Ambrus Volatility Fund: Call Follow-Up (2026.03.26)

Built from Transcripts/Ambrus 2026.03.26/Ambrus 2026.03.26_transcript.txt, an Otter.ai transcript of about 74 minutes. Speaker labels are partly generic and names are often mis-transcribed: "Amherst", "embers" and "Ambros" all mean Ambrus, "QBR" means QVR, and "Robert cancel" means Robert Kanter. Speaker attributions below are inferred where the transcript is unclear. The transcript sits outside sources/. Manager statements below are what was said on the call, not verified facts. §9 checks them against sources/.

---

1. Meeting Snapshot

• Meeting topic: Introductory diligence on the Ambrus carry-neutral tail-risk strategy, covering the tail sleeve, the flow sleeve, monetization, risk, capacity and next steps.
• Participants:
  • Allocator side: VX (founder, "VS Partners"; describes a cross-border allocator to offshore hedge funds and onshore China); Speaker 3, who led the technical questions and referred to "our previous notes" from about five years ago; Speaker 2.
  • Ambrus: Speaker 1, inferred to be Kris Sidial (co-CIO) from his career history and the "Chris will take the reins" handoff.
  • Unknown Speaker opened the call and asked about timeline and ticket size, likely on the Ambrus or intermediary side. "Leo" is mentioned as an earlier contact.
• Date: 26 March 2026 per the folder and file names. The Otter summary header says "April 20, 2026", probably the export date. The call content is consistent with March 2026: it mentions "this month" market stress, Liberation Day as "last year", and a multi-strat drawdown "last March".
• Purpose of the call: The allocator wants to re-enter tail-risk allocations after exiting about two to three years ago. It wanted to understand the strategy before an expedited allocation.
• Overall outcome: The tone was positive and fast-moving. Before the call, VX had already told operations to start onboarding, which VX said skipped the firm's usual research process. Ambrus will send approved materials. The allocator is targeting deployment "within a month, if not faster". No size was committed.

2. Executive Summary

• The allocator is already onboarding Ambrus ahead of completing its own research, and aims to deploy within a month. VX said sizing is market-regime dependent. In 2020 the firm allocated about 15% to tail managers, then cut to 3%, then exited.
• Kris described Ambrus as a "trading-based" rather than "solutions-based" tail fund, using the Peak6/CTC/O'Connor prop model: hold deep OTM options and pay for them with flow trading. He named Universa, One River and 36 South as solutions-based peers.
• Unverified claims:
  • Ambrus is "the top performer in the space over the last five years".
  • It was "volatility hedge fund of the year" last year.
  • It is backed by "the largest multi-strat hedge fund in the world", which called Ambrus the only tail fund it has ever invested in.
• Economics as stated:
  • The tail book holds 7.5–15% of the fund in premium terms.
  • It burns about 1–2% a month, or 12–20% a year.
  • The flow book must therefore earn 12–20% a year to stay flat.
  • Kris said a "really bad year" (both books losing) would be about -15%, and that this has never happened.
• AUM is now $700M, with a $250M pipeline. Capacity is about **$2.0–2.25B**, after which Ambrus will hard-close. The multi-strat and a large multi-family office hold capacity rights they have signalled they will exercise. Ambrus says new investors need to be in formal ODD/IDD within about two months or risk losing capacity.
• Month-to-date (March 2026): about +4.5%, with the S&P down about 5.5% and VIX about 26. The book is long 1–1.5 year S&P downside and two-month mid-delta VIX calls. Ambrus has not monetized. Kris's base case is a 1999–2002-style grind lower with vol in the 20s to 40s, not a COVID-style crash.
• Data limits: a contract with the multi-strat restricts what Ambrus can share. A sample portfolio from just before Liberation Day (March 2025) can be shared. P&L attribution and the current portfolio were not approved by compliance.

3. Key Takeaways

• Takeaway: Manager-stated unit economics are 12–20% annual tail burn, offset by 12–20% from the flow book.
  • Why it matters: The whole case rests on a flow sleeve earning mid-teens or better gross on a small capital footprint, at $700M and rising. That is the single largest unverified claim, and attribution by sleeve is exactly what Ambrus says it cannot yet share.
• Takeaway: Capacity scarcity is being used to set the pace: a $2–2.25B hard close, capacity rights held by two anchors, and a two-month ODD/IDD window.
  • Why it matters: This creates pressure to shortcut diligence. The allocator has already skipped its usual process. Investors holding capacity rights also suggest side letters and possibly preferential terms or information rights.
• Takeaway: AUM of $700M stated on the call is about 60% above the DDQ's $437.6M firm AUM as of 1 January 2026.
  • Why it matters: Either AUM grew by about $260M in under three months (concentration and capacity-decay risk), or the figures are defined differently. This needs a dated, administrator-backed reconciliation.
• Takeaway: Risk controls in the flow book are concrete and rule-based.
  • Each trade risks 5–10 bps, with 5–10 trades open at once, for a stated worst-case day of 25–50 bps.
  • After a bad day risk is halved, halved again after a second, and switched off after a third.
  • Size is also cut when a strategy behaves outside its observed distribution, even when it is making money.
  • Why it matters: This is consistent with the DDQ's tiered drawdown controls and gives specific numbers to test against daily P&L.
• Takeaway: Key-person concentration is operational, not just intellectual. Kris said no trade reaches an exchange until he or William clicks accept.
  • Why it matters: This is a sensible safeguard, but it puts every execution through two people. It also contrasts with the DDQ, which lists the execution trader as authorized to place orders.
• Takeaway: Realized monetization was described as selective.
  • In August 2024 Ambrus sold only "a very tiny part" of the book.
  • In April 2025 the book was "up about 98%" at the peak and about 30% was monetized that month, from dislocated positions that made up less than 10% of the book.
  • Why it matters: "Up ~98%" against the reported +25.21% for April 2025 needs explaining. It may refer to the tail book only, to unrealized gains given back, or to a different measure.

4. Decisions Made

No decision on allocation size or final approval was made.

5. Action Items

6. Open Questions

• Question: How does $700M AUM reconcile with the DDQ's $437.6M (1 January 2026)? What is the net-flow history, and how large are the multi-strat's and the MFO's stakes?
  • Why it matters: It tells you about concentration, capacity usage, and whether the flow edge is being diluted.
• Question: What exactly are the capacity rights (amount, price, terms, expiry)? Do the anchors have side letters with better fees, liquidity or transparency? Is there an MFN?
  • Why it matters: It affects alignment and fairness to later investors. The multi-strat contract already limits disclosure to others.
• Question: Can attribution by sleeve (tail burn vs. flow P&L, monthly since May 2022) be shared under NDA, or verified by the administrator?
  • Why it matters: This is the core proof of the carry-neutral claim. Without it the edge can't be tested.
• Question: What did "book up about 98%" in April 2025 measure, and how much of that gain was kept versus given back?
  • Why it matters: The fund reported +25.21% that month. The gap is a test of monetization skill.
• Question: Is trade allocation across the master fund, SMAs, Medici and Sundial done after the trade or before it?
  • Why it matters: The call says "post trade allocation… pro rata" (attribution is unclear in the transcript). The DDQ says allocations are fixed before or at order entry and never reallocated afterwards [DDQ26 §9.5]. This is an ODD point.
• Question: What is the hard capacity of the flow sleeve on its own, and what evidence supports $2–2.25B given that the deck calls the edges "capacity constrained"?
  • Why it matters: At $2B, a 12–20% flow return is about $240–400M a year from short-horizon trades risking 5–10 bps each.
• Question: The call did not cover the restated 2020–May 2022 history, the 2021 change from short vega to "no short optionality", GP capital, fees, the redemption fee, or team depth.
  • Why it matters: These remain the main gaps from the first-pass note (Ambrus_Volatility_HF-Diligence-2026-10-20.md).
• Question: Is the "top performer over five years" claim based on live or hypothetical returns, and against which peer universe? What is "volatility hedge fund of the year", and who awarded it?
  • Why it matters: Five years reaches back into the period the deck labels hypothetical.

7. Risks, Blockers, or Watchouts

• Process risk on the allocator side: onboarding began before research was finished, under a manager-imposed capacity deadline. Combined with restricted data, this is how diligence gets compressed.
• Information asymmetry: a contract with the multi-strat limits what Ambrus can share, and attribution and current holdings are not approved. Anchor investors may be getting transparency that others are not.
• Conflicts with sources/: AUM ($700M vs. $437.6M), trade allocation (post-trade vs. at order entry), and who can place orders (two co-CIOs vs. also the execution trader) all differ. See §9.
• Narrative drift on the flow book: Kris said the trades are "the same exact things" as five years ago and only the messaging changed "to protect our footprint". The 2021 materials described capped short-vol VIX contango and short VXX call spreads [Pitch21 pp.20, 23, 27], while the 2025 deck says "no short optionality" [Deck25 p.27]. Whether the risk profile really changed needs confirming.
• Strategy change on the tail book: the allocator's 2021 notes recorded a put basket on about 40 single names. Kris now says the tail book avoids single-name puts and uses only VIX, S&P and sector ETFs.
• Dependence on anchor investors: two anchors could absorb the remaining capacity. Their flows, or their exit, could dominate the fund.
• Transcript quality: Otter mislabels speakers and garbles numbers, notably the allocator's size answer and some attributions around the trade-allocation exchange. Treat exact figures as needing confirmation.

8. CRM Note

CRM Note: 26 March 2026 intro and diligence call between the allocator team (VX, Speaker 3) and Kris Sidial (co-CIO, Ambrus) on the carry-neutral tail strategy. The allocator is re-entering tail risk after about two years out, has already started onboarding, and targets deployment within a month. Size is market-dependent and not committed. Ambrus says AUM is $700M with a $250M pipeline and a $2–2.25B hard close, says two anchors (a multi-strat and a large MFO) hold capacity rights, and wants new investors in formal ODD/IDD within about two months. Next step: Ambrus emails approved materials (pre-Liberation Day sample portfolio, investor letters, with the COO copied). Attribution and current holdings are pending compliance, and the allocator still needs sleeve-level P&L and an AUM reconciliation before sizing.

9. Call Claims vs. sources/

Suggested housekeeping: if this transcript should count as evidence under the project rules, move or copy it into sources/. Today it sits in Transcripts/, so findings that rely on it are manager statements, not evidence.