Hedge Funds interview preparation

Questions

100

Traced to a firm

39

Firms

16

Updated

September 2026

Asked at

All firmsMan Group10Balyasny Asset Management7Bridgewater Associates3DED.E. Shaw3Apollo Global Management2KKR2Oaktree Capital Management2Point722SCSquarepoint Capital2ACAQR Capital Management1BGBaupost Group1Coatue Management1HPS Investment Partners1Northern Trust1Viking Global Investors1Wolverine Trading1

Topic

All topicsStrategy taxonomy8Stock pitch10Short selling6Portfolio construction8Risk and drawdown8Performance and alpha7Event-driven and merger arb8Distressed and credit5Fund structure and economics7Financing, NAV and operations6Compliance and research process5Quant and systematic6India and Category III AIFs5Career and fit11

Level

AnyCoreIntermediateHard

Type

AnyTechnicalMarket viewBrainteaserCaseFit

1. 047How would you evaluate a track record you were thinking of allocating to?Performance and alphaHardsuperdayFund of fundsMulti-manager platforms

   Say this

   Work out what generated the return, whether it is repeatable at the size they want to run, and whether the business around it can survive a bad year. Return level is the least interesting thing on the page; the interesting things are attribution, capacity and operations.

   Then walk it

   1. First, decompose. Factor regression on the monthly series plus position-level attribution if they will share it. A track record that is 70 percent explained by long momentum and short volatility is an expensive way to buy two factors.
   2. Second, ask what the sample contains. Which regimes did it live through? A book started in 2019 has seen a crash and a violent recovery; a book started in 2023 has seen one direction. No stress event in the sample means the tail is unmeasured.
   3. Third, concentration of the result. If the top three positions made the whole number over five years, the process claim is much weaker than the return suggests.
   4. Fourth, capacity. What was AUM through the period, how liquid were the positions relative to size, and what does the expected return look like at three times the assets? Almost every disappointing allocation is an asset growth story.
   5. Fifth, the people and the business. Is the performance attributable to one person, what is the team turnover, what is the fee and expense load, and how much of the manager's own money is in the fund?
   6. Sixth, operations, which is where allocators actually lose capital. Independent administrator, audited by someone real, who marks illiquid positions, what are the gates and lock-ups, and what is the history of using them. Then the terms question: if I want out in a bad market, can I get out?

   Where candidates lose it

   Focusing on returns and Sharpe. Those are the inputs to the question, not the answer. The differentiated parts are capacity at future AUM and the operational due diligence, and a candidate who mentions the administrator, the auditor and who marks the book sounds like they have sat in an allocator's seat.

   Expect next

   • What single question would you ask the manager?
   • How would you test capacity?
   • What would make you decline a fund with excellent numbers?
2. 051How do you assess deal-break risk?Event-driven and merger arbHardsuperdayMerger arbitrageEvent-driven

   Say this

   By working through the conditions in the merger agreement one at a time and asking which one could actually fail. In practice almost all breaks come from four places: antitrust or regulatory, financing, the shareholder vote, or a material adverse change claim by a buyer who wants out.

   Then walk it

   1. Regulatory is the biggest and the slowest. Overlap between the parties, market share in the relevant definition, which agencies have jurisdiction, whether a second request or a phase two review is likely, and whether remedies are available. Cross-border adds Chinese and European approvals, which have their own political weather.
   2. Financing next. Is it fully committed, is there a financing condition, is there a ticking fee, and has the credit market moved against the buyer since signing. A buyer whose debt got 300 basis points more expensive has an incentive to find a problem.
   3. Then the vote. Who owns the target, are there activist holders arguing the price is too low, is a proxy adviser recommending against, and is the premium defensible against the unaffected price.
   4. Then the contract itself, which is where the real work is. Read the definition of a material adverse effect and the carve-outs, look at the outside date and extension mechanics, the break fee in both directions, and whether there is specific performance.
   5. Then read the incentives. Strategic buyers close; the risk is regulatory. Sponsor buyers have financing risk and a history of renegotiating price when the world changes. A buyer who has walked before is a different underwriting.
   6. Then price the downside honestly. Undisturbed price, adjusted for how the market has moved since, plus the chance of another bidder. And say the base rate: historically around 5 to 8 percent of announced deals break, so any model implying a 1 percent break probability is wrong.

   Where candidates lose it

   Answering with 'regulatory risk' and stopping. Merger arb is a documents business. Naming the material adverse effect definition, the outside date, the break fee and whether specific performance is available is what distinguishes someone who has read an agreement from someone who has read a headline. Also know the historical break base rate.

   Expect next

   • What is in a typical MAE carve-out list?
   • Would you rather own a spread with a strategic or a sponsor buyer?
   • What do you do when the spread widens on news you already knew?
3. 053A deal you own gets a second request from the antitrust authority. Walk me through what you do.Event-driven and merger arbHardsuperdayMerger arbitrageEvent-driven

   Say this

   First reprice the trade rather than react to the print. A second request lengthens the timeline and raises break probability, so the spread should widen; the question is whether it has widened by more or less than the new facts justify. Then decide whether the position is still correctly sized.

   Then walk it

   1. Step one: re-derive the implied probability from the new spread. If the spread went from 3 percent to 9 percent and the undisturbed downside is 20 percent, the market is now implying a much higher break risk. Compare that with your own estimate.
   2. Step two: re-underwrite the substance. What is the theory of harm, is the overlap horizontal or vertical, how large is the combined share in the market definition the agency will use, and are divestiture remedies plausible? Base rates matter: most second requests still end in completion, historically the large majority.
   3. Step three: reset the timeline. A second request typically adds six to twelve months, so the annualised return on the remaining spread can actually fall even as the gross spread widens. Recompute it, because that is where people fool themselves.
   4. Step four: check the agreement. Is the outside date far enough out to survive the review, who bears the obligation to litigate, and is there a reverse break fee if the buyer walks on regulatory grounds.
   5. Step five: size. Higher variance and a longer hold means less capital, not more, unless my own probability estimate is genuinely above the market's. And I would check what else in the book has the same regulatory exposure, because arb books accumulate correlated antitrust risk without noticing.
   6. Then the honest self-check: am I adding because I have new information, or because the position is down and the spread looks attractive? The second is how merger arb books turn a break into a disaster.

   Where candidates lose it

   Automatically adding because the spread widened. Widening on genuine new information is not an opportunity, it is a repricing. Also, forgetting that a longer timeline can reduce the annualised return even when the gross spread doubles. Do that arithmetic out loud and check correlated regulatory exposure across the rest of the book.

   Expect next

   • What proportion of second requests end in a block?
   • Who pays the reverse break fee and when?
   • How would you hedge regulatory risk across the whole book?
4. 057What is the fulcrum security, and how do you find it?Distressed and creditHardsuperdayDistressed debtSpecial situations

   Say this

   The fulcrum is the most senior claim that does not get paid in full, so it is the layer that converts into the equity of the reorganised company. You find it by valuing the business, then walking the capital structure down in priority order until the value runs out. Whichever tranche is sitting where the money stops is the fulcrum.

   Then walk it

   1. Worked example. Enterprise value 800. Secured bank debt 500, senior unsecured 400, subordinated 200. The banks are covered in full, the seniors receive 300 against 400 claims, so they recover 75 cents and take the new equity. Senior unsecured is the fulcrum; the subs and old equity get nothing or a nuisance tip.
   2. So the first job is the enterprise value, and the whole answer hinges on it. Use a multiple on normalised through-cycle EBITDA plus a liquidation floor on hard assets, and be explicit that you are valuing a restructured business without the current debt burden.
   3. Then build the waterfall properly, which means reading the documents. Structural seniority from where debt sits in the group, collateral and whether the lien is perfected, guarantees from operating subsidiaries, intercompany claims, and any leakage from drop-down or J. Crew style transactions that moved assets away from lenders.
   4. Then add the claims people forget: DIP financing, which is super-priority; administrative and professional fees, which in a long case are enormous; pension deficits; tax claims; and rejected lease and litigation claims that crystallise in the process.
   5. Why it matters: owning the fulcrum means owning the equity upside for a debt price, and it gives you a seat at the negotiating table because your class has to vote on the plan. That control is often worth more than the recovery arithmetic.
   6. The honest limitation: the fulcrum moves. A change in the EBITDA estimate or the exit multiple of one turn can shift it a whole layer, and the process itself can reallocate value through negotiation rather than arithmetic. So I would buy the fulcrum with a margin of safety, or buy the layer just above it and give up some upside for structural protection.

   Where candidates lose it

   Identifying the fulcrum from the capital structure table without an enterprise value. The fulcrum is defined by where the value breaks, so no valuation means no answer. Second trap: ignoring administrative costs and DIP priority, which routinely push the break a layer higher than a clean model suggests.

   Expect next

   • What happens to the fulcrum if your EBITDA estimate is 20 percent too high?
   • What is a DIP loan and why does it price so well?
   • How does a drop-down transaction hurt existing lenders?
5. 060Tell me about recent trends a distressed manager would be affected by.Distressed and creditHardtechnicalOaktree Capital ManagementRisk · Los Angeles · 2022

   Say this

   Three things dominate: the maturity wall of debt raised in the cheap-money years now refinancing at much higher coupons, the rise of private credit changing who holds the paper, and much weaker documents, so lenders have fewer protections than the last cycle. Together they mean more stress with slower, messier resolution.

   Then walk it

   1. Start with rates and the refinancing wall. Loans issued at low single-digit coupons are repricing several hundred basis points higher, and for a leveraged borrower that can consume most of its free cash flow. Interest coverage, not leverage, is the binding constraint this cycle.
   2. Then private credit. A very large share of leveraged lending has moved from syndicated markets to direct lenders, so stress shows up in negotiated amendments rather than in visible secondary prices. Marks are held by the lender, which delays price discovery and makes the cycle look calmer than it is.
   3. Then documentation. Covenant-lite is standard, so there is no maintenance covenant to trip. Defaults happen at a payment date instead of early, which means companies arrive at restructuring with less value left and lenders have less negotiating leverage.
   4. Then liability management, which is the defining feature of the current cycle. Drop-downs, uptiering and other creditor-on-creditor transactions move collateral away from non-participating lenders. That makes intercreditor documents the central analytical exercise and raises the value of a blocking position.
   5. Then the composition. Stress has been concentrated in specific pockets rather than economy-wide: commercial real estate with office valuations and refinancing, some healthcare and consumer services, and highly levered businesses with floating rate debt.
   6. How that changes the job, which is the point of the question: fewer clean recoveries and more negotiation, so a manager values legal capability and the ability to build blocking stakes over screening for cheap paper. And a risk-first manager would say the honest part out loud, that spreads spent long periods too tight to be compensated for this, so patience and dry powder were the correct posture rather than forcing deployment.

   Where candidates lose it

   Giving generic macro commentary. This question is asking whether you follow the credit market specifically. The three details that land are the coverage ratio squeeze rather than leverage, private credit delaying price discovery, and liability management exercises. Refresh the numbers the week of the interview and name a real situation.

   Expect next

   • What is an uptier transaction and why do lenders sue over it?
   • Where in the market would you look for distressed opportunities today?
   • Why have default rates stayed lower than the rate move implied?

   Reported by candidates at Oaktree Capital Management (Risk, Los Angeles, 2022). Source: Wall Street Oasis.

6. 066Explain a hurdle rate, a clawback and a crystallisation period to an investor.Fund structure and economicsHardtechnicalFund of funds

   Say this

   A hurdle is the return the fund must beat before any performance fee is earned. A clawback returns fees already paid if later losses show they were not deserved. A crystallisation period is how often the performance fee is locked in and taken. All three exist because the performance fee is an option and investors are trying to make it less of one.

   Then walk it

   1. Hurdle: typically cash, SOFR plus a spread, or a benchmark. With a 5 percent hurdle and a 12 percent return, the 20 percent fee applies to 7 points, not 12. Then ask whether it is a hard hurdle, fee on the excess only, or a soft hurdle, fee on everything once cleared. The difference is real money.
   2. Crystallisation: monthly, quarterly or annual. More frequent crystallisation favours the manager, because fees are locked in on a good quarter even if the year ends flat. Annual with a genuine high water mark is the investor-friendly standard.
   3. Worked example of why frequency matters: up 10 percent in the first half, down 10 percent in the second, roughly flat for the year. With quarterly crystallisation the manager has banked a performance fee on the first half. With annual, nothing is due.
   4. Clawback is more common in private funds than hedge funds, and it is the fix for that problem: fees paid on interim gains are returned if the final outcome does not support them, usually held in escrow.
   5. The equalisation problem sits behind all of it. Investors subscribing at different times have different high water marks, so funds either run separate series per subscription or use equalisation accounting with depreciation deposits. It is administratively ugly and it is why the administrator matters.
   6. The plain conclusion for an investor: the headline two and twenty tells you almost nothing. Hurdle type, crystallisation frequency, high water mark treatment and the expense load determine what you actually pay, and two funds with identical headline terms can differ by hundreds of basis points a year.

   Where candidates lose it

   Defining the three terms in isolation. The value of this answer is showing how they interact and which combinations transfer money to the manager. The hard-versus-soft hurdle distinction and the crystallisation frequency example are the two specifics that make it convincing.

   Expect next

   • Which is better for the investor, a hard or a soft hurdle?
   • Why is crystallisation frequency worth arguing over?
   • What is equalisation and why does it exist?
7. 067I would like you to evaluate my LP stake in a fund. How much would you be willing to pay for it?Fund structure and economicsHardsuperdayBGBaupost GroupEquity Hedge · Boston · 2018

   Say this

   I would start from reported NAV, then adjust it for three things: whether the marks are believable, what the liquidity terms let me do with it, and what fees I inherit. For a hedge fund LP interest that usually means paying a discount to NAV, and the size of the discount is the whole answer.

   Then walk it

   1. First ask what I am actually buying. A limited partnership interest in the fund, with its capital account, its high water mark, its lock-up status and its place in any side pocket. Not a portfolio of securities.
   2. Then interrogate the NAV. What percentage of the book is level one, exchange-priced and verifiable, versus level two and level three marked by the manager? I would take reported NAV on the liquid sleeve and haircut the hard-to-value sleeve materially, 20 to 40 percent depending on who marks it and whether the auditor tested it.
   3. Then the liquidity terms, which determine the discount as much as the assets. Am I locked for two more years, is there a gate, is a side pocket attached? Discount for the time I cannot get out, at my own required return. Two years locked at a 12 percent required return is roughly 20 percent of value before anything else.
   4. Then the fees I inherit. The seller's high water mark is a real asset to me: if the fund is below it, I get performance-fee-free return until it recovers, which is worth paying for. If the fund is at a peak, I inherit a full fee load.
   5. Then the qualitative discount: is the manager's team intact, is the strategy still in capacity, and why is the seller selling? Motivated sellers are the reason this market exists, and an LP selling because they know something is a genuine risk.
   6. Then say a number and defend it, because refusing to is the real failure. Something like: 'For a fund with 70 percent liquid marks, a one-year remaining lock and no side pocket, I would start around 85 to 90 percent of NAV, and I would go to 70 if a quarter of the book is level three.' Then name the one piece of information that would move the bid most, which is almost always the valuation policy on the illiquid sleeve.

   Where candidates lose it

   Answering NAV. If NAV were the answer there would be no secondary market. The analytical content is the mark quality, the liquidity discount and the inherited high water mark. And the interviewer here is explicitly testing whether you will commit to a price under uncertainty, so produce a number with a range and the reasoning behind it rather than more questions.

   Expect next

   • Why is the seller selling?
   • How much would you pay if a third of the book is level three?
   • How does an inherited high water mark change your bid?

   Reported by candidates at Baupost Group (Equity Hedge, Boston, 2018). Source: Wall Street Oasis.

8. 070How does margin financing on a long-short book actually work?Financing, NAV and operationsHardtechnicalPrime brokerageLong-short equity

   Say this

   The prime broker requires margin against your total positions, long and short, calculated either by a fixed rule like Reg T or by a risk-based portfolio model. Your equity supports the whole book, so the constraint on gross exposure is the margin requirement, and the cost is the spread you pay on the borrowed amount.

   Then walk it

   1. Two regimes. Reg T is rules-based: 50 percent initial margin on longs and 150 percent of the short's value including proceeds, which caps you at roughly 2 times gross. Portfolio margin or a risk-based model looks at the net risk of the whole book and can allow 6 to 8 times for a hedged, diversified portfolio.
   2. That is why hedged books get more leverage. Under a risk model, a long and an offsetting short in the same industry attract far less margin than two directional positions, so factor neutrality is rewarded by the financing as well as by the risk team.
   3. The economics: you pay a financing rate on the long borrowings, roughly the overnight rate plus a spread of maybe 40 to 100 basis points, and you receive a rebate below the overnight rate on short proceeds. On a 300 percent gross book the net financing line is a large, recurring cost.
   4. Margin is marked daily. Losses reduce equity, which raises the required margin as a fraction of what is left, so the constraint tightens exactly as you lose money. That reflexivity is the core mechanic to understand.
   5. Hence the margin spiral: losses, margin call, forced selling, more losses. It is the same mechanism in LTCM, in 2008, in the 2020 basis unwind and in Archegos. The trade did not have to be wrong for the fund to die; the financing ran out first.
   6. The defences are practical and worth naming: hold an excess cash buffer above the requirement, negotiate term financing or locked haircuts where you can, multi-prime so no single counterparty can force you, and stress test the margin requirement rather than just the P&L. Most funds stress the portfolio and forget to stress the financing.

   Where candidates lose it

   Describing leverage as a single number. The insight is that the margin requirement rises as your equity falls, so leverage is reflexive rather than static. And stress testing the margin requirement, not just the portfolio value, is the answer that sounds like someone who has watched a treasurer work.

   Expect next

   • How much gross could you run under portfolio margin versus Reg T?
   • What happened at Archegos, in financing terms?
   • How would you stress test your financing?
9. 071How would you value an illiquid position for the monthly NAV?Financing, NAV and operationsHardsuperdayDistressed debtFund administration

   Say this

   By a written policy applied consistently, not by judgement each month. Hierarchy: an observable transaction price if there is one, then broker quotes, then a model calibrated to comparable market data, with the manager's own view last. Then document it, get it reviewed independently, and disclose the level.

   Then walk it

   1. Start at the top of the fair value hierarchy. Level one is an exchange price and there is nothing to decide. Level two is observable inputs, so indicative broker quotes on a similar bond or a recent trade in the same issuer's other paper. Level three is a model with unobservable inputs, which is where the real work and the real risk sit.
   2. For a private or restructured equity stake, a defensible approach is a multiple on the latest reported earnings against listed comparables, plus an illiquidity discount, cross-checked against the last primary round or any secondary transaction.
   3. For a stressed loan, recovery analysis: enterprise value, the waterfall, then a discount rate reflecting the time and uncertainty of the process. Mark the claim, not the face amount.
   4. Governance is most of the answer. Broker quotes from at least two independent sources where possible, a pricing committee that signs off, a written valuation policy reviewed by the board, the administrator striking the NAV rather than the manager, and an annual audit that tests the level three marks.
   5. Consistency matters more than precision, and this is the part to say out loud. A defensible method applied every month is better than a more accurate method applied when it suits. Changing methodology when a mark is inconvenient is the classic abuse.
   6. Then the disclosures that let an investor judge: percentage of NAV in each level, the discount applied, and a sensitivity showing what NAV would be at plus or minus a reasonable range on the key input. If the position is large and genuinely unmarkable, the right answer may be to side-pocket it rather than to invent a number.

   Where candidates lose it

   Giving a valuation technique and no governance. The question is really about controls: who marks it, who checks the marker, and how you stop the method from changing when the number is unhelpful. Also remember to mention that side-pocketing can be the correct answer rather than marking something you cannot mark.

   Expect next

   • What if the two broker quotes differ by 15 points?
   • Who signs off on a level three mark?
   • When should a position be side-pocketed instead?
10. 074You are on an expert network call and the expert starts describing their current employer's unreported quarter. What do you do?Compliance and research processHardsuperdayMulti-manager platformsCompliance

    Say this

    Stop the call immediately, say clearly that you cannot receive that information, and end it rather than steer it. Then report it to compliance the same day, document what was said, and let them decide whether the name goes on a restricted list. You do not get to make that judgement yourself.

    Then walk it

    1. Interrupt, do not redirect. 'I have to stop you there. I cannot discuss unreported financial results for your employer.' Trying to move the conversation along while having already heard it does not help you.
    2. End the call. Continuing after a breach, even on other topics, looks like you kept fishing, and the call is recorded or logged by the network.
    3. Report it immediately and in writing to compliance, with the date, the expert, the network, and what was said. Self-reporting is the single most protective thing you can do, and delaying it is what turns a mistake into a career-ending problem.
    4. Expect the consequence and accept it: compliance will likely restrict the name, meaning nobody at the firm can trade it until they clear it. Even if you were already long, you may be frozen. That is the correct outcome.
    5. Say the structural controls that should have prevented it, because it shows you understand the framework rather than just the etiquette. Pre-approved question lists, chaperoned calls, prohibitions on speaking to current employees of public companies you cover, mandatory network training, and post-call logging.
    6. Then the honest point about incentives. The information would have been valuable, and that is exactly why the rule has to be absolute rather than a judgement call in the moment. Every insider trading case involving expert networks, including the ones that put people in prison, started with someone deciding this once was probably fine.

    Where candidates lose it

    Giving a soft answer: 'I would change the subject' or 'I would not use it in my model'. Both are wrong. Once you possess material non-public information you are restricted whether you use it or not. The three required beats are stop, end, report. And never say you would check with the PM first; compliance is the escalation path.

    Expect next

    • What if your PM already has a large position in that name?
    • What is the firm's obligation once you report it?
    • How do you structure expert calls to avoid this?

Firm tags come from public, anonymous candidate reports on Wall Street Oasis: strong signal, not sworn testimony. Firms are named as the places a question was reported, not as partners of Fin Maverick. Answers are written for this page to show how to think out loud; they are not scripts to recite.