Market microstructure is the one interview topic that maps directly onto the job. Probability brainteasers test how you think; microstructure questions test whether you understand what a market maker actually does all day. Prop shops and HFTs — Optiver, SIG, IMC, Hudson River Trading, Jump — lean on it heavily for trader roles, and it shows up in quant researcher loops at any firm that trades intraday. If you can explain why the spread exists with actual math, you separate yourself from candidates who memorized definitions.
Why firms ask microstructure questions
Three reasons. First, it filters for genuine interest: someone who has never wondered how a limit order book works probably hasn't thought hard about trading. Second, it connects to the sequential rounds — the same interviewer will often pivot from "what is adverse selection?" straight into a make-me-a-market game and watch whether you apply the concept you just defined. Third, the math is elementary but the reasoning is not: most microstructure questions are conditional-probability questions in disguise, which is why they pair naturally with Bayes' theorem drills.
The core toolkit
You should be able to explain, without notes, each of the following:
- Order book mechanics. Limit orders rest and provide liquidity; market orders cross the spread and take it. Matching is price-time priority at almost every major venue: better prices trade first, and within a price level, earlier orders trade first. Queue position is therefore an asset.
- Spread decomposition. The bid-ask spread compensates the market maker for three things: order-processing costs, inventory risk (holding a position you didn't want), and adverse selection (trading against someone who knows more than you). Interviews focus overwhelmingly on the third.
- Adverse selection and the Glosten-Milgrom logic. Quotes must be set so that the price you trade at already reflects the information content of the trade itself. Ask $=$ E[value $\mid$ someone buys from you]; bid $=$ E[value $\mid$ someone sells to you].
- Price impact. In Kyle-style models, informed order flow moves price linearly: $\Delta p = \lambda \cdot q$, where $\lambda$ rises with the informativeness of flow and falls with the amount of noise trading. You won't be asked to derive Kyle's model in a first round, but "what determines $\lambda$?" is fair game.
- Inventory management. A market maker skews quotes against their position — long inventory means a lower bid and lower ask — to attract offsetting flow. This is the single behavior interviewers most want to see in game rounds.
| Concept | One-line interview answer | Typical follow-up |
|---|---|---|
| Bid-ask spread | Compensation for processing costs, inventory risk, adverse selection | "Which component dominates in liquid names?" |
| Adverse selection | Your quotes get hit precisely when they're wrong | "So how should quotes respond to a fill?" |
| Price-time priority | Best price first, then earliest order | "Why is queue position valuable?" |
| Kyle's lambda | Per-unit price impact of order flow | "What happens to $\lambda$ if noise trading doubles?" |
Worked example: setting a zero-profit spread
This is the canonical question, asked in some form at nearly every market-making firm.
A stock is worth either \$110 or \$90, each with probability $1/2$. A fraction $\alpha = 20\%$ of incoming traders are informed and know the true value; the rest are uninformed and buy or sell with probability $1/2$ each. Where do you quote?
Set the ask so you break even conditional on being lifted. An informed trader buys only when the value is \$110, so
$$P(\text{buy} \mid V = 110) = \alpha \cdot 1 + (1-\alpha)\cdot \tfrac{1}{2} = 0.2 + 0.4 = 0.6,$$
and the unconditional probability of a buy is $0.2 \cdot 0.5 + 0.8 \cdot 0.5 = 0.5$. By Bayes,
$$P(V = 110 \mid \text{buy}) = \frac{0.6 \times 0.5}{0.5} = 0.6.$$
So the zero-profit ask is $0.6 \times 110 + 0.4 \times 90 = \$102$. By symmetry the bid is \$98, giving a spread of \$4 around the \$100 mid. Notice the clean general result under these symmetric assumptions: the spread equals $\alpha \, (V_H - V_L) = 0.2 \times 20 = 4$. The follow-ups write themselves: double the informed fraction and the spread doubles; add a fixed processing cost and both quotes shift outward; let trades arrive sequentially and each fill updates your posterior, so quotes should drift toward the true value — which is exactly the price-discovery story interviewers want you to tell. The probability machinery here is the same as in our expected value bank and core probability questions.
The traps that sink candidates
- Quoting the mid and calling it fair. If you quote a zero-width market, informed flow picks you off on both sides. The spread is not greed; it's the Bayesian correction for who trades with you.
- Ignoring the information in your own fills. Getting lifted repeatedly on the offer is evidence the value is higher. Candidates who re-quote the same market after three one-sided fills fail the market-making game regardless of their spread math.
- Treating last price as value. The last trade tells you where liquidity was taken, not where fair value sits — especially after your own market order impacted the book.
- Forgetting inventory. Even with zero adverse selection, a risk-averse market maker skews quotes when positioned. Saying "I'd hold the position because my price is fair" signals you've never carried risk.
- Hand-waving queue priority. If you claim you'd "just join the best bid," expect the follow-up: joining a 500,000-share queue at position 500,001 earns you the fills nobody else wanted.
Firms where this topic is a first-class round include Optiver and SIG; both routinely chain a definition question into a live quoting exercise.
Practice next
Work through our market microstructure question bank for the full set of order-book, spread, and adverse-selection problems with worked solutions, then pressure-test the concepts live in the make-a-market game or the broader trading games suite — the fastest way to make quote-skewing and inventory management instinctive before a real interview.
More topic guides
- The Airplane Seat Problem: Why the Answer Is 1/2 (Three Proofs)
- Bayes' Theorem in Quant Interviews
- Behavioral Interview Questions at Trading Firms (With Answer Frameworks)
- Coin Flip Questions in Quant Interviews
- C++ Low-Latency Interview Questions at HFT Firms
- Dice Questions in Quant Interviews
- Fermi Estimation Interview Questions at Trading Firms
- Gambler's Ruin in Quant Interviews
- The Kelly Criterion in Quant Interviews
- Linear Regression Interview Questions: OLS Assumptions, R² Traps & Regression to the Mean
- Machine Learning Quant Interview Questions: Overfitting, Cross-Validation & Feature Leakage
- The Market Making Game Interview: How to Answer 'Make Me a Market'
- Markov Chains in Quant Interviews
- Martingales in Quant Interviews
- Mental Math for Trading Interviews: Training Plan, Zetamac Benchmarks & Firm Tests
- The Monty Hall Problem — and the Variants Interviews Actually Ask
- Number Sequence Tests in Trading Interviews: The 8 Pattern Types & How to Practice
- Optimal Stopping in Quant Interviews
- Options Pricing Interview Questions: Black-Scholes, Greeks & Put-Call Parity
- Quant Interview Cheat Sheet: Probability, Markov Chains, Options & Linear Algebra (Free PDF)
- Random Walks in Quant Interviews
- Statistics Interview Questions for Quant Roles: Hypothesis Testing, MLE & p-Value Traps
- Stochastic Calculus Interview Questions: Ito's Lemma, SDEs & Brownian Motion
- Time Series Interview Questions: Stationarity, ARMA & Autocorrelation Traps
- Top 50 Quant Interview Questions (With Full Solutions)
- All guides & explainers
Frequently asked questions
What market microstructure questions come up in quant interviews?
The most common questions cover how a limit order book matches trades (price-time priority), why the bid-ask spread exists, and how adverse selection forces market makers to widen quotes. A standard worked question gives you a two-value asset with some fraction of informed traders and asks you to compute the zero-profit bid and ask using conditional expectations. Firms like Optiver, SIG, and IMC often follow the theory question with a live make-me-a-market exercise.
What is adverse selection in a market making interview?
Adverse selection is the risk that your quotes get filled precisely when they are wrong: informed traders buy from you when the asset is worth more than your ask and sell to you when it is worth less than your bid. The correct response is to set quotes at the expected value conditional on the trade happening, which mechanically creates a bid-ask spread. In the standard symmetric example, the spread scales linearly with the fraction of informed flow.
How do you answer the bid-ask spread interview question?
Decompose the spread into order-processing costs, inventory risk, and adverse selection, then show the adverse-selection component with math. Set the ask equal to the expected value given that someone buys and the bid equal to the expected value given that someone sells, using Bayes' theorem to update on the trade direction. Interviewers then typically ask how the spread changes if informed trading increases (it widens) or if you accumulate inventory (you skew both quotes against your position).
Do I need to know Kyle's model or Glosten-Milgrom for interviews?
You need the intuition more than the full derivations. For Glosten-Milgrom, be able to compute conditional-expectation quotes in a simple two-value setup, since that exact calculation is asked frequently. For Kyle, know that price impact is linear in order flow, and that lambda increases with the informativeness of flow and decreases with the amount of noise trading; full derivations only appear in research-heavy PhD-level loops.
Practice the real thing
QuantVault has 2,800+ quant interview problems with full solutions, intuition, and hints, firm-by-firm interview funnels, and an auto-graded coding judge. Start free.