The House Edge

Appearance

11. Who's at a Disadvantage, and Why They Don't Move ​

Cold open ​

In May 2025, the Pump.fun launchpad — Solana's dominant memecoin-issuance venue — reported that approximately 30.1% of all trading wallets ended a measured period profitable. Twelve months later, in April 2026, that figure stood at approximately 73.3%.[1] The platform appeared, by the most-cited measure of retail outcomes, to have produced a roughly 2.4× improvement in retail profitability over the year. The number is true. The framing is misleading. Over the same period, Pump.fun's monthly active wallet count collapsed from approximately 5.2 million to approximately 1.8 million — a 65% exodus.[2] The platform's "improvement" in retail-trader outcomes was not produced by losers becoming winners. It was produced by losers leaving the table. The wallets that did not leave were, on average, the more skilled ones; the remaining 73.3% profitability rate reflects survivorship, not learning. This chapter is about the trader, the LP, the market maker, and the validator who didn't leave the table — and the structural reasons they didn't. Part III documented three chain architectures (Solana's vertical integration that compressed the extraction surface; Hyperliquid's in-consensus matching that eliminated it; Ethereum's PBS + L2 + restaking trifecta that dispersed it). This chapter documents the four chronic-loser categories that exist across all three: retail traders unaware of toxic flow, passive LPs in informed-flow pools, slow market makers, and validators without infrastructure relationships. The question is not whether each category loses — that is the empirical anchor of the prior ten chapters. The question is why each category continues to participate anyway, and why "just leave the table" is harder than it sounds.

What this chapter answers ​

  • Who are the four chronic-loser categories on-chain, and how do their losses appear in the data Chapters 1–10 documented?
  • Why does each category continue to participate, given that the structural disadvantages are documented?
  • Where is the consent gap — the difference between informed participation in a known disadvantageous structure and uninformed participation in a structure the participant does not understand?
  • What does "just leave the table" actually look like for each category, and why is that exit path harder than the framing suggests?

The setup ​

The chapter's organising frame: participation is not the same as consent. A passive LP who deposits into a Uniswap V3 ETH/USDC pool knowing that the LVR framework predicts a roughly 5–7% annual loss to informed flow has consented to a known structural disadvantage in exchange for accepting volatility-as-payment.[3] A retail trader on a Solana memecoin venue who believes they are competing against other retail traders on equal terms, when they are in fact trading against prop-AMM operators with privileged routing visibility, has not consented to that structural disadvantage because they do not know it exists. The four loser categories the chapter develops differ along two axes: (a) the magnitude of the structural loss they bear; (b) the informational gap between the loss as experienced and the loss as understood.

The chapter previews three structural findings. First: the loser categories are not symmetric. Retail traders bear small per-trade losses across high transaction counts; passive LPs bear continuous LVR drag on slow-moving capital; slow market makers face displacement — loss of the competitive position itself, not loss-per-trade; validators without infrastructure relationships bear opportunity costs in the form of unrealised top-quintile revenue rather than direct losses. Second: each category's exit path is structurally constrained. Retail's alternative is centralised exchanges in their jurisdiction (often unavailable, particularly in the US); LPs' alternative is depositing nowhere (the LVR-aware LP is rare in 2026 retail behaviour); market makers' alternative is pivoting to prop-AMMs (an option only available to firms with prop-AMM-grade capital and technology); validators' alternative is exiting the network (which forfeits the staking yield itself). Third: the consent gap is the chapter's clinical core. The chapter does not argue that the structures are illegitimate — three of the four loser categories include participants who knew the structure when they entered it. The chapter argues that the consent gap is largest for retail, and that the structural mechanisms that produce this gap (UX abstraction, information asymmetry between platform and user, the absence of mandatory disclosure of toxic-flow routing) are the chapter's load-bearing finding.

The worked examples ​

Three short cases, one per category-with-named-actors. The slow market maker category is developed at the category level rather than via a single named firm — named-firm events reflect volatility within the winner set rather than displacement of the loser category, and the structural argument is sharper at category level.

Worked Example A — Carlos, the Pump.fun chronic loser. Carlos — the trader introduced as the Ethereum case in Chapter 3 and reused here as a generic Solana retail name — is one of the approximately 1.8 million wallets still trading on Pump.fun in April 2026. Per the platform's survivorship data, Carlos is statistically more likely to be profitable than the average 2025 user — but only because the unprofitable 65% have left. Carlos's structural position: trading against prop-AMM operators routing through Helius and Jupiter Beam (Chapter 8); paying the cumulative ~$8–15 take per $10,000 notional (Chapter 8 Alice trace); does not know that approximately $0.50–$1 of his per-swap cost is captured at the Helius/Phantom rebate layer and split 50/50 (Chapter 7) without his explicit consent. Carlos's continued participation is explained by: (a) UX advantage — Phantom is the easiest crypto wallet experience available; (b) absence of equivalent CEX access — he is US-domiciled and most major centralised exchanges that allow Solana memecoin trading are no longer fully available to US retail post-2024; (c) the per-period rather than per-trade loss framing — the BIS Bulletin 69's "73–81% of retail crypto investors lose money" finding does not produce behavioural change at the individual level because Carlos sees each swap's outcome, not the aggregate of his year.

Worked Example B — Bob's Uniswap V3 LP revisited. Bob — introduced in Chapter 2 as the LP case — deposited $50,000 into the ETH/USDC 5-bps pool on Uniswap V3 in late 2024. By May 2026, Bob has earned approximately $4,500 in fees and incurred approximately $7,200 of LVR-driven impermanent loss against the equivalent buy-and-hold benchmark — a roughly $2,700 net loss on the position. Bob's behaviour is informed: he read the Milionis, Moallemi, Roughgarden, and Zhang LVR paper after his second quarter of net-negative performance, and he understands the structural mechanism.[3:1] Bob's continued participation is explained by: (a) the absence of a better LP venue at the capital scale he wants to deploy passively; (b) the "fees-earned" framing on Uniswap's UI which displays gross fees prominently and LVR not at all; (c) the structural fact that withdrawing crystallises the loss as realised, where leaving the position open allows Bob to mentally treat the loss as unrealised. The HLP on Hyperliquid (Chapter 9) is the structurally different passive-LP-equivalent — depositors who consented to a clearinghouse-default-fund-style role and were JELLY-tested on 26 March 2025 with a 65% TVL drawdown.[4] The HLP depositors are informed in a way Bob is not; the structural exposure is comparable.

Worked Example C — Erika, the SFDP-removed Solana validator. Erika operates a small Solana validator with approximately 8,000 SOL of external stake plus 25,000 SOL of Solana Foundation Delegation Program stake. In April 2026, the Foundation announced that approximately 150 validators would lose SFDP delegation under the May 2026 ASN-concentration and data-centre-concentration rules (Chapter 8).[5] Erika's data centre — a smaller European facility outside the Frankfurt-Amsterdam axis — exceeds the new 15% concentration threshold; her SFDP stake will be revoked. Her structural position: stock Agave-Jito client; no BAM Node relationship; no Harmonic Performance strategy; no Frankfurt or NY5 colocation; per Syndica's March 2026 telemetry, her per-block priority-fee capture sits approximately at or below network median. Post-SFDP revocation, her per-block revenue falls to the ~6,000 SOL minimum economically viable threshold; she has 30 days to either acquire enough external stake to remain in the Foundation-stake program or wind down. Erika's continued participation is explained by: (a) the sunk-cost / hardware-asset framing — her one-time $12,000 colocation CapEx; (b) Alpenglow's option value — expected late-Q3/early-Q4 2026 mainnet activation with the Validator Admission Ticket dropping the profitable-validator threshold from ~4,850 SOL to ~450 SOL; if she can survive the SFDP transition, the post-Alpenglow economics may restore viability; (c) the structural fact that exiting forfeits validator-set membership entirely, where remaining preserves the option value.

The mechanics, in detail ​

Retail traders unaware of toxic flow ​

The empirical anchors. The BIS Bulletin 69 retail-loss aggregate: approximately 73–81% of retail crypto-app users likely lost money on their initial Bitcoin investment; average retail investor loss approximately 47.89% of invested funds.[6] Coverage: 95 countries, August 2015 through December 2022; major crypto trading platform monthly active users rising from approximately 100,000 in August 2015 to approximately 30 million in November 2021. The structural finding: as prices rose, smaller users bought while the largest holders ("whales") sold — making returns at smaller users' expense. The losses are direct and measurable. Crypto's retail loss profile is not a niche phenomenon of memecoin trading or sandwich attacks; it is the structural shape of the asset class as the BIS measured it.

Pump.fun's survivorship-driven "improvement" — 30.1% in June 2025 → 73.3% profitable wallets in April 2026, coinciding with a 65% monthly-active-wallet collapse from approximately 5.2 million to approximately 1.8 million — is the cleanest single 2026 data point on the "losers leave, platform metrics improve, structural losses persist for remaining participants" pattern.[7] In the April 2026 cohort, the largest profitable group (approximately 2.05 million wallets, or ~65% of profitable wallets) earned only $1–$500 for the month. Approximately 793,000 wallets — roughly a quarter of active — lost between $1 and $500 in the median month; approximately 46,000 wallets (~1.5%) lost more than $500. The platform is healthier-looking because it lost the worst-performing 65% of its 2025 users, not because the remaining 35% improved their per-trade outcomes.

Banana Gun's 1.3 million users and 25.3 million executed trades — with $16.09 billion cumulative trading volume and an average trade size of $635 — are the chapter's clearest named-retail-category structural anchor.[8] Banana Gun's commercial model is documented in Chapter 7: users pay 0.5%–1% per trade on Ethereum, the bot routes exclusively to Titan, Titan retained 2,271.26 ETH of the 4,466.89 ETH user fees (~50.8%), and Titan's builder share rose from <1% to >50% on the strength of that single arrangement. The 1.3-million-user count is the under-developed fact in Chapter 7's treatment: this is not a marginal product. It is approximately the same order of magnitude as Coinbase's monthly active US retail user base for crypto trading.[9] At least 1.3 million retail users are routing through one bot operator into one exclusive builder contract, paying approximately half of their on-chain costs to a firm whose existence is not disclosed in any product surface the bot's interface displays.

The FCA's Cryptoasset Consumer Research 2025 (Wave 6) documented that UK crypto ownership declined from 12% to 8% of UK adults over 2024–2025 — from approximately 7 million to approximately 4.5 million users — while average holdings per investor rose to approximately £1,842 (~$2,500).[10] Awareness of cryptoassets remained high (91%), but deeper understanding of stablecoins, DeFi, and crypto lending was variable. 73% of UK users purchased through centralised exchanges (up 4% from 2024). The structural finding: the FCA's Financial Promotions Rules (October 2023) reduced participation but did not reduce per-trade losses for the residual participants — the wallets that remained were not better at trading, they were simply the harder-to-deter subset. Disclosure as a regulatory intervention shifts who participates without changing what happens inside each trade.

The structural mechanism Carlos cannot see. The Phantom + Jupiter UI shows him his SOL balance and his executed-trade price. It does not show him: the Helius 50/50 wallet rebate split (Chapter 7); the Jupiter Beam private-bundle routing into Jito (Chapter 8); the +101% Harmonic Performance validator-side capture (Chapter 5, Chapter 7); or the prop-AMM internalisation that is quoting against his swap (Chapter 2, Chapter 8). The per-trade loss Carlos pays is real; the per-layer trace of where it goes is, in the chapter's framing, abstracted away by the very UX that makes the experience accessible to him in the first place. This is the chapter's first instance of the consent gap: the trade is consented; the structural take that is bundled with it is not, because it is not disclosed in the surface Carlos sees.

Passive LPs in informed-flow pools ​

The empirical anchors. The Loss-Versus-Rebalancing (LVR) framework is the chapter's central passive-LP anchor. Milionis, Moallemi, Roughgarden, and Zhang's 2022 paper (arXiv 2208.06046) formalised the LP problem as the difference between the LP's P&L and a constant-mix rebalancing portfolio's P&L.[11] The key result: an LP in an AMM with 5% daily price volatility loses approximately σ²/8 per unit time — which annualises to approximately 11% per year to arbitrageurs alone, before fees. For high-volume pairs like ETH/USDC and BTC/USDC, the empirical LVR loss runs approximately 5–7% of deposited capital per year.

Topaze Blue's 2022 study found that approximately 49.5% of Uniswap V3 LPs realised negative returns net of impermanent loss across 17 V3 pools (~43% of V3 TVL): $199.3 million in collected fees against $260.1 million in IL — passive LPs were collectively approximately $60.8 million worse off than HODL.[12] No comprehensive 2025–2026 re-measurement of this finding has been published; subsequent commentary (MEXC Research 2025) places the share of LPs in volatile pairs losing money net of fees at approximately 54.7%, directionally consistent with the Topaze Blue finding.

Uniswap Labs's own 2023 research found that V3 passive LPs in the ETH/USDC 5-bps tier — the highest-volume DeFi pool — underperformed V2 by approximately 68%.[13] Across all tiers and pools studied, V3 passive LPs earned higher fees than V2 LPs by an average of approximately 54%; but in the dominant ETH/USDC 5-bps tier, the most-traded pool is the worst case for the passive LP. The mechanism: the most-traded pool is the pool where counterparties most often know more than the pool does, and the passive LP absorbs adverse selection at the LP's expense. The dominant DeFi LP pool is, by Uniswap's own research, where passive LPs lose worst.

Hyperliquid's HLP is the structurally different case the chapter develops as a comparison.[14] HLP TVL fell from approximately $500 million pre-12 March 2025 to approximately $177 million by 1 April 2025 — a 65% drawdown — on the back of a single targeted whale attack on the JELLY perpetual. The Hyper Foundation delisted JELLY and closed positions at $0.0095 (a $700,000 gain to HLP); HLP recovered to approximately $383 million TVL by May 2026 (Chapter 9). Historical HLP returns: approximately 1.75% per month / ~20% annualised through 2025; a $700 million February 2026 liquidation generated approximately $15 million in HLP profit. HLP is structurally distinct from a Uniswap V3 passive LP — the depositor knows she is funding the chain's house and absorbs both upside (3% of HLP-routed trading fees) and tail risk — but the category-level loser dynamic is comparable: any HLP depositor participating during a single JELLY-style event eats a fraction of the drawdown. The HLP is the chapter's clearest case of an informed passive-LP class with explicit consent; Bob's position is the chapter's clearest example of the same structural role without explicit consent.

The structural mechanism underlying passive-LP loss. Toxic flow is arbitrage order flow that forces the LP to buy or sell against the market; other flow is "uninformed" (mostly retail). CrocSwap's series on toxic-flow detection has identified two clusters of negative-PnL wallets with high notional swap sizes, and aggregator-routed flow (Jupiter, CoW Swap) appears to remain a credible signal of non-toxicity not easily gameable by arbitrageurs.[15] The structural consequence: passive LPs in the most-traded pool absorb the most-toxic flow; the solver markets and aggregators have routed retail flow away from passive pools and toward prop-AMMs and PMM solver inventory. The passive LP is the structural residual — left holding the toxic component because the non-toxic retail component was siphoned off by the aggregators.

Bob's continued participation is the structural finding: the absence of an LVR-aware passive LP venue at his capital scale; the Uniswap UI's gross-fees-prominent, LVR-absent display; the unrealised-loss psychology that makes withdrawal feel like crystallising a loss rather than stopping a structural bleed. The consent gap here is partial: the LVR research has existed in published peer-reviewable form since 2022 and is one Google search away from any LP with two quarters of negative results. The structural mechanism is not hidden; the surface where the LP deposits does not display it.

Slow market makers ​

This subsection is developed at the category level rather than via a single named firm. Named-firm events — Wintermute's mid-May 2026 BTC + ETH liquidity compression on Hyperliquid (approximately a 90% drop in combined BTC + ETH resting notional), Auros Global's near-simultaneous reduction — reflect volatility within the winner set rather than displacement of the loser category.[16] The structural displacement Chapter 11 documents is the displacement of categories of competitive form, not of specific firms.

Phoenix's revenue collapse from $3.7 million in Q2 2024 to $68,604 in Q1 2026 — a roughly 54× decline — is the canonical "market makers that lost the spot CLOB war" case.[17] Ellipsis Labs renamed Phoenix to Phoenix Legacy in early 2026 and pivoted to perpetuals (Phoenix Perpetuals, announced Breakpoint 2025). The displacement was not a single firm exiting; it was the entire competitive form (on-chain spot CLOBs on Solana) failing to compete with prop-AMMs and aggregator-routed flow. Manifest at approximately $3.9 billion of 30-day spot volume captures less than 2% of the $284.5 billion Q1 2026 Solana DEX volume; OpenBook sits in the same compressed band. The slow market maker on Solana is not a named firm — it is the category of market makers whose business depended on tight-spread quoting against a public CLOB and who could not compete with the prop-AMM internalisation that ate their flow.

The Ethereum slow market maker is the firm that did not capture a builder-direct contract under Wu et al.'s 75 EOF arrangements (Chapter 7).[18] Pahari and Canidio's December 2024 measurement: 77.2%–84% of total fees paid in winning Ethereum blocks come from exclusive transactions not available through public order flow. The original-generation L1 market makers — the firms that quoted into the public mempool and the pre-MEV-Boost-private-order-flow architecture — have either pivoted (Beaverbuild migrated into BuilderNet; Flashbots stopped operating a centralised builder December 2024) or compressed (smaller firms that did not have the capital to build a builder operation themselves). The 2026 winners are SCP and Wintermute (more than 90% of UniswapX volume between two firms); Barter (targeting 50%+ of CoW solver share); the named approved counterparties for the regulated venues (Bitwise BHYP per Chapters 7 and 9: Wintermute, Flowdesk, FalconX, Nonco).

The Solana market maker without Frankfurt / Equinix FR5 colocation, BAM Node access, or Harmonic strategy enablement earns at-or-below network median priority fees regardless of operational discipline — the +101% Harmonic Performance gap measured against the median (Chapter 5, Chapter 7, Chapter 8) is the chapter's "speed-and-relationship moat" anchor.[19] Syndica March 2026 measurements: Frankendancer Harmonic Performance +101%, Frankendancer Harmonic Balanced +39%, Agave Harmonic +36% per-block priority fees versus network median. Chorus One's August 2025 finding: combined operational gains from timing games plus scheduler optimisation amount to approximately +3% in total rewards (~27 basis points annualised). The ratio is approximately an order of magnitude, and the slow market maker on Solana is the operator at network median or below, by definition.

The structural reasons slow market makers continue to participate. (a) Sunk costs — existing infrastructure, established team, and operational expertise that cannot be easily repurposed. (b) Option value — the structural shape may change. BAM open-sourcing mid-Q2 2026 (Chapter 8) could in principle reset the relationship-based moats, and the firms that have remained ready to deploy are positioned to benefit. (c) The structural difficulty of pivoting to prop-AMM operation — which requires the firm to originate trading flow rather than to quote on it. Wintermute (Tessera V on Solana) and Ellipsis Labs (Phoenix Perpetuals) are the named pivoters; the smaller firms — those running standalone quoting bots on Phoenix, OpenBook, or any of the on-chain CLOBs that compressed — do not have the option to internalise prop-AMM economics because the operational stack requires multi-million-dollar capital outlay. The slow market maker is, in the chapter's framing, the loser category most able to consent to its position because the firms understand the structural mechanism. What they lack is the capital or the operational ability to exit.

Validators without infrastructure relationships ​

The empirical anchors. Solana's validator count has collapsed from approximately 2,560 active validators in mid-2024 to approximately 770 in March 2026 — a roughly 70% decline.[20] Approximately 150 validators are projected to lose Foundation stake under the May 2026 SFDP concentration rules (Chapter 8). The remaining ~770 are heavily concentrated in the named Equinix FR5 / Amsterdam / London anchor data centres (Frankfurt 19%, Amsterdam 16%, London 12% of stake — Chapter 8 anchor). The validator exit is the cleanest single piece of empirical evidence in the book that the structural concentration the chapters have documented is producing observable participant-loss: the small-validator category is not theoretical; it is exiting in real time.

The structural mechanism. The Syndica March 2026 measurement (Chapter 5, Chapter 7, Chapter 8) places per-block priority-fee captures at +33% to +101% above network median for validators running access-optimised configurations (Frankendancer with Harmonic Performance or Balanced strategy; Agave Harmonic). Chorus One's August 2025 measurement places the operational ceiling — block-production timing manipulation plus scheduler optimisation — at approximately +3% in total rewards (~27 basis points annualised). Placeholder VC's September 2025 measurement places the Solana validator profit Gini coefficient at approximately 0.93 — extreme inequality, more unequal than the most-unequal US states or the most-unequal countries globally. The bottom-quintile Solana validator is the operator at-or-below network median running stock Agave with no SFDP delegation, no BAM Node relationship, no Frankfurt colocation, no Harmonic strategy. The structural disadvantage is not one of operational discipline. It is one of access.

Hyperliquid's small-stake validator faces a different structural constraint.[21] Slot #21 in the active set requires approximately 525,000 HYPE (~$26 million at $50 per HYPE), even though the protocol's self-delegation minimum is only 10,000 HYPE locked for one year. The set is mid-expansion from 24 to 27 (Hyper Foundation announcement, 18 May 2026). Any operator can register a validator with 10,000 HYPE self-delegation; only the top 24 (expanding to 27) by total delegated stake form the active set. A validator entering "undelegate-only mode" if self-delegation drops below threshold. The 525,000 HYPE / 10,000 HYPE ratio is approximately 52× — a small-stake operator with the technical capacity to run a validator faces a stake-acquisition barrier roughly 52× her self-delegation minimum. The shape is different from Solana's (where the floor is operational economics) and Ethereum's (where the floor is the 32-ETH minimum, which Vitalik Buterin has proposed lowering to 1 ETH). On Hyperliquid the floor is delegated stake competition — a permissionless-but-zero-sum competition for the 21-to-27 active slots.

The Ethereum solo home staker is the named comparison.[22] Solo home stakers constitute approximately 5.4% of staked ETH (per the ethstaker June 2024 dataset) and earn approximately 4–5% APY versus Lido's 3.5–4% (Lido takes a roughly 10% fee; solo stakers keep it all). Three operators — Lido at 24.2% of staked ETH, Coinbase Cloud at approximately 5.1%, Binance at approximately 9.1% — control roughly 38% of the validator set. The structural argument: the Ethereum solo staker is a winner by raw yield but a loser by aggregate share and by asymmetric access to MEV-Boost optimal-relay routing, builder relationships, and the institutional staking-as-a-service infrastructure (Coinbase Cloud, Kiln, Figment, Lido's curated operator set, per Chapter 10). Vitalik Buterin's advocacy for lowering the staking minimum to 1 ETH is not a protocol commitment; the structural shape is set.

The structural reasons small validators continue to operate. (a) Hardware sunk costs — the one-time CapEx of $8,000–$15,000 for Solana validator hardware (Chapter 8) and the operational continuity that has accumulated around it. (b) Alpenglow option value — expected late-Q3/early-Q4 2026 mainnet activation with the VAT dropping the profitable-validator threshold from ~4,850 SOL to ~450 SOL (Chapter 8); the small-stake operator who survives the SFDP transition may see her economic viability restored. (c) Validator-set-membership option value — exiting forfeits the membership entirely; staying preserves it. On Solana, the exit is fast and visible (the 2,560 → 770 trajectory). On Hyperliquid, the entry is structurally bounded (the 52× capital ratio). On Ethereum, the participation is structurally subsidised by the protocol's solo-staker advocacy and the absence of an explicit minimum-stake-concentration rule.

The chapter's editorial contribution: the four loser categories differ in their informed-consent relationship to the structural disadvantage.

CategoryLoss magnitudeLoss visibility (Layer 1: per-trade)Loss visibility (Layer 2: per-layer trace)Consent type
Retail tradersSmall per-trade × high count (cumulative ~$8–25 per $10K on Solana / Ethereum L1)Visible per-trade (slippage shown by wallet UI)Invisible per-layer (RPC rebate, validator capture, builder margin)Uninformed — UX abstraction obscures the per-layer trace
Passive LPsContinuous LVR drag (~5–7% per year on ETH/USDC)Invisible (fees displayed, LVR not displayed)Mathematically visible to those who read Milionis et al.Partial — LVR research exists but is not surfaced at deposit time
Slow market makersDisplacement, not loss-per-trade (Phoenix $3.7M → $69K; 54× decline)Visible at firm level (volume share data)Visible at category level (Phoenix decline; EOF research)Informed — firms understand the structural mechanism
Validators without relationshipsOpportunity cost (top-quintile vs median per-block)Visible per-epoch (rewards data)Visible cross-validator (Gini ~0.93; +33–101% telemetry)Informed — operators have access to comparison data

The structural argument: the consent gap is largest for retail, and the consent gap is what makes the retail loser category structurally different from the other three. The other three categories include participants who know the structure and have made informed decisions (often constrained by lack of alternatives) to remain in it. The retail category includes a meaningful share of participants who do not know the structure — and whose continued participation cannot be straightforwardly described as consensual in the way the other categories' can.

The chapter does not argue that the structures are illegitimate. The chapter argues that the asymmetry of informed consent across the four loser categories is the most important structural fact about who-loses-and-why on-chain in 2026, and that the regulatory frameworks documented in earlier chapters — the EU PFOF ban (Chapter 7), the SEC's withdrawal of the Order Competition Rule (Chapter 7), the ETF wrappers (Chapters 9, 10) — do not address the consent gap because they regulate the asset-exposure layer rather than the trade-execution layer.

Why "just leave the table" is harder than it sounds ​

A clinical summary of the structural reasons each category does not exit.

Retail. The alternative is centralised-exchange access (often unavailable by jurisdiction; US retail post-2024 faces progressively restricted CEX options) or non-participation (which forfeits the asset exposure entirely). The behavioural finding from the FCA's 2025 UK study: disclosure rules reduce participation modestly but do not improve per-trade losses for the residual participants. The DEX-to-CEX spot volume ratio more than tripled between 2021 and 2025 — from 6.0% to 21.2% as of November 2025 — and spot DEX share reached approximately 14% of global spot in January 2026 after peaking at 24.5% in mid-2025.[23] The shift is in the opposite direction of "retail moves to safer venues." Retail is moving toward self-custody on DEXs, not away from them.

Passive LPs. The alternative is depositing nowhere — there is no comparable passive yield product at the capital scale of typical LP positions. The HLP is a structurally different option but requires the depositor to consent to clearinghouse-style tail exposure. Centralised lending products carry counterparty risk the LP is explicitly trying to avoid. Solutions exist at the protocol level — CoW AMM batch auctions, Panoptic, oracle-based AMM designs — but none has replaced the dominant V3-style design at scale.

Slow market makers. The alternative is pivoting to prop-AMM operation (which requires originating-trading-flow capability — a different business that requires capital, technology, and the operational stack only the largest firms have); or exiting market making entirely (which forfeits the firm's competitive position). Wintermute (Tessera V) and Temporal (HumidiFi) and Ellipsis Labs (Phoenix Perpetuals) are the named pivoters. The smaller firms — the ones running standalone quoting bots on the on-chain CLOBs that compressed — do not have the option to internalise prop-AMM economics because the operational stack requires multi-million-dollar capital outlay. The "just pivot" answer is available to the largest 5–10 firms. For the rest, the exit is to non-crypto trading or to a smaller crypto venue with worse economics.

Validators without infrastructure relationships. The alternative is exiting the validator set, which forfeits the staking yield itself and the option value of post-Alpenglow improved economics. On Solana the exit is fast and visible (the ~150-validator SFDP removal in May 2026; the 2,560 → 770 trajectory since 2023). On Hyperliquid the entry is structurally bounded (the 52× capital ratio for active-set membership). On Ethereum the participation is structurally subsidised by the protocol's solo-staker advocacy and the absence of an explicit minimum-stake-concentration rule, but the share-compression to Lido + Coinbase Cloud + Binance + Kiln continues.

The chapter's editorial reading: "just leave the table" is a behavioural injunction that ignores the structural costs of leaving for three of the four categories, and ignores the consent gap for the fourth.

What changes when… ​

Part IV's first chapter has been the chronic-loser categorical finding. Chapter 12 closes the book with the forward-looking trends — the in-protocol PBS arrival, the decentralised-sequencer roadmap (Espresso, Astria, Superchain), shared-state cross-chain interop, the appchain thesis, and the intent-based architecture trend (Anoma, CoW Swap, UniswapX). The question Chapter 12 must engage with: do any of these trends materially address the consent gap Chapter 11 has documented? The chapter's tentative answer is "in some categories yes, in retail not yet" — but the development belongs in Chapter 12. The book is one chapter away from its end, and the question of whether the structure changes is the closer's job.

Footnotes and sources ​

  1. CoinGecko, Pump.fun Traders Are Making a Comeback: Profitability Hits Two-Year High, https://www.coingecko.com/research/publications/pump-fun-traders-are-making-a-comeback; CoinInsider, Pump.fun Hits Two-Year High as 73% of Traders Turn Profitable in April 2026, https://www.coininsider.com/news/pump-fun-hits-two-year-high-as-73-of-traders-turn-profitable-in-april-2026; Phemex, CoinGecko Reports 73.3% of Pump.fun Traders Profitable in April 2026, https://phemex.com/news/article/coingecko-reports-733-of-pump-fun-traders-profitable-in-april-80114. Pump.fun profitable-wallet share: 30.1% (June 2025) → 73.3% (April 2026). Accessed 2026-05-14. ↩︎ ↩︎ ↩︎

  2. CoinGecko and CoinInsider sources per [1:1]. Pump.fun monthly active wallets: ~5.2M (May 2025) → ~1.8M (December 2025) → 3.14M (April 2026); 2.30M profitable in April 2026, with the largest profitable cohort (~2.05M, ~65% of profitable wallets) earning only $1–$500 for the month; ~793K wallets (~25% of active) lost between $1 and $500; ~46K (~1.5%) lost more than $500. Accessed 2026-05-14. ↩︎ ↩︎

  3. Milionis, Moallemi, Roughgarden, and Zhang, Automated Market Making and Loss-Versus-Rebalancing, arXiv 2208.06046, https://arxiv.org/pdf/2208.06046; CoW, What Is Loss-Versus-Rebalancing (LVR), https://cow.fi/learn/what-is-loss-versus-rebalancing-lvr; a16z crypto, LVR: Quantifying the Cost of Providing Liquidity to AMMs, https://a16zcrypto.com/posts/article/lvr-quantifying-the-cost-of-providing-liquidity-to-automated-market-makers/. LVR loss ~σ²/8 per unit time, ~11% per year to arbitrageurs before fees in a 5%-daily-volatility AMM; ~5–7% capital/year for ETH/USDC and BTC/USDC pairs. Already cited Chapter 2. Accessed 2026-05-14. ↩︎ ↩︎ ↩︎ ↩︎

  4. Hyperliquid Docs, Protocol Vaults — HLP, https://hyperliquid.gitbook.io/hyperliquid-docs/hypercore/vaults/protocol-vaults; Monolith VC, Hyperliquid After the JELLY Attack: Collapse or Comeback, https://medium.com/@monolith.vc/hyperliquid-after-the-jelly-attack-collapse-or-comeback-f380ce55472c. HLP TVL drawdown $500M → $177M (65% decline) during the 26 March 2025 JELLY incident; HLP closed positions at $0.0095 for a $700K HLP gain; recovered to ~$383M TVL by May 2026. Already cited Chapter 9. Accessed 2026-05-14. ↩︎ ↩︎

  5. Solana Foundation, Delegation Criteria, https://solana.org/delegation-criteria; Phemex News, Solana Foundation Updates Validator Delegation Requirements for May 2026, https://phemex.com/news/article/solana-foundation-updates-validator-delegation-requirements-for-may-2026-69090; Blockworks, Solana Foundation Begins Pruning Validators, https://blockworks.co/news/solana-foundation-pruning-validators-delegation; Helius, Solana Foundation Delegation Program and Challenges for Validators, https://www.helius.dev/blog/solana-foundation-delegation-program-sfdp. May 2026 SFDP rules: <25% ASN/hosting-provider concentration; <15% data-centre concentration; ~150 validators projected to lose Foundation stake. Already cited Chapters 5, 8. Accessed 2026-05-14. ↩︎ ↩︎

  6. Aramonte, Huang, and Schrimpf, Crypto trading and Bitcoin prices: evidence from a new database of retail adoption, BIS Bulletin No 69, https://www.bis.org/publ/bisbull69.pdf; Investment Executive, Most Retail Investors Lose on Bitcoin Trades: BIS, https://www.investmentexecutive.com/news/research-and-markets/most-retail-investors-lose-on-bitcoin-trades-bis/; Ecofin Agency, Bitcoin Retail Investors Lost 47.89% of Invested Funds on Average, https://www.ecofinagency.com/finance/1104-44420-bitcoin-retail-investors-lost-47-89-of-invested-funds-on-average-bis-bulletin. BIS Bulletin 69 (2022): ~73–81% of retail crypto-app users likely lost money on their initial Bitcoin investment; average retail loss ~47.89% of invested funds. Coverage 95 countries, Aug 2015 – Dec 2022. Accessed 2026-05-14. ↩︎

  7. CoinGecko, Pump.fun Traders Are Making a Comeback, op. cit.; Phemex, CoinGecko Reports 73.3% of Pump.fun Traders Profitable, op. cit. Detail per [1:2] and [2:1]. Accessed 2026-05-14. ↩︎

  8. CoinGecko, Top Telegram Trading Bots, https://www.coingecko.com/learn/top-telegram-trading-bots; BlockBase, Banana Gun Research, https://insights.blockbase.co/banana-gun-research/. Banana Gun ~1.3M users; ~25.3M executed trades; $16.09B cumulative trading volume; ~$635 average trade size; 0.5%–1% per-trade fee. Already cited Chapter 7. Accessed 2026-05-14. ↩︎

  9. Cross-reference Chapter 1 framing of Coinbase's US retail user base; the order-of-magnitude comparison is the chapter's structural observation rather than a single-source claim. Accessed 2026-05-14. ↩︎

  10. Financial Conduct Authority, Cryptoassets Consumer Research 2025 (Wave 6), https://www.fca.org.uk/publications/research-notes/cryptoassets-consumer-research-2025; FCA, Cryptoasset Consumer Research 2025 Wave 6 PDF, https://www.fca.org.uk/publication/research-notes/cryptoasset-consumer-research-2025-wave-6.pdf; CoinDesk, Number of Crypto Users in the UK Drops Even as Amount Held Increases, https://www.coindesk.com/markets/2025/12/16/number-of-crypto-users-in-the-uk-drops-even-as-amount-held-increases. UK crypto ownership 12% (2024) → 8% (2025); ~7M → ~4.5M users; average holdings rose to ~£1,842; awareness remained 91%; 73% of UK users purchased through CEXes (up 4% from 2024). Accessed 2026-05-14. ↩︎

  11. Milionis et al., arXiv 2208.06046, op. cit.; CoW source per [3:2]; a16z crypto source per [3:3]. Already cited Chapter 2. Accessed 2026-05-14. ↩︎

  12. CryptoSlate, New Report Shows 50% of Uniswap V3 Liquidity Providers Are Losing Money, https://cryptoslate.com/new-report-shows-50-of-uniswap-v3-liquidity-providers-are-losing-money/. Topaze Blue + Bancor 2021 study: ~49.5% of V3 LPs realised negative returns net of impermanent loss across 17 pools (~43% of V3 TVL); $199.3M in collected fees vs $260.1M in IL; collectively ~$60.8M worse off than HODL. Already cited Chapter 2. Accessed 2026-05-14. ↩︎

  13. Uniswap Labs, When Uniswap V3 Returns More Fees for Passive LPs, https://blog.uniswap.org/fee-returns. V3 ETH/USDC 5-bps tier underperformed V2 by ~68%; across all tiers V3 LPs earned ~54% higher fees on average but the dominant pool was the worst case. Already cited Chapter 2. Accessed 2026-05-14. ↩︎

  14. KuCoin, Maximizing the Liquidation Alpha: How Hyperliquid's HLP Vault Converts Whale Losses into LP Yield, https://www.kucoin.com/news/articles/maximizing-the-liquidation-alpha-how-hyperliquid-s-hlp-vault-converts-whale-losses-into-liquidity-provider-yield; ARX, Hyperliquid Vaults Explained, https://arx.trade/blog/hyperliquid-vaults-explained/; cross-ref Chapter 9 sources per [4:1]. HLP historical returns ~1.75%/month / ~20% annualised through 2025; $700M February 2026 liquidation generated ~$15M HLP profit. Accessed 2026-05-14. ↩︎

  15. CrocSwap, Discrimination of Toxic Flow in Uniswap V3 Part 4, https://crocswap.medium.com/discrimination-of-toxic-flow-in-uniswap-v3-part-4-c09656ec016e; Sandmark, Toxic Flow: The Hidden Cost of Providing Liquidity, https://www.sandmark.com/news/features/toxic-flow-hidden-cost-providing-liquidity. Toxic flow defined as arbitrage order flow forcing the LP to buy/sell against the market; aggregator-routed flow (Jupiter, CoW Swap) is a credible signal of non-toxicity not easily gameable. Accessed 2026-05-14. ↩︎

  16. TechFlow, BTC Liquidity Plummets 90% — Hyperliquid Market Makers Withdraw Coverage, https://www.techflowpost.com/en-US/article/31620. Mid-May 2026: Wintermute and Auros Global simultaneously withdrew ~$100M of aggregate liquidity from Hyperliquid; Wintermute's combined BTC + ETH resting liquidity dropped from ~$40M to ~$4M (~90% decline). Withdrawal occurred three days after CME and ICE jointly pressured US regulators to review Hyperliquid. Accessed 2026-05-14. ↩︎

  17. DefiLlama, Phoenix Protocol Page, https://defillama.com/protocol/phoenix; Ellipsis Labs, Introducing Phoenix Perpetuals, https://www.ellipsislabs.xyz/blog-posts/introducing-phoenix-perpetuals. Phoenix peak Q2 2024 revenue $3.7M; Q1 2026 revenue $68,604; ~54× decline; renamed Phoenix Legacy; Ellipsis Labs pivoted to perpetuals (Phoenix Perpetuals, announced Breakpoint 2025). Already cited Chapters 2, 8, 9. Accessed 2026-05-14. ↩︎

  18. Wu et al., arXiv 2405.01329, https://arxiv.org/html/2405.01329v4; Pahari and Canidio, arXiv 2509.16052, https://arxiv.org/abs/2509.16052. 75 EOF arrangements account for ~71% of trading-related Ethereum builder revenue; 77.2%–84% of fees in winning Ethereum blocks come from exclusive transactions; only ~7% of exclusive value comes from senders routing exclusively to one builder (most exclusivity is dynamic). Already cited Chapter 7. Accessed 2026-05-14. ↩︎

  19. Syndica, Deep Dive: Solana Onchain Activity, March 2026, https://blog.syndica.io/deep-dive-solana-onchain-activity/; Chorus One, Timing Games on Solana, https://chorus.one/reports-research/timing-games-on-solana-validator-incentives-network-impacts-and-agaves-hidden-inefficiencies; Placeholder VC, Leveling the Stakes on Solana, https://www.placeholder.vc/blog/2025/9/15/leveling-the-stakes-on-solana. Per-block priority-fee captures vs network median: Frankendancer Harmonic Performance +101%, Balanced +39%, Agave Harmonic +36%. Combined timing-games + scheduler optimisation: ~+3% total rewards (~27 bps annualised). Solana validator profit Gini ~0.93. Already cited Chapters 5, 7, 8. Accessed 2026-05-14. ↩︎

  20. CCN, Solana Validators Plunge 68% in 3 Years, https://www.ccn.com/news/crypto/solana-validators-plunge-68-3-years-dead-chain-sybil-purge/; Helius source per [5:1]. Solana active-validator count ~2,560 (mid-2024) → ~770 (March 2026); ~70% decline. Accessed 2026-05-14. ↩︎

  21. Hyperliquid Docs, Staking, https://hyperliquid.gitbook.io/hyperliquid-docs/hypercore/staking; cp0x on X, Slot 21 ~525K HYPE, https://x.com/cp0xdotcom/status/1920482476821991466; Coin Edition, Hyperliquid to Increase Validators 24 → 27, https://coinedition.com/hyperliquid-to-increase-validators-amid-transparency-and-security-discussions/. Slot #21 requires ~525,000 HYPE delegated; self-delegation minimum 10,000 HYPE locked for one year; set mid-expansion from 24 to 27 over 30-day onboarding window. Already cited Chapters 5, 9. Accessed 2026-05-14. ↩︎

  22. CoinDesk, Figment Outpaces Rivals in Ether Staking Growth; Lido's Decline Eases Dominance Concerns, https://www.coindesk.com/tech/2025/08/14/figment-outpaces-rivals-in-ether-staking-growth-lido-s-decline-eases-dominance-concerns; Datawallet, Ethereum Staking Statistics 2026, https://www.datawallet.com/crypto/ethereum-staking-statistics-and-trends; AInvest, Ethereum Staking 2026: Lido's 10% Fee Drag vs Solo Stakers' 4–5% Yield Edge, https://www.ainvest.com/news/ethereum-staking-2026-lido-10-fee-drag-solo-stakers-4-5-yield-edge-ignites-liquidity-conviction-battle-2603/; Cointelegraph, Vitalik Buterin Advocates Lowering Solo Staking ETH, https://cointelegraph.com/news/vitalik-buterin-advocates-lowering-solo-staking-eth; ethstaker, Solo Stakers Dataset, https://github.com/ethstaker/solo-stakers. Solo home stakers ~5.4% of staked ETH (June 2024 dataset); earn 4–5% APY vs Lido ~3.5–4%; Lido at ~24.2%, Coinbase Cloud at ~5.1%, Binance at ~9.1% — three operators control ~38% of validator set. Already cited Chapters 5, 10. Accessed 2026-05-14. ↩︎

  23. CCN, Crypto Users Are Shifting from CEXs to DEXs — Here's Why, https://www.ccn.com/news/crypto/crypto-users-are-shifting-from-cexs-to-dexs-heres-why/; CoinGecko, DEX to CEX Spot Volume Ratios, https://www.coingecko.com/research/publications/dex-to-cex-ratio; FCA, Cryptoassets Consumer Research 2025, op. cit. DEX-to-CEX spot volume ratio: 6.0% (2021) → 21.2% (November 2025); spot DEX share reached ~14% of global spot in January 2026 after peaking at 24.5% in mid-2025. Accessed 2026-05-14. ↩︎