IVYPATH BLOG · SHSAT

← All articles

How Many Questions Can You Miss on the SHSAT and Still Get Into Stuyvesant?

If your child is preparing for the SHSAT and aiming for Stuyvesant, you've probably typed some version of this into a search bar: "How many questions can you miss on the SHSAT and still get in?" It's a fair question — and an important one for planning. But the honest answer surprises most NYC parents: there is no fixed number.

That isn't a dodge. It's a direct consequence of how the SHSAT is actually scored. Once you understand the mechanics, you'll have a much more realistic sense of the margin for error at a school like Stuyvesant — and why "just don't miss more than X" is the wrong way to think about it.

First, what's actually on the test

The test changed in 2026. Beginning with the fall 2026 administration, the SHSAT is a computer-adaptive test. It still has two sections, English Language Arts (ELA) and Math, but each section now contains 50 questions rather than 57, for 100 in total, with a standard testing time of 180 minutes. Every student answers the same number of questions and is tested on the same standards, content areas and item types.

What changed underneath is more important than the count. The test now adjusts as your child works: it opens with an item or passage set of average difficulty, and each answer moves the estimate of their ability up or down, which changes what comes next. There is still no penalty for a wrong answer, so every question should get an answer. But navigation is tighter than on the old paper exam. Math items and stand-alone ELA items cannot be revisited once answered. Inside an ELA passage set, answers can be changed until that set is submitted. Students choose which section to open first.

Curious how your child handles SHSAT-style questions under timed conditions? Our free SHSAT diagnostic gives an estimated score today; for Stuyvesant-level students, our full-length practice test is the better measure. We send it free after a 15-minute consultation.

Why there's no fixed "miss count"

This is the part most families don't realize, and the 2026 change makes it more true rather than less. Your number of correct answers is not your final score. Scoring is based on the pattern of correct and incorrect responses in relation to the difficulty of the items served, and that is then transformed into the reported SHSAT score, with the two sections combining into a composite.

Under an adaptive test, two students who miss the same number of questions can receive different scores. A student who answers difficult items correctly and slips on a hard one late is not in the same position as a student who missed easier items early and was routed to a gentler path. That is the whole point of adapting: the exam is measuring where a student's ability sits, not counting their errors. Which is precisely why a fixed "you can miss this many" number no longer describes anything real, and why any raw-to-scaled conversion chart you find online cannot apply to an adaptive administration.

A few consequences follow directly from this:

So when someone tells you "you can miss 12 and still get into Stuyvesant," they're describing one specific year's curve — not a rule you can count on.

What Stuyvesant actually requires

Stuyvesant has the highest cutoff of all eight specialized high schools, and the cutoff is simply the score of the last student admitted in a given year. It floats based on that year's applicant pool and that year's curve.

Recent Stuyvesant cutoffs have landed in the mid-to-high 550s:

Against a theoretical maximum near 700, a cutoff in the 550s can sound like there's comfortable room to spare. There isn't. Because the scale compresses heavily at the top, those scores correspond to very high accuracy.

So, realistically, how few mistakes?

While no exact figure exists, the pattern among students who score at Stuyvesant's level is consistent: they typically answer in the neighborhood of 80–85%+ of scored questions correctly. On 94 scored questions, that's roughly 75–80 correct — meaning admits often miss only somewhere in the range of a dozen-and-a-half questions across the entire exam, and the strongest Stuyvesant scorers miss noticeably fewer.

Treat that as a planning benchmark, not a guarantee. In a year with an easier form, the cutoff in raw-score terms effectively tightens; in a harder year, it loosens slightly. The honest takeaway is this: to be safe for Stuyvesant, your child should be aiming to miss as few as possible — single digits is the goal, not "how many can I afford to lose."

That's also why diagnostic accuracy matters more than chasing a magic number. The right question isn't "how many can I miss?" It's "which question types am I missing, and why?" Two students with the same raw score can have completely different paths to improvement. A free SHSAT diagnostic points to the skills where your child is losing the most points, so prep time goes where it counts.

Frequently asked questions

How many questions are on the SHSAT, and how many count?

From fall 2026 there are 100 questions total, 50 ELA and 50 Math, with 180 minutes of standard testing time. Earlier administrations used 57 per section, so material describing 114 questions is out of date. Answer everything: there is no penalty for a wrong answer.

Is the SHSAT curved like a class grade?

Not exactly. It's equated, not curved. Curving compares you to other students; equating compares your performance to the difficulty of your specific test form. This keeps scores fair across different versions, but it means the raw-to-scaled conversion changes every year.

What was Stuyvesant's cutoff score recently?

Stuyvesant's cutoff has been in the mid-to-high 550s in recent years (about 556 in 2025 and 561 in 2026), out of a composite scale topping out near 700. It's the highest cutoff of any specialized high school and shifts annually with the applicant pool.

Should I aim for a specific "safe" score?

Aim high rather than to a threshold. Because cutoffs move year to year and the scale is steep at the top, the smart target is near-perfect accuracy. Use a diagnostic to find weak spots early, then close them well before test day.

See where your child stands

A free 35-minute diagnostic that gives an estimated score next to last year's cutoffs, reviewed by a 99th-percentile tutor on a free call.

Start the free diagnostic →