Articles · Optiver

Optiver Quant Trading Internship 2027: Mental Math, Games, Probability, and Cutoffs

Preparing for Optiver's 2027 quant internship? Learn what is official, what candidates report, and how to train mental math, games, probability, and cutoffs.

The hardest part of preparing for an Optiver quantitative internship is not the arithmetic. It is deciding which version of the process to believe. One candidate reports mental math and probability games; another mentions an order-book simulation; a third advances after leaving part of a section unfinished.

Here is the useful answer: Optiver confirms online assessments for some 2027 quantitative internships, but it does not publish one global test sequence or a universal cutoff. Office, role, and recruiting cycle matter. Your invitation is the final authority on sections, timing, permitted tools, and deadlines.

Start with Optiver interview questions to practice the probability, expected-value, sequence, and market-making skills behind the assessment. This guide separates official information from current candidate reports, then turns both into a focused preparation plan.

Quick verdict: what should you expect?

Question	Best current answer	What to do
Is there an online assessment?	Yes for the Amsterdam 2027 Quantitative Trading Internship if an application advances. Current US candidates also report quantitative assessments, but the exact mix varies.	Read every section name and timer in your own invitation before preparing.
What topics recur?	Mental arithmetic, probability, number sequences, cognitive games, and market decisions recur in current reports.	Train speed only after your methods are reliable.
Is 80-in-8 always included?	No official 2027 page promises that exact test. It is a well-known candidate-reported format, not a universal rule.	Practice rapid arithmetic, but do not build your entire plan around one legacy name.
What is the cutoff?	Optiver publishes no universal section score or pass mark.	Aim for accurate, sustained performance and avoid inventing a target from anonymous posts.
Is a perfect score required?	No public evidence supports a universal perfect-score requirement. Recent candidates report progressing despite incomplete sections.	Keep working through the full assessment instead of treating one miss as the end.

The practical strategy: prepare for a portfolio of short quantitative tasks rather than one monolithic exam. Your result is likely interpreted with your resume and later interviews, not as a single magic number candidates can reverse-engineer from Reddit.

What Optiver officially confirms for 2027

Optiver's current US listing is titled Quantitative Intern (Summer 2027). It says candidates may be considered for either a Quant Trading or Quant Research internship based on strengths, interests, and interview performance. The role asks for mathematics, probability, statistics, logical reasoning, and programming or scripting experience.

The Amsterdam Quantitative Trading Internship (2027 Start) is more explicit about the first step. Candidates selected to proceed must complete online assessments within seven days. The ten-week internship itself includes probability, market structure, strategy, simulated trading, research, and AI-assisted problem solving.

Those official pages do not specify a global number of assessment sections, a public scoring formula, a required arithmetic score, or a guaranteed interview timeline. They also describe different eligibility rules by region. For example, the Amsterdam page lists a global eight-month cool-off after a quantitative assessment or interview, while the Shanghai role describes one application per role per season.

This is why a careful guide must say "reported" when discussing named games. Optiver can change providers, section names, and screening order without making an old candidate experience false.

What current candidates report

Recent 2027-cycle discussions repeatedly mention combinations of mental math, probability, number logic, likelihood ranking, an order-book exercise, and short cognitive games. Reports also mention names such as Beat the Odds, NumberLogic, Zap-N, and 80-in-8.

These reports are useful for identifying skill families. They are weak evidence for a universal format. A US Quantitative Intern applicant, an Amsterdam Quantitative Trading applicant, and a research candidate may receive overlapping but different evaluations.

Assessment area	Evidence level	Safe conclusion
Online assessment after resume selection	Official for the Amsterdam 2027 Quantitative Trading role	Be ready to complete it within the stated seven-day window.
Mental math and probability	Strong recurring candidate signal and aligned with official role requirements	These are high-priority preparation areas.
Sequences, likelihood ranking, and cognitive games	Current candidate reports	Practice the underlying skills, not memorized screenshots or exact game names.
Order-book or market-making game	Current candidate and interview reports	Learn fair value, spread, inventory, and belief updates.
One universal cutoff	Not publicly supported	Do not present a rumored score as an Optiver rule.

Mental math: build reliable speed

Rapid arithmetic matters because it preserves working memory for the decision. Practice addition, subtraction, multiplication, division, fractions, percentages, and decimal conversions until common operations feel automatic. Include two-digit multiplication and percentage changes, but do not confuse frantic clicking with improvement.

Use short timed sets and record both throughput and accuracy. If you answer 60 items at 80% accuracy, your effective performance is not automatically better than 48 items at 98%. First remove repeated error types, then shorten the timer.

A better drill structure

Run three rounds: two minutes at comfortable speed, two minutes at assessment pace, and two minutes slightly faster than assessment pace. Afterward, classify each miss as fact recall, carrying, sign, decimal placement, or panic. Repeat only the weakest category before taking another mixed set.

For market-style arithmetic, estimate before calculating. If 17% of 240 must be near 40, an answer of 4.08 should trigger an immediate decimal check. That habit matters more than shaving a second from an unchecked calculation.

Probability: recognize the model before calculating

Prioritize conditional probability, independence, combinations, expectation, variance, Bayes' rule, and simple stopping problems. Cards, coins, dice, and urns are surface stories; the real skill is translating them into outcomes and events quickly.

For each question, state four things mentally: the sample space, what is conditioned on, whether draws are independent, and whether a complement is simpler. Only then calculate. Many timed misses come from solving the wrong model perfectly.

Expected value questions often evolve into decisions. A positive-EV bet is not automatically appropriate if bankroll constraints, correlation, variance, or an asymmetric downside is introduced. Be ready to explain what new fact would change your choice.

Use simulation as a check, not a substitute

If coding is permitted in a later interview, a small simulation can validate an intuition. But a timed probability assessment usually rewards direct reasoning. Practice deriving the exact answer first, then use simulation during study to expose bad assumptions.

Games and order-book tasks: show controlled adaptation

Cognitive games may measure memory, attention, reaction, pattern recognition, or decision consistency. You cannot cram a new nervous system overnight, but you can improve by understanding instructions, using the practice round, and avoiding speed that destroys accuracy.

An order-book or market-making exercise is different. Begin with fair value, quote a bid and ask, and track inventory. When new information arrives, update the estimate rather than anchoring to the first quote. Explain why the spread widens when uncertainty rises and why position size should reflect both edge and risk.

Do not try to reverse-engineer personality or cognitive sections by choosing answers you imagine a trader would select. Consistency, attention, and honest decision-making are safer than a manufactured profile.

How Optiver cutoffs probably work, and what is not known

There is no official public table that says a particular mental-math total, probability score, or combined percentile guarantees an interview. Candidate posts often use the word "cutoff" for three different things: a minimum section threshold, an overall assessment threshold, or the score at which a particular applicant happened to advance.

Those are not interchangeable. Recruiting may consider section balance, role fit, office demand, resume strength, and later verification. Optiver's US listing explicitly says the process helps determine whether Quant Trading or Quant Research is the better fit, which suggests the assessment can be more than a binary pass/fail gate.

Recent candidates also report moving forward after failing to finish an order-book game. That does not prove incomplete work is harmless; it shows only that perfection is not a defensible universal requirement. Treat every section seriously, but keep going after a difficult one.

What happens after the assessment?

Current reports describe recruiter or behavioral conversations followed by technical interviews featuring probability, mental math, and market-making games. A recent US candidate described a first round dominated by probability puzzles and market-making rather than LeetCode-style coding. Another 2027 candidate received an interactive game-like cognitive interview through Optiver's own assessment platform.

These are snapshots, not a promised sequence. Prepare a concise "Why Optiver and why quantitative trading?" answer that connects your background to fast decisions under uncertainty, research, markets, or competitive problem solving. Also prepare one story about learning from a loss or incorrect hypothesis.

During a live game, narrate your model, quote, inventory, and update. Interviewers can evaluate calm recalibration only if they can follow your reasoning.

A seven-day Optiver preparation plan

Day	Focus	What to do
Day 1	Baseline and rules	Read your invitation, then complete one untimed probability set and one short mental-math benchmark.
Day 2	Arithmetic accuracy	Drill fractions, percentages, multiplication, and division; build an error log before increasing speed.
Day 3	Probability models	Practice conditional probability, combinatorics, independence, expectation, and complements under a timer.
Day 4	Sequences and cognition	Run short pattern sets and attention games; stop when fatigue begins to corrupt the data.
Day 5	Market making	Price simple dice or card contracts, quote two-sided markets, and update after new information.
Day 6	Full simulation	Mirror the section order and break policy in your invitation; record accuracy, pace, and recovery after mistakes.
Day 7	Review and logistics	Review recurring misses, rehearse motivation and learning stories, and confirm device, browser, room, and deadline.

Common preparation mistakes

Memorizing a rumored section list: section names can change. Train probability, number sense, attention, and market judgment so your preparation transfers.

Chasing speed with no error log: repeated arithmetic mistakes become faster mistakes. Diagnose them before raising the pace.

Assuming one bad section means rejection: finish the assessment. A difficult order-book game or missed item does not give you enough information to predict the outcome.

Ignoring the invitation: candidate posts cannot tell you your exact timer, retake rules, accommodations process, or permitted tools. The invitation can.

Frequently asked questions

Does every Optiver Quant Trading applicant get an assessment?

The Amsterdam 2027 posting says selected applicants proceed to online assessments. That is not the same as saying every applicant worldwide receives one. Resume screening, region, and role can affect the sequence.

Is 80-in-8 still part of the Optiver assessment?

It remains widely associated with Optiver in candidate accounts, but Optiver does not publicly promise that every 2027 applicant will receive an assessment with that name or format. Prepare rapid arithmetic without depending on an exact legacy section.

What score do I need to pass?

No universal official cutoff was found. Treat anonymous thresholds as personal observations, not policy. Aim for high accuracy across sections and complete every component you can.

Do I need finance experience?

The role pages emphasize quantitative reasoning and interest in markets rather than prior professional trading. Learn expected value, bid and ask, spread, inventory risk, and how market makers provide liquidity.

Will there be coding?

The US Quantitative Intern listing asks for programming or scripting experience, and candidates can be considered for Quant Trading or Quant Research. Coding depth can therefore vary. Be comfortable with clean Python, basic data structures, and simple simulation, but follow your invitation.

How quickly does Optiver respond after the OA?

Recent candidates report responses ranging from the next business day to several days, but Optiver publishes no universal turnaround. Record the completion date, follow any stated timeline, and continue other applications while waiting.

Final takeaway

The Optiver Quant Trading Internship 2027 assessment is best understood as a set of quantitative signals, not one rumored cutoff. Build accurate mental math, recognize probability models quickly, stay composed in cognitive games, and explain how new information changes a market decision.

The goal is not to imitate somebody else's assessment. It is to become reliable across the versions Optiver may actually send.

Sources and Further Reading

Research note: This guide was checked on August 23, 2026. Optiver can change assessments by role, office, and recruiting cycle; your invitation remains the source of truth.