Papertrade Price Manipulation Controversy: How Hyperliquid BBO Risks and Position Limits Work

Papertrade Price Manipulation Controversy: How Hyperliquid BBO Risks and Position Limits Work

Papertrade Tightens Trading Limits Following Leveraged Price Manipulation

Papertrade, a synthetic perpetuals venue launched on HyperEVM on October 10, 2026, faced quick scrutiny after on-chain data showed a single wallet realizing approximately $1.28 million in profits within hours by opening large leveraged positions and simultaneously trading thousands of ETH on Hyperliquid to shift the reference price. The platform prices every open and close at the midpoint of Hyperliquid’s best bid and offer, settles against a protocol-owned USDC liquidity pool, and offers up to 1000x leverage with no funding rates or conventional slippage. Papertrade’s own documentation had already identified best-bid-and-offer manipulation as a primary unresolved protocol-level risk. Within a day, the team announced it was tightening open-interest caps both on-chain and at the relayer level, attributing the vulnerability window to deliberately permissive launch parameters intended to manage congestion.
 
The pattern shows the concrete economic consequences when a high-leverage synthetic contract inherits a single order-book midpoint without independent deviation checks, time-weighted averaging, or multi-venue aggregation. The Papertrade incident demonstrates that reliance on an unfiltered Hyperliquid BBO midpoint, combined with initially loose directional open-interest headroom of roughly $1 billion per side and a $10 million per-position notional ceiling, created a profitable path for coordinated price influence that the subsequent tightening of those same caps is now designed to render nearly unprofitable.

Launch Congestion and Permissive Open-Interest Headroom Created the Initial Window

Papertrade trading opened at 10 a.m. ET on October 10, 2026, on HyperEVM and immediately attracted substantial notional volume, with Bitcoin open interest reported near $3 billion and Ether near $2.66 billion within minutes. To manage frontend congestion and spam, the protocol routed activity through a relayer and set directional open-interest headroom at current open interest plus $1 billion of additional exposure on each side of each market. A keeper was programmed to leave that limit unchanged while remaining inside a 10 percent band between $900 million and $1.1 billion. Individual atomic positions were capped at $10 million notional, yet traders could open multiple positions.
 
The combination allowed aggregate directional exposure far larger than the liquidity pool, which began near zero and was funded solely by trader losses. Because confirmation times were uncertain under heavy load, timing a precise open-move-close sequence was more difficult, yet the size of available headroom still permitted positions large enough that a 0.1 percent reference-price shift at 1000x leverage could equal the full posted margin. Papertrade later stated that the permissive settings were intentional to keep queues moving and that tighter caps during peak demand would have slowed the experience further. Once congestion eased, the same parameters became the first control the team adjusted.

Hyperliquid BBO Midpoint Functions as the Sole Settlement Oracle

Papertrade’s smart contract does not place any corresponding perpetual order on Hyperliquid. Instead, it creates a synthetic swap between the trader and the protocol’s liquidity pool. At the moment a position opens, the contract reads the best bid and best ask through Hyperliquid’s BBO precompile at address 0x000000000000000000000000000000000000080e and records the midpoint as the entry price. The same midpoint is read again at close, and the resulting profit or loss is settled directly against the pool. The precompile returns consensus-verified HyperCore state with no off-chain signer or separate oracle committee.
 
The oracle rejects a zero value or a crossed book, yet it contains no maximum-deviation check, no time-weighted average, and no lookback breaker. Consequently, any non-zero, uncrossed quote, even one shifted by a single unfilled limit order placed inside the natural spread, can pass and become the settlement price. Documentation published before launch explicitly listed this path as a major protocol-level risk. Because the pool, rather than another trader, bears every profitable close, a brief displacement of the midpoint directly transfers value from liquidity providers to the trader who timed the move.

Documented BBO Manipulation Path Relies on Small Orders Moving the Inside Quote

Papertrade’s risk documentation describes a lower-cost variant of the same vulnerability: an attacker can rest a limit order inside Hyperliquid’s natural spread without expecting a fill, thereby changing the best bid or best ask by one or two ticks and shifting the midpoint that Papertrade will later use for settlement. Once the Papertrade position is closed against the displaced midpoint, the order on Hyperliquid can be cancelled. At 1000x leverage, a 0.1 percent move equals the entire margin of a fully leveraged position, so the economic incentive scales with position size rather than with the capital required to move the quote.
 
Hyperliquid’s own order book depth for perpetual bids has at times been thinner than major centralized venues, increasing the relative impact of sizeable market or limit orders. The absence of an independent reference price means Papertrade cannot reject a manipulated but still a valid BBO. Economic mitigations such as a 0.2-basis-point deadband and an asymmetric-impact haircut that applies the largest reduction to the smallest profitable moves reduce the net payoff yet do not eliminate the possibility that a carefully timed sequence remains profitable when open-interest headroom is large.

On-Ledger Records Show $1.28 Million Realized from Coordinated ETH Trades

On-chain trading data and Papertrade account records reviewed by multiple outlets show that wallet 0x2025\ldots840e opened large long and short positions on Papertrade and simultaneously bought or sold thousands of ETH on Hyperliquid. Between 6:01 p.m. and 6:02 p.m. ET on October 10, the wallet purchased approximately 6,060 ETH, lifting the price by roughly $3 or 0.12 percent; one minute later, it sold about 5,210 ETH, and the price fell approximately $3.50. Later the same evening, it opened thirty separate ETH short positions at 660x leverage totaling nearly $297 million notional and sold roughly 6,590 ETH, driving the reference price lower. The wallet traded approximately $231 million in notional ETH on
 
Hyperliquid that day and paid about $52,000 in fees. Papertrade portfolio pages displayed $1.28 million in cumulative realized profit on $3.18 billion of volume, with withdrawals of roughly $2.48 million against $1.2 million in deposits. A subsequent set of thirty short positions that remained open more than three hours was liquidated, producing a $450,000 loss that reduced earlier gains. The final net figure reported across independent analyses remained approximately $1.28 million.

Per-Position Notional Caps Were Bypassed by Splitting Exposure

Although each atomic position was limited to $10 million notional, the protocol permitted a single trader to hold multiple concurrent positions. The $297 million short exposure was therefore constructed as thirty separate positions of approximately $9.9 million each, remaining just under the individual ceiling. Directional open-interest headroom of roughly $1 billion still contained sufficient unused capacity, so the aggregate exposure did not trigger an immediate revert. Because settlement occurs against the shared liquidity pool rather than against matched counterparties, the economic impact of the price displacement was concentrated on the pool rather than diluted across many traders.
 
Papertrade’s later statement confirmed that the launch-period caps had been set loosely precisely to accommodate expected demand and that the same parameters were the first levers adjusted once congestion declined. The episode illustrates that a per-position limit alone does not constrain total directional risk when multiple positions can be opened under a generous aggregate headroom.

Tightening of On-Chain and Relayer Caps Aims to Shrink Attack Profitability

At 15:42 UTC on October 11, Papertrade publicly acknowledged overnight manipulation attempts against its BTC and ETH markets and stated that it had begun tightening open-interest caps both in the smart contract and at the relayer level. The team emphasized that the activity was “not an oversight in the mechanism design” developed over six months but rather a consequence of the elevated launch headroom. Lower directional caps reduce the maximum position size that can be opened against a moved quote, thereby lowering the expected payoff of any successful displacement. Caps continue to operate separately for long and short sides; an opening trade that would push either side above its limit simply reverts.
 
Papertrade noted that one user who initially realized gains later returned nearly all of that profit through new accounts, and that some participants agreed to reverse trades characterized as manipulated in exchange for retaining farmed PAPER tokens. The statement did not release wallet addresses, transaction hashes, or a quantified pool loss figure. Operators retain additional controls: a market keeper can reduce a side’s cap to zero without freezing the live midpoint used for closes, and a pause guardian can retire a market identifier while still allowing existing positions to close at the frozen price.

Asymmetric Impact Haircut and Deadband Reduce Net Payoffs on Small Moves

In addition to position limits, Papertrade applies two economic filters at settlement. A 0.2-basis-point deadband discards extremely small gains before profit calculation proceeds. An asymmetric-impact curve then imposes its largest haircut on the smallest profitable price moves, so that a trader who captures only a few ticks receives a proportionally larger reduction in realized gain. A 2 percent win fee is also charged on profitable closes after the haircut. These mechanisms lower the net extraction possible from a marginal BBO displacement, yet documentation acknowledges that they do not prove manipulation is unprofitable when larger moves or larger positions remain available.
 
Because the liquidity pool is the sole counterparty, every dollar paid out after haircuts still reduces the bankroll available for other winners and for eventual distribution to PAPER stakers once the pool exceeds its reward cap. The scheduled increase of that reward cap from $5 million to $10 million after 18:45 UTC on October 17 will alter the timing of excess distribution, making the effectiveness of the newly tightened directional caps a near-term test of residual risk.

Liquidity Pool Started Near Zero and Grew Solely from Trader Losses

Papertrade’s liquidity pool begins at zero and expands exclusively from the realized losses of traders; winning positions are paid from the same pool. When the pool cannot immediately cover a profitable close, the unpaid profit enters a first-in, first-out queue that is later refilled by subsequent losses, while the original margin is returned at once. On the day after launch, the pool stood at approximately $5 million, and the protocol had distributed 25 million PAPER to stakers. Snapshots around the time of the reported activity showed pool balances fluctuating between roughly $3.14 million and $4.30 million with an empty payout queue.
 
Because a successful BBO displacement causes the pool to pay out against a moved price, the capital available to future winners and to stakers is directly affected. The design therefore concentrates both the upside of high leverage and the downside of reference-price influence onto a single shared bankroll rather than distributing it across matched counterparties.

Hyperliquid Native Safeguards Differ from Papertrade’s Settlement Reference

Hyperliquid’s own perpetual markets employ separate oracle and mark-price systems for margin calculations, unrealized profit-and-loss, and liquidations. Those mechanisms aggregate information beyond a single order-book midpoint and are not the reference used by Papertrade. Consequently, the reported activity does not constitute a breach of Hyperliquid’s native risk controls; it concerns only the way Papertrade chose to import the BBO midpoint without additional filters. Hyperliquid order-book depth for perpetual bids has on occasion been thinner than that of major centralized venues, which can amplify the price impact of sizeable orders.
 
Papertrade’s decision to treat the midpoint as authoritative for both entry and exit therefore imports that depth characteristic directly into its settlement process. Analysts have suggested that substituting Hyperliquid’s official multi-venue oracle price for the raw BBO midpoint would remove the single-book vulnerability, yet such a change would also alter the zero-slippage, zero-funding character that defines the current product.

High-Leverage Synthetic Design Amplifies Small Reference-Price Displacements

At 1000x leverage, a 0.1 percent movement in the reference price equals the full margin posted on a position. When that reference price is taken from a single order-book midpoint that can be influenced by relatively modest capital, the economic return on the capital used to move the quote can become highly asymmetric. The same leverage also multiplies the absolute size of any loss borne by the liquidity pool. Papertrade’s zero-slippage execution at the midpoint removes the usual market-impact cost that would otherwise deter large directional bets, further concentrating risk on the accuracy of the single price source.
 
The combination of extreme leverage, synthetic settlement against a shared pool, and an unfiltered external midpoint therefore creates a structural incentive for coordinated activity whenever open-interest headroom is large relative to the cost of moving the quote. Tightening the headroom reduces the maximum size of positions that can be opened against any given displacement and is the primary parameter adjustment Papertrade has implemented so far.

Parameter Adjustments Continue While Residual Risk Remains Acknowledged

Papertrade has stated that no system can completely eliminate manipulation and that its objective is to set parameters that render attempts rarely profitable. The tightening of directional open-interest caps on-chain and at the relayer level is the first concrete step taken after the launch-day activity. Additional operator controls, market-specific cap reductions, market retirement with frozen midpoints for existing positions, and a global gate that halts new opens while allowing closes, remain available.
 
The forthcoming increase in the liquidity-pool reward cap from $5 million to $10 million will change the threshold at which excess bankroll becomes distributable to stakers. Whether the revised caps sufficiently constrain future attempts will be observable in subsequent open-interest levels and in any further public statements quantifying residual exposure. Documentation continues to list BBO manipulation among the protocol’s unresolved risks, underscoring that parameter tuning rather than architectural redesign is the current mitigation path.

Outlook for Traders and Liquidity Providers After the Incident

Traders evaluating the venue must now weigh the reduced maximum position size against the still-available leverage and the continued dependence on a single midpoint. Liquidity providers and PAPER stakers face the reality that any successful future displacement, however less profitable under tighter caps, still draws from the shared pool that ultimately funds rewards. The absence of independent deviation checks means that even after the recent adjustments a non-zero, uncrossed BBO remains authoritative.
 
Outlooking of real-time open-interest headroom, pool balance, and any further parameter announcements therefore becomes part of ordinary risk assessment. The episode supplies a concrete data point for other synthetic or high-leverage designs that inherit external order-book midpoints: the economic cost of an unfiltered reference price scales directly with the leverage offered and with the size of the directional exposure the protocol permits.

🔥 Beyond the Headlines: What KuCoin 5.0 Means for You

Market news moves fast — but where you act on it matters just as much. This October, KuCoin launches KuCoin 5.0, transforming KuCoin into a rebuilt platform. Here's what actually changes for you:
 
  • One account for everything. Older platforms split your money across separate "spot," "margin," and "futures" accounts and expected you to understand why. KuCoin 5.0's unified account removes that entirely — deposit once, and everything is simply there (only available to VIPs for now).
  • Stocks, indices, and commodities. KuCoin 5.0 expands beyond crypto into global markets. When crypto chops sideways and equities rally (or the reverse), you rotate in minutes instead of opening a brokerage account and waiting days for fiat rails.
  • Real-world assets (RWA). Tokenized exposure to traditional assets like commodities, right inside your crypto account. One of the fastest-growing segments in global finance is no longer reserved for institutions — you access it from the same balance you trade with.
  • Earn while you learn. Not ready to trade? KCUSD lets your stablecoins earn daily, auto-compounding interest. The lowest-stress way to put your idle deposit to work for 4% yield.
  • An AI assistant in plain language. Ask questions, get market context, understand what you're looking at — built into the platform, no jargon required.
  • An app that doesn't overwhelm. Faster, cleaner, and consistent — intuitive from the first tap, not after a tutorial.
  • Safety you can check, not just trust. A MiCAR-licensed EU entity, Proof of Reserves you can verify yourself, and internationally certified security (SOC 2 Type II, ISO 27001:2022).
 
Create your account in minutes — and start on the platform built for where crypto is going, not where it's been.

FAQs

How does Papertrade determine the price at which positions open and close?

Papertrade’s smart contract reads the midpoint between Hyperliquid’s best bid and best ask through a dedicated precompile at the exact moment a position is opened and again at the moment it is closed. That single midpoint becomes both the entry and the exit price. No separate oracle committee, multi-venue median, or time-weighted average is applied. The design produces zero conventional slippage and eliminates funding rates because no actual Hyperliquid perpetual is ever traded, yet it also means any temporary displacement of the Hyperliquid top of book is imported directly into settlement.
 

What specific open-interest limits were in place at launch?

At launch each market and each side carried approximately $1 billion of additional open-interest headroom above existing open interest. A keeper left the limit unchanged while headroom stayed inside a 10 percent band between $900 million and $1.1 billion. Individual atomic positions were limited to $10 million notional, but multiple concurrent positions were allowed. These figures were later reduced on-chain and at the relayer level after the reported activity.
 

Why did splitting positions into thirty separate trades matter?

The $10 million per-position ceiling could be observed by opening thirty positions of roughly $9.9 million each, producing an aggregate short exposure near $297 million. Because directional headroom remained available, the protocol did not reject the set of opens. The economic exposure to the subsequent price move therefore scaled with the combined notional rather than with any single position.
 

Did the liquidity pool lose a quantified amount?

Public statements from Papertrade did not release a precise loss figure attributable to the reported activity. Pool balances observed around the period fluctuated between approximately $3.14 million and $5 million. Because the pool is the sole counterparty to every trade, any profitable close against a displaced midpoint reduces the capital available for other winners and for eventual staker distributions.
 

How do the asymmetric haircut and deadband affect realized profits?

A 0.2-basis-point deadband discards extremely small gains before profit calculation. An asymmetric-impact curve then applies its largest percentage reduction to the smallest profitable moves. A 2 percent win fee is charged after the haircut. These filters lower the net amount extracted from marginal displacements yet do not prevent larger, carefully timed moves from remaining profitable when position size is still substantial.
 

What operator controls remain available after the cap tightening?

A market keeper can reduce a long or short cap to zero without freezing the live midpoint used for closes. A pause guardian can retire a market identifier, freeze its midpoint, and still allow existing positions to close at the frozen price. A global gate can halt all new opens while closes continue against live prices. These controls complement the revised directional headroom.
 
Disclaimer: This content is for informational purposes only and does not constitute investment advice. Cryptocurrency investments carry risk. Please do your own research (DYOR).