The House Edge

Appearance

Chapter 6 — Review Notes (Phase 3 self-check) ​

Status: REVIEW NOTES (Phase 3 — written by agent, accompanying DRAFT.md) Date: 2026-05-14 Working chapter: 06 — The Infrastructure Layer Word count: ≈7,100 words including footnotes (≈6,500 words of prose plus 22 footnotes plus inline Mermaid stack flowchart and firm-by-layer table). Above the Bible's 6,000 target — the chapter has 15+ named firms to develop and the content directives required the room.

The eight required questions ​

  1. Could a smart business reader with zero crypto background follow this chapter on first read?Yes for the structural argument; harder for the firm-by-firm depth. The cold open ("five firms touched one trade") is the right hook for a TradFi reader, and the paired Helius / Flashbots anchor profiles give the chapter a clear narrative. The four H4 mechanics subsections each follow the same pattern — define the layer, name the firms, develop one in depth. The Goldman MD will recognise the structural argument from her day-job experience with US equity infrastructure (DTCC, wholesalers, market-data vendors). Where she might slow down: the Solana subsection's depth (Helius's six layers, Jito's product stack, Ben Coverston as CEO of two firms, JIP-24, Paladin's reward split) is the chapter's densest stretch. The firm-by-layer table at the end of the verdict is meant to consolidate the load.

  2. Is every actor named, and is it clear how each one makes money?Yes. Helius (tiered RPC subscriptions $49–$999+/mo, validator stake yield, 50/50 wallet-rebate split, ETF partnership), Jito Labs (Block Engine + BAM tips routed to DAO post-JIP-24, JitoSOL 4% management fee), Flashbots (BuilderNet operator share, MEV-Share auction commission), Titan (~17.75% builder margin under exclusive flow, plus Titan Relay's payload-share commissions), BuilderNet (equal-split open-refund-rule among operators), bloXroute (subscription tiers for BDN, relay commissions on both Max-Profit and Regulated), Consensys/SMG (acquired the MEV-Blocker revenue stream from CoW DAO January 2026), CoW DAO (50% price-improvement share + 0.1% volume fee), Kolibrio (OFA auction at RPC layer), Triton, Astralane, Anza, Jump Crypto, Temporal, Paladin, Rakurai, Harmonic — every named firm has at least one paragraph and a footnote.

  3. Is there a worked example with specific dollar amounts threaded through the chapter?Yes — paired anchor profiles rather than a single-trade trace. Helius and Flashbots get full profile treatment in §4, with each named firm in the mechanics subsections developed against those two anchors. The dollar amounts are firm-level rather than trade-level: $67.6M Helius raised; $43B+ shielded by Flashbots Protect; 51.55% of Ethereum blocks built by Titan; $89.6M raised by bloXroute; $25M Jito Labs Q4 2024 Block Engine revenue. The chapter's verdict assembles a bottom-up aggregate — mid-hundreds-of-millions to low-billions of USD annual run rate across both chains — with explicit methodology.

  4. Does the chapter end with a clear "who wins, who loses" verdict?Yes. The verdict develops the winners (Helius, Jito Labs/Foundation, Titan/Gattaca, BuilderNet, Flashbots, bloXroute, the smaller specialist firms across both chains) with sourced firm-level numbers; identifies the losers (Eden Network defunct August 2025; Beaverbuild migrated into BuilderNet; retail traders paying cumulative cuts at every layer; validators without infrastructure relationships); and closes on the chapter's structural concern: the trader's question "who am I trading against, and who is taking a cut?" has an answer the trader cannot easily discover. The closing observation — "the chains' formal openness has been operationally enclosed by a B2B layer the average user does not know exists" — is the chapter's sharpest single sentence.

  5. Are all numbers sourced in footnotes?Yes. 22 footnotes. Every firm-level dollar amount, percentage, customer count, and named relationship has a footnote with URL and access date. Three explicit hedges called out in the prose: Titan/Gattaca's funding history is not publicly disclosed (footnote 8 says so); Quasar's founders are not publicly disclosed (developed in the prose as a chapter-level point); Astralane's funding and founders are not publicly disclosed (footnote 6 says so).

  6. Does the chain comparison box exist and contain real differences?Yes. Three paragraphs in §7 making structurally different points: Solana (operator-concentration across layers — same firms appearing at multiple layers; Helius's six-layer span; Ben Coverston as CEO of two firms); Hyperliquid (no significant third-party infrastructure layer; the protocol internalises the functions); Ethereum (layer-concentration — different firms but a small number at each specific function; Titan's 51% at one layer; bloXroute's 25% at another). The structural summary: Solana concentrates by operator; Ethereum concentrates by layer.

  7. Did I avoid every banned move from the Book Bible?Yes, with two flagged near-misses.

    • No hype words. No doom words. No tribal chain endorsements. No "Web 1.0." No throat-clearing. Specific firms, specific dollar amounts, specific dates throughout.
    • Near-miss 1: The verdict's closing line — "the chains' formal openness has been operationally enclosed by a B2B layer the average user does not know exists" — is editorial-by-degrees. It is a synthesis claim, not a directly sourced one. I think it is defensible — the chapter spent 6,500 words documenting the enclosure firm-by-firm — but it is the chapter's sharpest single sentence and the most likely candidate for Nick to soften or sharpen.
    • Near-miss 2: The Quasar paragraph develops the firm's anonymity as a chapter-level point. The framing — "a firm running ~15% of Ethereum blocks with no publicly identifiable operators is itself a chapter-level point" — is more pointed than the chapter's tone elsewhere. The Quasar example is, however, the cleanest case of the chapter's "you've never heard of them" framing.

  8. Would the Goldman MD finish this chapter without checking her phone?I think so, with reservations on the Solana subsection. The cold open is exactly the kind of TradFi-fluent framing she will respond to (the five-firms-touched-one-trade scene maps cleanly to the equity-infrastructure mental model). The Helius and Flashbots anchor profiles in §4 give her two firms to track through the chapter. The mechanics subsections — RPC, builders, relays, OFA — match her existing mental model of equity-infrastructure layers (data vendors, wholesalers, exchanges, ATSs). The chain-comparison closes the cross-chain argument with a clean Solana-vs-Ethereum distinction. The verdict's bottom-up aggregate gives her the scale anchor she'll want. The Solana subsection density (Helius's six layers, Jito's product stack, the Ben Coverston paragraph, JIP-24, Paladin, Temporal) is the chapter's densest stretch but the matrix table at the end consolidates it. The 7,100-word total is above the Bible's 6,000 target but each section earns its length.

What changed between phases — and what's load-bearing ​

Claims dropped from RESEARCH.md ​

  • Eden Network's full wind-down narrative. The research surfaced rich detail on Eden's August 2025 shutdown (treasury distribution, EDEN token holder payout, founder cited reasons). The chapter mentions Eden in one sentence in the verdict and one footnote — sufficient for the "firms that did not survive" point without overweighting a defunct firm.
  • Detailed JTO market-cap mechanics. The chapter cites JTO market cap (~$231M on 14 May 2026) once. The research had more detail on token unlock schedules, circulating supply trajectories, etc. — dropped to keep the chapter focused on infrastructure layer structure rather than tokenomics.
  • Hyperliquid validator-set details. Already developed in Chapter 5. Chapter 6 just mentions Hyperliquid as "no significant third-party infrastructure layer" without re-developing the architecture.

New claims added in the draft (and where they came from) ​

  • The chapter's structural argument that "Solana concentrates by operator; Ethereum concentrates by layer." Synthesis claim, supported by the named-firm pattern but not stated directly in any source. The chapter's central comparative argument.
  • The Helius-as-PFOF-analogue framing in the RPC subsection. Synthesis claim — Helius's 50/50 split with wallets is described as "structurally similar to a US equity broker's payment-for-order-flow relationship with a wholesaler" with the direction-of-flow note. The framing connects the chapter to Chapter 1's PFOF treatment (SEC DERA paper).
  • The "Helius is the largest single Solana validator" framing. The 15M+ SOL stake figure is well-sourced, but the chapter's positioning of Helius as a single firm with six-layer span on a single chain is composed by the chapter rather than directly cited.
  • The Ben Coverston paragraph. The CEO overlap between Temporal and Harmonic is documented in multiple sources (the Lightspeed podcast names him as Harmonic's CEO; DL News names him as Temporal's lead). The chapter's framing — "the same individual operating multiple firms at adjacent layers of one chain's infrastructure stack — is the kind of concentration the chapter's argument predicts but that no other chain has produced quite as visibly" — is synthesis.
  • The aggregate take-rate estimate. Built bottom-up from named-firm numbers with explicit methodology. The range (mid-hundreds-of-millions to low-billions annual run rate) is directional rather than precise; the chapter says so.

Things I'm uncertain about ​

  1. The chapter's length. At 7,100 words it is above the Bible's 6,000 target. The content directive (15+ named firms) genuinely needs the room, but if Nick wants to trim, the most cuttable content is some of the Tier 2 brief-firm coverage (e.g., the Paladin paragraph could shrink; Astralane could compress).

  2. The Quasar anonymity framing. I developed the firm's unknown founders as a chapter-level structural point. The argument lands — a firm running 15% of Ethereum blocks with no public operator identity is striking — but it's a stronger claim than the chapter makes elsewhere about other firms. If Nick wants this softer, the framing can be neutralised.

  3. The PFOF analogue in the RPC subsection. Helius's 50/50 wallet rebate is structurally like PFOF (a downstream pipeline operator paid by an upstream extraction party for access to retail flow) but mechanically different (flows in opposite direction; explicit rather than embedded). I made the analogy carefully. If a careful reader objects to the analogy, the framing can drop.

  4. The JIP-24 paragraph weight. The 2025 JIP-24 passage that routes 100% of Jito Block Engine + BAM tips to the Jito DAO (rather than the prior 3%/3% split) is the cleanest case in the book of an infrastructure firm voluntarily restructuring its revenue. I developed this in a paragraph. If Nick wants more weight (it could be a sidebar) or less (it could be a footnote), easy adjustment.

  5. The aggregate take rate estimate. I built it bottom-up because no published source aggregates it. The methodology is in the prose. If Nick wants the estimate dropped — preferring "the layer is significant but the precise number is not public" — that's a one-paragraph cut.

Places where the prose got technical and might lose the reader ​

  • The Solana subsection of the block-builders mechanics section. Jito Labs full treatment + JIP-24 paragraph + Temporal + Nozomi + the Ben Coverston paragraph + Paladin's reward split — five named firms / structures in one ~700-word stretch. The matrix table helps, but the prose density is real. If a reader stops here, the chapter is hard to recover from.

  • The OFA subsection. Four named services with overlapping mechanics (Flashbots Protect, MEV-Share, MEV-Blocker now at SMG, CoW DAO, Kolibrio). I distinguished them carefully but the variant landscape is dense. The Bible's "diagrams replace prose" rule could apply here — a comparison table specifically for OFA operators would land cleaner than the current prose.

  • The Meet the Infrastructure Provider sidebar. Long compared to prior sidebars (Meet the Validator, Meet the Builder, Meet the Searcher, Meet the Market Maker). The chapter has six fields and the content is dense. If the sidebar feels heavy, the Founded field can drop (the firms have varying founding dates that the prose already covers).

Files written/modified in Phase 3 (this chapter) ​

  • book/chapters/06_infrastructure/DRAFT.md — new, ≈7,100 words (22 footnotes; inline Mermaid stack flowchart + 19-firm matrix table; Meet the Infrastructure Provider sidebar)
  • book/glossary/GLOSSARY.md — appended 5 entries: Infrastructure firm (B2B trading infrastructure), JIP-24, MEV refund, Order Flow Auction (OFA), RPC provider. Now 75 total entries.
  • book/OUTLINE.md — Chapter 6 entry updated with subtitle and final section headings.

Chapter 3 back-edit (JIP-24) ​

book/chapters/03_mempool/DRAFT.md footnote 8 — added a 2025 JIP-24 update note: "The entire 6% Block Engine and BAM tip take now routes to the Jito DAO treasury; Jito Labs (the company) no longer takes a direct cut of these flows. The Q2 2025 numbers cited above were pre-JIP-24. Chapter 6 develops the firm vs. DAO restructuring." Ch 5 was checked and does not have a JIP-24-affected footnote — Ch 5 only mentions "Jito tips and BAM tips" generically in the Meet the Validator sidebar.

Phase 3 is complete. The chapter is now in Nick's review queue.

Per production order, the next chapter is Chapter 7 — Exclusive Order Flow, the structural pay-off of the actor chapters. Chapter 6 ends with "What changes when one of these infrastructure firms not only spans multiple layers but explicitly contracts with one specific searcher or trading firm for exclusive flow?" — that is Chapter 7's question.