The digital SAT is 98 scored questions: 54 in Reading and Writing (two modules of 27) and 44 in Math (two modules of 22). There is no fixed "miss this many for a 1500," because the test is adaptive and equated: the exact number you can miss shifts with the form and with which second module you're routed to. As a rough guide, a 1500 usually means missing only a handful across the whole test, and a 1400 gives you a bit more room. But which questions you miss matters as much as how many.
First, the counts, since that's half the question. The digital SAT has 98 scored questions split into two sections. Reading and Writing is 54 questions across two modules of 27, in 64 minutes. Math is 44 questions across two modules of 22, in 70 minutes. That's it: 98 questions, about two hours and fourteen minutes of testing, plus a break.
98 questions. The number everyone actually wants, though, is "how many can I whiff?"
Why there's no fixed "miss this many" number
Every prep forum has someone insisting you can miss exactly seven for a 1500. They're guessing, because two features of the digital SAT make a fixed cutoff impossible:
- Equating. Each form is scored on its own curve so that a given ability maps to the same scaled score regardless of which questions you saw. A slightly easier form forgives fewer misses; a harder one forgives more. The raw-to-scaled table is different every time.
- The adaptive second module. Your first module routes you to an easier or harder second module, and the two paths have different score ceilings. Miss the same number on the easy path and the hard path, and you get different scores.
"You can miss exactly 7 for a 1500." Sources: a guy on Reddit. The court finds this unpersuasive.
So anyone quoting a precise miss-count for a target score is selling false precision. The real relationship is a range, and it moves with the form. (If you want the full mechanics, see how the adaptive modules work and how the digital SAT is scored.)
Realistic ranges (with the caveat attached)
With "it depends on the form" firmly in mind, here's roughly how forgiving each target tends to be across the whole 98-question test:
- 1500+: only a handful of misses, often in the low single digits per section. The margin is thin.
- 1400: noticeably more room, on the order of several misses per section.
- 1300: more forgiving still, with room for a genuine handful of misses across each section.
These are ballparks, not promises. On a given form your real number could be a couple higher or lower. Treat them as orientation, not a budget you get to spend down.
It is not only how many you miss. It is which. A rushed easy one and an unbeatable hard one are not equal in the eyes of the score.
Stop asking "how many can I miss" and start asking "which misses are avoidable." The test doesn't care whether you count your errors; it cares which ones you make. Two students miss the same eight questions: one loses them to careless slips they could have caught, the other to genuinely brutal items. Same count, completely different problem, and only one of them is easy to fix.
What to do this week
Forget the miss-budget. Take your last test and sort every wrong answer into "avoidable" (rushed, misread, careless) and "genuinely hard." The avoidable pile is your real answer to "how many can I get back," and it's usually bigger than people expect. Work that pile first: it's the cheapest way to turn missed questions into a higher score, no matter what the equating table says on test day.
This is exactly the read Forge gives you. Instead of a raw count, it watches how you work a diagnostic and separates the misses you could have caught from the ones you couldn't, across five dimensions of reasoning. So "how many did I get wrong" becomes "here's which ones to win back first." The diagnosis and the fix.
Don't count your misses. Sort them. That's where the points are.