The Optiver online assessment: 80 in 8, NumberLogic, and the Zap-N games

"Optiver 80 in 8" might be the most-searched phrase in quant-trading prep, and most of what ranks for it is either marketing or unsourced folklore. This guide does it differently: every claim below is labeled as official (from Optiver's published materials), candidate-reported (from public first-person accounts, with dates), or third-party (from prep guides, which describe formats in detail but are not primary evidence). Where those disagree, we show the range instead of picking the flattering number.

Last reviewed 2026-08-16. Optiver's exact assessment set varies by role, office, and recruiting cycle, so treat this as a well-sourced map, not a guarantee.

What Optiver says officially

Optiver's own materials describe the online assessment stage without detailing its components. Two officially sourced facts are worth planning around:

• The firm's campus-recruiting guidance says the assessment evaluates quick problem solving and logic, and that practicing mental math and strategy games can help. (Official: Optiver US campus recruiting FAQs.)
• Assessments can only be attempted once every 8 months, with no resets or retakes within that window once an attempt has been started. (Official: Optiver Europe recruitment FAQ, published June 2026.)

That second point is the strategic one. You get one shot per cycle, so the time to discover that negatively-marked arithmetic changes how you should answer is in practice, not in the real attempt.

The arithmetic screen ("80 in 8")

The famous one. Candidate accounts going back over a decade describe a short, heavily speed-weighted arithmetic test with negative marking: a 2012 first-person account describes "lots of easy arithmetic, but negatively marked", with commenters reporting 80 questions in 8 minutes and a passing bar in the mid-50s. (Candidate-reported, 2012.) Third-party prep guides describe the same format today: 80 questions in 8 minutes, roughly a point lost per wrong answer, and a passing score above 55. (Third-party.)

The tactical consequence of negative marking: skipping is free, guessing is not. Candidates who treat it as an accuracy test under time pressure, skipping anything they cannot do instantly, consistently report better outcomes than candidates who race.

Train it in the real shape: our Optiver-style 80-in-8 preset runs 80 questions in 8 minutes with penalty scoring and practice benchmarks. The benchmarks are our simulated bars, not Optiver's real cutoffs, which are not published.

The sequences test (NumberLogic)

Candidate reports and third-party guides consistently describe a second timed test built on number sequences: you see a partial sequence and pick the number that completes the pattern. A 2012 candidate account describes a sequences test of 26 questions in 30 minutes; current third-party guides describe 24 to 26 questions in about 25 minutes, with 5 to 7 visible terms and five answer options. (Candidate-reported plus third-party; the counts vary by source and cycle, so we quote the range.)

The pattern families that show up in practice materials are the classics: arithmetic and geometric progressions, Fibonacci-style sums, alternating and interleaved sequences, and polynomial steps. Our Sequence Trainer generates exactly these families under time pressure.

The cognitive games (Zap-N)

This is the least-documented part of the battery, so the labeling matters most here.

What is genuinely candidate-reported: the battery exists. A July 2024 quant-research intern report lists the assessment as "a sequences test, a quick multiple choice probability test, a set of 8 neuro cognitive assessment games, and a hackerrank"; an August 2023 software-engineering report describes "cognitive test styled games" in a long online assessment; and a September 2023 first-person write-up names the component "Zap-N". (All candidate-reported, with dates.)

What is third-party only: the specific games. Prep guides describe a shape task (match or classify shapes quickly under distraction), a digit-recall task (a sequence flashes, you re-enter it in order, sometimes reversed), and a code-comparison task (find the exact replica of a short character code among lookalikes), among others. We have not found verifiable first-person accounts describing these mechanics, one prep vendor openly notes its information is aggregated from candidate write-ups and "not employer-confirmed", and the game set appears to vary by cycle. Treat the specifics as indicative.

We built two practice games modeled on this reported battery, and we label them the same way on their own pages: Shape Match trains rapid same-or-different perception (find the identical figure among rotated, mirrored, and subtly edited lookalikes, with penalty scoring), and Memory Recall trains ordered sequence recall (a symbol sequence flashes, you rebuild it, and the sequence grows each level). They are our own designs, not replicas of any firm's test.

Putting it together

If you are preparing for an Optiver-style screen, the honest training order is:

1. Arithmetic under penalty scoring first. The 80-in-8 preset is the highest-leverage hour you can spend; learn what your instant-answer threshold is and skip everything above it.
2. Sequences second. The Sequence Trainer until the classic families are recognition, not derivation.
3. The aptitude games as a warm-up habit. Shape Match and Memory Recall are short by design; a run of each before arithmetic practice mirrors the variety of the reported battery.
4. Then the full sitting. The quant trading OA simulation chains a speed-math section and a probability sprint at real pace, forward-only, the way an actual assessment feels.

The Optiver firm guide keeps the rounds, focus areas, and sources in one place, with a badge on every claim.

Sources, confidence, and corrections

Built from Optiver's published recruiting materials (US campus FAQs; Europe recruitment FAQ, June 2026), public first-person candidate reports (August 2012 blog account; Blind reports from August 2023 and July 2024; a September 2023 Medium write-up), and third-party prep guides for format detail, all accessed 2026-06-07 or 2026-08-16. Formats vary by role, office, and cycle; where sources disagree we quote ranges. Our full sourcing and labeling rules are public at /methodology.

QuantReady is an independent prep platform and is not affiliated with, endorsed by, or sponsored by Optiver or any other firm. All trademarks belong to their respective owners. If you sat this assessment recently and something here is out of date or wrong, tell us via /contact and we will review it against the sourcing standard above.

QuantReady is an independent prep platform and is not affiliated with, endorsed by, or sponsored by any firm we write about. All company names and trademarks belong to their respective owners. Read how we source our content.