Comfy Sunday Spaces v22 10/10, CZ vs Star and Metals

ThinkingUSD avatar
Flood
@ThinkingUSDFeb 5, 2026

The Spaces brought Flood and co-host Kyle into a candid re-underwriting of L1s, arguing that Ethereum/Solana have weakening investment cases as large institutions prioritize control, reliability, and monetization over credible neutrality. They explored why enterprises may opt for own/permissioned chains (e.g., Robinhood’s chain, DRW’s Canton), and how DeFi’s hack record and legal ambiguity on tokenized ownership undercut institutional adoption. Kyle reframed token value via “community money” while noting Hyperliquid’s HIP-3 shift to stables-denominated perps that accrue value to the venue. A deep dive into CEX market structure covered the 10/10 crash, empty books, algorithmic halts, and the limited impact of stablecoin depegs, calling for full transparency vs DEX auditability. They discussed HIP-3 metals’ rapid product-market fit, regulatory considerations for RWA perps, and the need for tokenized spot to improve hedging and liquidity. The duo criticized “crime season” listings, urging exchanges to raise standards and curate higher-quality assets, with Robinhood cited as a model. Additional insights included Kyle’s on-chain study of Binance’s BitMEX trades during COVID, Justin Sun’s sharp trades and acquisitions, portfolio positioning (long HYPE, reduced BNB), and Q&A on Hyperliquid’s growth path via builder codes and feature refinement.

Session overview and participants

  • Hosts: Flood (primary host) and Kyle (co-host). Tone: reflective and critical; aim to “re-underwrite” old theses and reassess what’s working now.
  • Format: Market check-in → L1 thesis re-underwriting → Hyperliquid HIP‑3 (metals/RWAs) → CEX market-structure drama (10/10) → listings/meta quality → hiring → Q&A.
  • Additional participants (Q&A): Beefed, shindirec, Matt (aka Nomat), Nice, Archer, Tim Brower, yallo.

Re-underwriting the L1 thesis (ETH/SOL vs enterprise reality)

  • Flood’s core thesis: Public L1s (Ethereum, Solana) have become tough investments.
    • The original selling points (credible neutrality, open access) are not necessarily advantages for mega‑cap enterprises and asset managers (e.g., BlackRock, State Street).
    • Enterprise priorities: fee generation, operational efficiency, trust and stability. Public L1s do not obviously maximize these; security incidents and unclear rollback norms are intolerable for institutions with fiduciary/brand risk.
    • Legal risk in tokenized ownership: If stock/ownership records live on-chain and keys are compromised (wrench attack/hack), who legally owns the asset? Current DeFi precedent offers poor recourse.
    • Big Tech threat: If Google/Amazon deem L1s existential (paralleling how AWS/Azure/Google Cloud emerged), they could deploy hybrid/permissioned L1s with SLAs, control, and tailored compliance—directly competing with public L1s for enterprise workflows.
    • Questions posed:
      • Why assume asset managers will deploy/custody on public L1s they don’t control or majority-own?
      • Why wouldn’t they prefer infrastructure from a counterparty whose equity value dwarfs ETH/SOL—thus offering stronger assurances and bespoke support?
  • Kyle’s perspective and partial alignment:
    • Ethereum’s early product-market fit reflected 2016–2018 crypto-native preferences (credible neutrality). Around ~2021 that user preference faded.
    • ETH persisted as “regulatory cover” (minimum decentralization threshold to build permissionlessly), not an engineering-first choice.
    • Today, that cover is less necessary; enterprises want their own chain: retain rollback/control, but expose an open, standardized API/Explorer to external developers.
    • Examples: Robinhood chain ambitions; Tempo; DRW’s Canton (Kyle disclosed association). Hyperliquid’s “builder codes” are a strong pattern: own the substrate and sovereignty, while offering an open, standardized dev surface.
    • Result: Many L1 tokens sit in a confusing spot versus clear enterprise needs.
  • Community money dynamic (Jordy Alexander):
    • On public L1s, native tokens naturally become “house money” for activity (NFT pricing, AMM base pairs), which bids the token in risk-on phases.
    • Hyperliquid’s metals/stocks perps highlight the counterexample: activity is stables‑denominated; the “house‑money” reflex is not required for PMF.

DeFi risk, onboarding cycles, and slowing metrics

  • Flood’s DeFi risk question: What’s the aggregate market cap of leading DeFi (Aave, Uniswap, SNX, Compound) vs cumulative hack losses? He suspects the loss share would shock outsiders, implying poor risk‑reward for institutions.
  • Onboarding arcs:
    • 2017–2020: Retail onboarded via Bitcoin exposure.
    • 2020–2021: ETH won the net new user funnel (Metamask, NFTs, on-chain swapping).
    • 2023–2024: Solana led net new onboarding (low fees/latency; meme coin wave). A “hyper‑growth tech stock” framing at $30–$40 SOL was valid.
    • Now: Key activity metrics (e.g., on-chain trading volume) have slowed on SOL; narrative pivoted from memes to on-chain equities. Open question: What is the growth rate and will SOL win the on-chain equities share?

Hyperliquid HIP‑3 (metals) and the RWA expansion

  • What changed:
    • HIP‑3 launched metals less than two weeks prior; volumes exploded into the single-digit billions. Crypto has underperformed equities; HIP‑3 is the first credible way to get equities/commodities-like beta within crypto.
    • Product uniqueness plus PMF → premium valuation. Diversified revenue streams (beyond crypto perps) should command higher multiples.
  • Structural note:
    • Perps let you list any asset with a robust oracle. Hyperliquid is demonstrating a multi‑asset exchange on-chain (crypto, commodities, equities proxies) that mirrors Robinhood’s arc from equities→crypto in reverse.
  • Market microstructure and participants:
    • Activity is margined in stables; no “house money” bid needed for token reflexivity.
    • Edge accrues to nimble two‑to‑three‑person market-making pods exploiting nascent/fragmented inefficiencies before large players fully arrive.
  • Portfolio stance (disclosure):
    • They hold HYPE (research only; not investment advice). They see Hyperliquid as “all‑weather” in both boring crypto regimes (access to non-crypto assets) and crypto-native bull runs (broader menu of markets).
    • Liquidity has improved year-on-year; public comps vs Binance liquidity have circulated.
  • Data points and regulation:
    • Reported that CME traded ~$130B in total derivatives one recent Friday; Hyperliquid’s silver market traded ~$3B in under 10 days—low single-digit percent of global metals derivs, but a true 0→1 moment for a decentralized venue.
    • Commodities aren’t securities but are regulated. As HIP‑3 scales, expect scrutiny of non‑KYC decentralized trading hosting a material share of global volume. Philosophically, Flood supports the freedom to transact but recognizes the need to “re-underwrite” regulatory risk as scale grows.
    • Kyle’s defensibility view: Commodities per se have less insider asymmetry; buying spot gold is not a regulated securities transaction. TradFi figures (e.g., Cathie Wood) are noticing Hyperliquid’s role as a rare bright spot.

CEX market structure, 10/10 event, and transparency deficits

  • BNB position: They own BNB (trimmed recently). Context: “Mom and dad are fighting”—public feud between Star (OKX founder Star Xu) and CZ (Binance founder).
  • 10/10 anecdote (Kyle):
    • During the sharp move, he observed near-empty order books, extreme spreads, and instant fills on multi‑million orders in a normally liquid instrument—indicative of systemic quoting/hedging halts.
    • He downplays the USDe (Ethena) depeg as a primary cause; more likely a trigger that broke “invariants” across algos (ADLs reappearing after years; stables not at $1), causing widespread cancel‑all and human restarts.
    • An exchange’s memo (“ByteDance” referenced in the call; context suggests a major CEX) cited API latency around collateral movement and portfolio balances and offered discretionary rebates. Hosts argue the post‑mortem was too thin for the damage to market trust; they call for a tape‑level incident report.
    • Black‑box problem: On CEXs, users can’t audit losses, rebate methodology, or balance‑sheet impacts. On-chain venues expose PnL holes and can be priced.
    • Speculative angle: In thin weekend conditions (post‑Friday close, no ETF flow), large players could find it advantageous to force vol, reset OI, and stress competitors. No specific accusation; an observation about incentives.
  • Historical parallel: COVID crash (2020) saw BitMEX downtime and a migration of trust/flow to Binance for better performance. A similar trust reallocation could happen again.
  • Star–CZ backstory: CZ worked at OKCoin (pre‑OKX) under Star Xu before founding Binance; rivalry endures.

BitMEX research and exchange trading anecdotes

  • Kyle’s 2018–2019 BitMEX chain analysis:
    • BitMEX used unique deposit addresses and pooled withdrawals, enabling clustering of profitable/unprofitable cohorts and linking multi‑account users.
    • Pre‑COVID trade example: A recurring large account deposited 20k BTC to BitMEX, put on ~80k BTC shorts (levered), then withdrew ~68k BTC profit. The withdrawal went to the Binance hot wallet—strongly indicating exchange‑level activity (users can’t credit to a hot wallet). Interpretation: exchange hedging/directional positioning via BitMEX.
  • Justin Sun segment:
    • Hosts praise his on‑chain trading acumen and strategic acquisitions (Poloniex, Huobi) that captured stranded, non‑KYC’d balances via custody fees—“200 IQ” balance sheet trades. Note litigation risk and “dust” accumulation dynamics across CEXs.

Listings meta, “crime season,” and exchange responsibility

  • “Crime season” label: Recent listing cycles are broadly bearish for buyers.
    • Chart cited: Average Binance listings in 2025 spike in week one, then trend down ~85% after one year.
    • CZ’s stance (town hall): Users choose to buy; exchanges don’t force it. Hosts counter that blasting listings to hundreds of millions of users predictably drives feeable churn; there’s a moral dimension.
  • 2017 context vs now:
    • Coinbase/Kraken were more selective; today, the “trash pumps” meta pushes venues to list or lose users to competitors willing to list everything.
  • Proposed improvements:
    • Tougher listing criteria and marketing—force transparency on vesting, token value accrual, and consider standardized buyback programs when projects are profitable (examples cited: pump.fun, Hyperliquid).
    • Curated retail menus and/or indices of “higher quality” tokens to help non‑expert users.
    • KPI idea: Rate exchanges by average return of listed tokens over time—align venue incentives with customer outcomes.
  • Structural drags on token performance:
    • Market saturation spreads retail attention thin; foundations employ locks/OTC hedges that depress secondary; net result: crypto indices underperformed equity indices, and even BTC/ETH/SOL lagged broad equities the last ~2.5 years.
  • Walled gardens and TradFi competition:
    • Robinhood currently offers the best retail “safety” menu among crypto brokers, per hosts.
    • IBKR/Schwab crypto: hosts tested IBKR stablecoin deposits; rumor of zero‑fee BTC/ETH at Schwab. Payment‑for‑order‑flow‑like models could come to crypto, compressing fees and improving UX for mainstream users.

Q&A highlights

  • Beefed (ecosystem breadth):
    • Spot equities on Hyperliquid: Felix x Ondo partnership noted.
    • NFTs on Hyper EVM: “Hypers” (Genesis airdrop), Tiny Hyper Cats, etc.; hosts own/participate.
    • Kyle: You can increasingly build an all‑weather on-chain portfolio (derivs + tokenized spot).
  • shindirec (US regulation trigger):
    • Flood: Possible even at today’s scale. Hyperliquid would likely argue decentralization akin to BTC/ETH/SOL. He avoids predicting specifics but flags this as a thesis variable to re‑underwrite as HIP‑3 grows.
  • Matt/Nomat (RWA perps market structure):
    • Latent demand is large; metals volatility/realized vol has flipped the “boring gold/silver” narrative.
    • MM participation is constrained: crypto-native MMs struggle to hedge in TradFi; TradFi MMs are unfamiliar with perps/permissionless venues.
    • Missing piece: liquid tokenized spot (on-chain) to enable efficient arbitrage and tighter spreads—benefiting all perp venues.
    • Competition is validation: Binance listing TSLA perps is additive for liquidity and lowers cost of capital across the category.
    • Kyle: As higher-quality RWAs arrive, the long‑tail alt “ticker game” loses allure.
  • Nice (distribution to day-traders beyond crypto):
    • Flood: Builder codes allow third parties to build bespoke UIs/experiences on Hyperliquid’s shared order book without brokerage agreements—monetization programmable (bps-level fees). Bull case: many frontends, one deep kitchen (liquidity/clearing) → varied UXs on a unified market.
    • Valuation framing: As HL evolves into a “real business,” comps to Schwab/IBKR/Robinhood become relevant (revenue, margins, per‑employee efficiency), even if TAM narratives remain expansive.
    • Kyle’s analogy: Many dining rooms, one kitchen. Builders own UX/brand; HL centralizes the hard parts (matching, risk, maintenance).
  • yallo (what’s next—HIP‑4/5/6?):
    • Flood: Expect growth driven by independent teams launching new markets (HIP‑3 pattern), while Hyperliquid expands multi‑asset capabilities.
    • Kyle: Near‑term focus is refinement/polish over “net new primitives.” Examples:
      • Oracles: Silver contracts currently clamp to prior close to avoid untethered moves when TradFi is shut; as HL hosts meaningful share of global price discovery, loosening that “leash” makes sense.
      • Mechanism hygiene: Revisit inherited defaults like funding‑rate clamps (13–15% annualized) and improve isolated margin/collateral UX.

Hiring and closing notes

  • Hiring:
    • Full Stock (fullstock.trade): trading front end; high‑performance engineering, QR/QT; Rust preferred; NYC.
    • Full Stack (family office/HFT): frontend, backend, blockchain engineers.
    • Intern program: 11‑weeks, in‑person/hybrid at Rameel Doc Capital; ~140 applicants so far.
  • Closing: Expect an “exciting” week ahead; continued focus on thorough, intellectually rigorous research while recognizing portfolio disclosures.