TL;DR
Papertrade is a synthetic perpetuals exchange on Hyperliquid's HyperEVM, offering only BTC and ETH markets with up to 1000x leverage. Users trade against a USDC pool that started at $0, and losing traders receive newly minted $PAPER. In its first 9 days, cumulative volume reached $1.35T across 22,869 users. DefiLlama reports cumulative revenue of $60.79M, of which $56.14M has been distributed to $PAPER stakers. By revenue relative to TVL, it is currently one of the most profitable on-chain protocols.
The model, however, essentially pits traders against the House:
- 85% of revenue comes from retained user losses: $51.5M of LP Surplus.
- House funds currently amount to only $3.38M, against approximately $99M of users' net long exposure.
- A simultaneous BTC/ETH rise of approximately 3–4% could exhaust House funds, putting profitable traders' withdrawals into a queue.
- $PAPER currently cannot be transferred and has no market price.
- Revenue is falling rapidly: approximately $9.3M per day over the first 3 days, versus only $2.37M over the latest 24h.
For an ordinary user, this looks more like a casino offering rebates on losses than an airdrop farm with an arbitrage opportunity.
Mechanics and Team
Papertrade's central design is an exchange with no order book, where users trade against a bootstrapped LP. At both opening and closing, contracts use a precompile to read Hyperliquid's BBO midpoint. Users enter synthetic swaps against the LP; no actual order is placed on Hyperliquid. There is therefore no funding, slippage, or spread, only an OI cap for each asset. Papertrade Docs
- Direct LP deposits are not permitted: the LP grows only through user losses. When it cannot pay, profitable traders receive queued debt, repaid on a first-come, first-served basis using subsequent losses.
- Fees are asymmetric: losing traders pay no additional fee. Profitable closes incur an “asymmetric impact” discount on profits: the smaller the price move, the larger the discount. A 2% fee then applies to the remainder. Profitable moves below 0.2bps receive no payout. airdrops.io
- Stakers have two revenue sources: (1) 1% of settled trade PnL; (2) all LP funds exceeding the $5M cap. Papertrade Docs
The cofounders are two well-known, semi-anonymous traders, Jez (@izebel_eth) and Blurr (@blurr). Bankless
- Contracts are upgradeable, controlled by a 2-of-3 multisig comprising izebel_eth, blurr, and UngusTrade, with a 7-day timelock. airdrops.io
- In May 2026, the project announced it had engaged Guardian Audits. IQ.wiki Third-party summaries, however, note that the documentation contains no public audit report. This needs independent checking.
- The project has no VC backing or fundraising record.
| Dimension | Situation | Assessment |
|---|---|---|
| Team | Jez and Blurr, semi-anonymous traders | Strong community influence; opaque legal entity |
| Contract permissions | 2-of-3 multisig, 7-day timelock | Centralized upgrade authority |
| Audit | Guardian Audits engagement announced; no public report seen | Medium risk |
| Funding | No VC; fair launch | No selling pressure from unlocks |
Trading and On-Chain Data
Volume has been enormous since launch, but positions and TVL have declined significantly from their launch peaks. The official Dashboard snapshot at approximately 06:15 UTC on 2026-10-19 shows:
| Metric | Value | Interpretation |
|---|---|---|
| Cumulative volume | $1.35T | $1.24T on 10-17; approximately $55B per day over the past two days |
| Cumulative liquidations | $40.7B | 3.0% of volume |
| Trade count | 1,460,633 | Approximately $924K average notional per trade |
| Users | 22,869 | 21,523 on 10-17; approximately 670 added per day |
| Open interest (OI) | $458.5M | $11B at launch, $693M the following day; continuing to contract |
| Open-position margin | $9.93M | Approximately 46x average effective leverage |
| TVL | $64.5M | ATH of $102.09M on 10-11 |
The materials provide only the following TVL observations, which cannot reconstruct daily changes between them:
| Observation | TVL |
|---|---|
| 2026-10-11 all-time high | $102.09M |
| 2026-10-19 snapshot at approximately 06:15 UTC | $64.5M |
Users are net long overall, leaving the House with exposure far larger than its funds. BTC positions total $203M long and $119M short, with net longs of $84.2M; ETH net longs are $15.2M. The House takes the other side of approximately $99M in total net longs, with only $3.38M in funds (LP balance of $4.69M and unrealized PnL of −$1.31M). A rough estimate suggests a simultaneous BTC/ETH rise of approximately 3.4% would exhaust House funds and trigger queued debt. Current queued debt is $0.
The $4.69M LP balance is also below the $5M distribution cap. Stakers therefore currently receive only the 1% PnL share, with no overflow.
Revenue sources and distribution, according to DefiLlama as of 2026-10-19:
| Revenue item | Amount | Distribution |
|---|---|---|
| LP Surplus (users' net losses) | $51.5M | All to stakers |
| Fees from profitable traders | $7.95M | 50% to developer addresses, or $3.98M |
| Liquidation fees | $1.33M | 50% to developer addresses, or $665.8K |
| Total | $60.79M | $56.14M to stakers; approximately $4.65M to the team (7.6%) |
Revenue decay is pronounced:
- Approximately $27.9M in the first 3 days, averaging approximately $9.3M per day.
- $30.56M over the latest 7 days, averaging approximately $4.4M per day.
- $2.37M over the latest 24h.
DefiLlama's “$2.37B annualized” extrapolation from cumulative data severely overstates sustainable revenue. DefiLlama
At wallet level, the Paperdash leaderboard shows leading traders with volume concentrated in the $0.7B–$1.9B range: for example, 0x7827…4756 has $1.89B and 0x0f72…a669 has $1.38B. Each wallet accounts for less than 0.2% of total volume, with no extreme concentration evident. These figures come from the Paperdash leaderboard, not an independent, comprehensive HyperEVM on-chain analysis. Paperdash
PAPER Token Supply
$PAPER is an ERC-20 token on HyperEVM with a fully fair launch: no premine, team or VC allocation, airdrop, or unlocks. Every token is minted when a user realizes a loss. Papertrade Docs
| Metric | Value (2026-10-19) |
|---|---|
| Total supply | 8,677,225,186 |
| Staked | 8,612,435,206 (99.25%) |
| Cumulative staking rewards | $54.06M USDC |
| Cumulative distribution per token | Approximately $0.0062 |
The minting curve works as follows: when the LP is below $2M, each $1 lost mints 100 PAPER. Above that level, the rate declines according to 100 × (120M / (120M + H))², where H is the pool's historical peak earnings and can only increase. On 10-17, H was approximately $222.7M, implying approximately 12.3 tokens minted per $1 lost. The acquisition cost per PAPER had thus risen from approximately 1 cent at launch to approximately 8 cents. airdrops.io
Important concerns:
- Early tokens acquired for 1 cent share the same reward pool as tokens now acquired for approximately 8 cents, substantially diluting returns for later entrants.
- Cumulative distributions per token are currently approximately 0.6 cents, below the current acquisition cost.
- If the LP falls below $2M, issuance returns to 100 tokens per $1, creating another round of dilution.
⚠️ Imitation-token warning: $PAPER currently cannot be transferred and has no official DEX pair. A token named PAPER exists on Solana with $0 in 24h volume; Coinbase also has a “papertradexyz” page labeled as Solana. Neither is the official token. Do not buy them.
Risk Incident and Main Risks
Oracle manipulation has already occurred. On launch day, 10-10, wallet 0x2025...840e traded ETH in size on Hyperliquid, moving its BBO price by approximately 0.12%. It used 1000x leverage on Papertrade to earn approximately $1.28M on $3.18B in volume, while paying only approximately $52K in Hyperliquid fees. The team subsequently tightened OI limits on-chain and at the relay layer. The Block via TradingView DefiLlama records this as an Oracle Manipulation security incident.
| Risk | Severity | Explanation |
|---|---|---|
| Queued payouts | High | House funds of $3.38M against approximately $99M net exposure; a one-way market can trigger queued debt, with no fixed payout time |
| Oracle manipulation | High | Large trades can move the BBO midpoint; the official risk page acknowledges this |
| No PAPER liquidity | High | Nontransferable, no price, and no scheduled date for enabling transfers |
| Dilution from hedged token farming | Medium | Delphi's tempest notes that long/short hedging can mint PAPER at low cost Bankless |
| Falling revenue | Medium | Daily revenue fell from approximately $9.3M to $2.37M |
| Centralization and infrastructure | Medium | A multisig can upgrade contracts; dependence on HyperEVM blockspace, with the Dashboard showing a 3M gas budget per block and 9.62M queued |
| Extreme leverage | High | At 1000x leverage, an adverse move of approximately 0.052% triggers full liquidation |
Conclusion
Papertrade's mechanism is genuinely novel. It passes 100% of House profits to token holders; the team receives only 50% of trading and liquidation fees, approximately 7.6% of total revenue. Within 9 days, it generated $60M in revenue and $1.35T in volume. Yet its revenue comes almost entirely from retail losses, without a sustainable fee-based moat. House funds and user exposure are severely mismatched, and oracle manipulation has already occurred. Revenue is falling rapidly, while retaining users depends on a continuing inflow of new loss-bearing funds.
What to watch: Focus on four indicators:
- Whether House funds can remain above $5M, determining whether stakers can resume receiving overflow distributions.
- Whether queued debt appears.
- When PAPER transfers are enabled, and whether its first market price exceeds the approximately 8-cent acquisition cost.
- Whether daily revenue can stabilize above $2M.
If you already trade BTC/ETH with high leverage, PAPER can be treated as a rebate on losses; remember to stake it. Entering solely to farm PAPER, however, means exchanging real losses for a token without a market, whose distributions depend on the next group of losing traders. I do not recommend doing so.
Data snapshot: Papertrade Dashboard and DefiLlama, 2026-10-19 at approximately 06:15 UTC. This content is not investment advice.
Source note: This article is organized from user-provided research, preserving the Dashboard and DefiLlama snapshot at approximately 06:15 UTC on 2026-10-19 and its original source conventions. No Surf calls or independent online verification were made. The 3.4% House stress estimate, token acquisition costs, and annualized revenue are conditional estimates from the materials, not a payout guarantee, market price, or forecast of future revenue. The original TVL chart placeholder has been replaced with the two observations supported by the materials; no historical series has been fabricated.