fixed mix of easy, medium, and hard questions across the four content areas (Heart of Algebra, Problem Solving and Data Analysis, Advanced Math, Geometry and Trigonometry).
Module 2 comes in two flavours: an “easier” version (more easy and medium questions, fewer hard ones) and a “harder” version (more medium and hard questions, fewer easy ones). Which Module 2 you get is determined by your performance on Module 1.
What “adaptive” really means here
The adaptive part is at the module level, not the question level. On a true item-adaptive test (like a typical online IQ test), each question you answer determines the difficulty of the very next question. The digital SAT is not like that. Within a module, the questions are fixed; the adaptation happens once, at the boundary between Module 1 and Module 2.
This matters because it means your Module 1 strategy and your Module 2 strategy can sensibly be different.
The score scale
The SAT Math section is scored on a $200$–$800$ scale (in $10$-point increments), the same as the old paper SAT. Your scaled score combines:
- which Module 2 you ended up in (easier or harder);
- how many questions you got right within each module.
The College Board does not publish the exact conversion table, and it varies slightly from one test administration to the next (a process called “equating” that adjusts for variations in question difficulty). But several broad observations are stable.
First: the maximum possible score on the easier Module 2 path is capped — usually around $580$–$620$. If you ace the easier module entirely, the highest you can score is somewhere in the low 600s, no matter how perfectly you do.
Second: scoring above about $650$ requires the harder Module 2. This is what the College Board means by “adaptive testing”: they need to see you handle harder questions before they will give you a high-end score.
Third: within the harder Module 2, the difference between an $680$ and a $750$ is usually a small number of additional correct answers. The high end of the scale is steep.
What this means for your Module 1 strategy
Module 1 is the gatekeeper. If you want a score above $650$, you have to perform well enough on Module 1 to be routed to the harder Module 2. Most analyses suggest the cutoff is around $14$–$16$ correct out of $22$ on Module 1 to land in the harder version, but this is not officially published and the exact threshold varies.
The strategic implication: on Module 1 you should prioritise getting a moderately high number of questions right, because that is what unlocks the higher score range. Skipping a hard Module 1 question that would take five minutes is often better than spending the time and getting it wrong — the time you saved lets you nail the next two medium questions, which are worth more for the routing.
Once you are in Module 2, the routing is locked in, and the strategy shifts to maximising the raw count.
What this means for your Module 2 strategy
If you got the easier Module 2: you are probably aiming to maximise the score within the cap. The easier module’s questions are mostly approachable; carelessness is the main enemy. Take your time on each question, double-check, and do not skip ahead unnecessarily.
If you got the harder Module 2: a few questions will be genuinely hard, possibly harder than anything you have seen on practice tests. The strategy is to get the medium questions right reliably and treat the hardest ones as bonuses. Do not let a single hard question drain ten minutes you could have spent securing five mediums.
Telling which Module 2 you are in: the College Board does not say explicitly, but the questions in the easier version are noticeably shorter and use simpler numbers. If everything in Module 2 looks straightforward, you are probably in the easier version. If Module 2 opens with a question that has you reading three paragraphs and extracting numbers, you are probably in the harder version.
The “cap” is the most important detail no one talks about
The College Board does not advertise the soft cap on the easier module path, but it is real and it is the single most consequential fact about the digital SAT scoring. A student who would have scored $700$ on the old paper SAT can score below $620$ on the new digital test if they botch Module 1 and get routed down. This is not a flaw in the student; it is a feature of the test design that you have to plan around.
The practical implication is that Module 1 deserves more careful preparation than its proportion of the questions would suggest. It is not just “the first half of the maths”; it is the qualifier that determines how high you can go.
Section vs total scoring
The Math section is one of two sections (Reading & Writing is the other). Each section is scored on the $200$–$800$ scale, and your total SAT score is the sum, on the $400$–$1600$ scale.
Each section runs its own independent two-module adaptive process. Doing well on Math has no effect on Reading & Writing’s adaptation, and vice versa. They are scored entirely separately.
What about wrong answers and guessing
There is no guessing penalty. Every blank is a guaranteed zero; every guess has at minimum a $25\%$ chance of being right (on the multiple-choice questions). So you should answer every question, including ones you have no idea about. This was already true on the paper SAT for the last decade, and it is still true.
On student-produced response questions, guessing is much less useful (you have to type a number, not pick from four options), but it is still better than leaving them blank.
Time per question
$35$ minutes for $22$ questions is just under $96$ seconds per question. This is enough to do all of them, but only if you do not get stuck. The main pacing rule: do the easy questions in the first $10$–$15$ seconds and bank the time for harder ones later. If you are spending $90$ seconds on a question that is clearly easy, you are slowing yourself down on the wrong type of question.
The other side: if a hard question is taking three minutes, mark it, skip it, and come back at the end if there is time. Spending five minutes on a single question and getting it wrong is the worst outcome, because you also missed the chance to do three or four other questions in that time.
The honest summary
The digital SAT’s Math scoring rewards a specific shape of performance: do well enough on Module 1 to qualify for the harder Module 2, then do as well as you can on a paper of genuinely hard questions. A student who is “great at the easy stuff and bad at the hard stuff” has a lower ceiling than they realise; a student who is “solid throughout but flashy on nothing” can do remarkably well, because the routing threshold is exactly what they hit.
The bigger lesson from understanding the scoring is that “just answer as many as you can” is the wrong mental model for the digital SAT. Module 1 has a different strategic weight from Module 2, and a slightly different approach to each makes a measurable difference. Knowing this in advance is worth several practice tests.
For the syllabus details, the SAT Math exam guide on this site breaks down the four content areas and the formulas the College Board does not give you on the reference sheet.