Market Making Games in Quant Interviews: How to Prepare

The market making game is the most distinctive part of a quant trading interview, and the one candidates least know how to prepare for. It looks casual: an interviewer deals some cards or quotes some prices, and you decide whether to trade. Underneath, it is testing the exact skill the job requires, which is pricing under uncertainty while someone tries to take advantage of you.

This guide explains what these games actually test, the common formats, the concepts behind them, which firms use them, and how to practice.

What the games test

A market maker is someone who stands ready to buy and sell. They quote a bid, the price they will buy at, and an ask, the price they will sell at. The gap between them is the spread, and it is how the market maker gets paid for taking risk.

The risk is adverse selection. Some of the people trading against you know more than you do, and if your price is wrong they will trade against you repeatedly until you lose. The game is a compressed version of this: hidden information, asymmetric knowledge, and real money consequences.

Three skills are being measured. Expected value: can you compute what a position is worth with incomplete information. Risk management: do you size positions sensibly rather than betting the farm. And composure: can you keep doing both while the interviewer leans on you.

The common formats

Mental arithmetic. Many firms start with a speed math assessment. Optiver is famous for its "80 in 8" test, eighty arithmetic questions in eight minutes, and PuzzledQuant has its own version for practice. The questions are simple. The time pressure is the point. Traders do arithmetic the way you breathe, without thinking about it.

The scoring is the part most candidates only discover mid-test, and it changes how you should play. It is +1 for a correct answer, −1 for a wrong one, and zero for a question you leave blank. Skipping is free; guessing is not. Six seconds a question means the winning strategy is to move fast on the shapes you recognise and abandon anything that stalls you, rather than fighting one ugly division while the clock eats three easy ones. Candidates who guess through the back half to "finish" routinely score worse than candidates who answered sixty and left twenty blank. Public estimates put the practical bar around 55, and competitive candidates aim closer to 70.

Card games. The interviewer deals cards face up and keeps some hidden, and you trade against them. A simple version: you see one card from a deck, the interviewer sees two, and you take turns quoting a price for a card yet to be drawn. Your job is to quote a bid and ask around the fair value, then watch how the interviewer's willingness to trade leaks information about what they hold. Susquehanna (SIG) is the firm most associated with this style, and it is worth understanding why: SIG built its trading culture on poker, teaches it formally in trader training, and structures its interviews around the same "think in bets" reasoning. You do not need to play poker to get an offer there, but you do need to be fluent in pot odds, bet sizing relative to edge, and updating on what a counterparty's action reveals. SIG's final stage also tends to include a group exercise with the other candidates, often auction or bidding flavoured, where they watch how you behave with your competition sitting next to you.

Dice and expected value games. A common setup is a hidden die or sum, where you get partial information and must price a payoff. The math is straightforward expected value, but the twist is that the interviewer's trades reveal information, and you must update in real time.

Live bid-ask simulators. The most realistic format: a screen shows a price moving around, and you quote and trade continuously, managing inventory and P&L. These are usually reserved for later rounds or assessment centers at firms like Optiver, IMC, and Flow Traders.

Estimation games. Loosely related but common enough to mention: "estimate X" questions test your ability to decompose an unknown quantity, which is the same skill as pricing an illiquid asset.

The concepts to know cold

Expected value is the foundation. Every game reduces to summing probability times payoff. If you cannot do this in your head quickly and accurately, practice arithmetic first.

Fair value and the spread. The fair value of a position is its expected value. Your bid sits below fair value and your ask above it, and the width of the spread reflects how uncertain you are and how much adverse selection you expect.

Adverse selection. This is the heart of the game. If the counterparty trades aggressively at your ask, they probably know your ask is too low. You should widen the spread or move the price, not stubbornly keep quoting. The interviewers are explicitly testing whether you will be exploited.

Bayesian updating. When a trade happens, it is information. Update your estimate of the hidden state, then re price. Candidates who treat each round as independent lose quickly.

Inventory risk. A market maker who accumulates a large position is exposed if the price moves against them. Part of the skill is recognizing when your inventory is too big and quoting defensively to flatten it.

Which firms ask

Market making games are the signature of prop trading firms and market makers specifically, which makes sense: it is literally their business.

Optiver, IMC, and Flow Traders run versions of the arithmetic and live trading simulations, often as online assessments before interviews. Susquehanna leans on cards, poker, and game theory. Jane Street runs market making rounds in later interviews. DRW, Akuna Capital, and others use variants. If you are interviewing at any Dutch or Chicago market maker, assume a game is coming.

The weighting differs, and it is worth targeting your practice accordingly. Optiver and IMC filter hardest on raw arithmetic speed. Jane Street weights clean reasoning and the quality of your narration. SIG weights decision-making under uncertainty above both. Same three skills, different scoreboards.

How to prepare

Preparation has three parts: speed, math, and instinct.

For speed, do mental arithmetic every day. The 80 in 8 game is a direct replica of the assessment style, and ten minutes a day moves the needle fast. Two weeks before the real assessment, start practising the skip discipline as well as the speed: run a full timed set, and afterwards count how many of your wrong answers were questions you should simply have left blank. That number is usually the gap between your score and your target.

For the math, drill expected value problems until the computation is automatic. Work through the probability chapters of "A Practical Guide to Quantitative Finance Interviews" and then re do them faster.

For instinct, you need repetition against a counterparty, which is the hard part to arrange alone. PuzzledQuant's market making games fill that gap. The dice market maker simulates the hidden dice setup where you quote a bid and ask and trade, with your P&L tracked. The card counting game trains the expected value and hidden information reasoning behind the SIG style card games. Play them until quoting around fair value feels automatic, then widen or tighten based on what the trades tell you.

If you have a friend also interviewing, do live practice. One person deals hidden cards, the other quotes. It is awkward for ten minutes and then it is the best preparation there is.

The mindset that wins

The candidates who do well treat the game as a conversation, not a math test. They narrate their reasoning. They admit when they are unsure. They update when the interviewer trades against them and say so out loud.

The candidates who do poorly go quiet, fixate on getting the exact right number, and get exploited because they treat every round as independent.

A market making game is not really about whether you can compute expected value. It is about whether you can price under uncertainty, read information from trades, and stay composed while doing it. That is the job, and the game is the interview in miniature.

Start with the market making games, build the arithmetic habit, and you will walk into the real thing knowing what to do when the cards come out.