Papertrade, a synthetic perpetuals venue built on HyperEVM, booked roughly $56 million in revenue during its launch week of October 10–17 — 3.2 times the $17.5 million Hyperliquid recorded over the same period — and paid out more than $50 million of that to stakers of its PAPER token. Research analysts Carlos Guzman and Slater Santer at trading firm GSR reported the figures in the firm's October 19 weekly, which also put Papertrade's launch-day take at $25 million.

The mechanism behind the numbers is unusual: the platform mints PAPER only when a trader loses money. Launch-week participants responded by deliberately taking losses, racing through liquidations or running delta-neutral positions to maximize tokens earned per dollar lost. The mint rate fell from 100 PAPER per dollar of qualifying loss to 11 within nine days, and a revised set of mint rules took effect on October 21. The setup inverts the usual logic of trading, in which the objective is to avoid losses, and offers a concrete example of how directly token emissions can steer on-chain behavior.

Every PAPER Token Began as a Losing Trade

Papertrade runs on HyperEVM, the smart-contract layer of Hyperliquid, a decentralized perpetuals exchange, and offers up to 1,000x leverage on BTC and ETH. Traders never face each other. Each position is a bet against a protocol-owned pool that, according to Papertrade's pool documentation, started at $0 with no seed capital and grows from user losses.

PAPER is minted only when a trader closes a position at a loss or is liquidated. The project's token overview calls it a "consolation claim on future protocol value." Because the token cannot yet be transferred between wallets, every PAPER in circulation was minted on a losing trade. Staked PAPER earns a pro-rata share of the protocol's USDC distributions, a revenue-sharing structure common across DeFi token designs.

Under the launch contracts, each dollar of qualifying loss minted 100 PAPER while the pool held less than $2 million, with fewer tokens issued as the pool's cumulative gains grew. GSR tracked the rate falling to 31 PAPER per dollar by October 12 and to 11 by October 19.

GSR identified two ways traders farmed the token. Some raced through liquidations to stack PAPER as quickly as possible. Others ran delta-neutral setups, pairing positions so their net market exposure stayed near zero while maximizing PAPER per dollar lost. Leverage set the pace: at 1,000x, $100 of margin controls a $100,000 position that a move of roughly 0.05% wipes out. GSR counted more than $678 billion of launch-day notional volume against about $137 million of deposits.

Following One Dollar From a Losing Trade to a Staker

The path from a loss to a payout runs through four steps, set out in Papertrade's token documentation and in DeFiLlama's published Papertrade adapter, which defines how the revenue figure is counted. GSR lists DeFiLlama among its data sources.

First, the loss lands in the pool. A losing close or liquidation moves the trader's lost margin into the pool and mints PAPER on that loss.

Second, a 2% fee is split. Closes pay 2% of realized profit loss, while liquidations pay 2% of the lost margin. Half goes to stakers; the other half goes to a developer fee recipient set by the contract owner.

Third, the pool keeps a $5 million reserve. Gains up to that cap stay in the pool to pay future winners.

Fourth, the surplus goes to stakers. Pool gains above the cap are swept to staked PAPER in USDC.

GSR's numbers show how much of the headline revenue traveled that path. The pool took in roughly $56 million in launch week and, net of close and liquidation fees, paid more than $50 million to stakers — at least 89 cents of every revenue dollar. For traders as a group, the week came close to a round trip: about $56 million in, more than $50 million out to holders of loss-minted tokens.

The developer's cut is half of the 2% fees, which DeFiLlama books as a separate revenue line. GSR counted $63.4 million of liquidated collateral through October 18, meaning the developer's half of liquidation fees comes to no more than about $634,000. GSR does not break out closing fees, which leaves the developer's full take for the week unquantified.

Hyperliquid's Revenue Comes From Fees, Not Losses

The two venues collect money in fundamentally different ways:

• Hyperliquid — Charged on: position size, via a 0.045% base taker fee on perps. Paid by: every taker, win or lose. Goes to: HLP vault, assistance fund, market deployers. Oct 10–17 revenue: $17.5 million.
• Papertrade — Charged on: realized losses, plus 2% of profit or loss. Paid by: losing traders, plus 2% of winners' gains. Goes to: PAPER stakers, minus the developer's half of fees. Oct 10–17 revenue: $56 million.

Hyperliquid's fee schedule states that its fees are "entirely directed to the community," with the assistance fund converting its share into HYPE that is then burned. Neither venue sends its fees to a company treasury.

The difference lies in what the trader pays for. A Hyperliquid taker pays a slice of position size in exchange for execution, while most of Papertrade's intake is money its traders lose. That distinction also prevents the two venues' volume figures from being compared directly. At Hyperliquid's base taker rate, Papertrade's $678 billion launch-day notional would have generated about $305 million in fees. Papertrade booked $25 million that day, because its income tracks profit and loss rather than position size.

Inside the Round Trip, Money Moved Two Ways

The group-level picture hides two transfers. The first runs from losers to winners. The pool's intake is net of winning trades it had already paid, so traders lost more than $56 million in gross terms.

The second runs from late minters to early ones. Papertrade's staking rules split each distribution in proportion to staked balances at that moment, with no lockup or cooldown. A loss taken on launch day therefore earned roughly nine times the payout share of the same loss taken on October 19.

With no open market for PAPER, the USDC it collects is the only return that can be measured. A farmer's result is that USDC minus the losses taken, adjusted for any gains on the winning leg of a hedge. No protocol-wide total can show that figure for an individual account.

The New Rules Pay Extra to Refill an Empty Pool

When the pool cannot cover a winning close in full, the unpaid profit enters a first-in, first-out queue and is cleared as later losses refill the pool. The trader's original margin comes back immediately, unless the position was opened with an existing queued balance.

Papertrade's second contract set went live on October 21, after GSR's measurement window. Under the revised mint curve, each loss mints a fixed share of the current PAPER supply per dollar, with no cap. While the pool sits below zero, the part of a loss that refills the deficit counts up to five times toward the mint — a multiplier that ramps up over three days. The pool documentation names an underwater pool as the point where PAPER mints fastest, so a shortfall owed to winners is met with a larger mint for losers.

The same upgrade slightly holds back winning closes on small moves and lets a keeper switch winners to a separate off-hours payout curve. Because each dollar of loss now mints a set fraction of existing supply, every new loss dilutes current stakers' share of future distributions. As of October 22, the mint-curve page had not published the live value of the constant that sets that fraction.

What Would Show the Loop Can Run Without Launch Incentives

Supporters of the design have a straightforward argument. PAPER launched with zero supply and no team, investor, or airdrop allocation, the pool has no outside depositors, and the developer takes half of a 2% fee. By that standard, launch week delivered a revenue share to the people who funded it.

The limitation is that every payout needs fresh losses, while the token reward for supplying them shrank sharply within nine days of launch. The October 21 contracts change how that reward is calculated, not the dependence on new losses.

The first full week under the new contracts, October 21 to 28, offers a direct test. If staker distributions stay near launch-week levels while the pool remains above zero, losses are arriving without help from the bonus. If the pool slips below zero and the five-times multiplier starts carrying the refill, Papertrade is paying extra to attract the losses that cover its winners.

This article is provided for informational purposes only and does not constitute financial or investment advice. Leveraged trading and token incentives can result in substantial losses.

This article originally appeared on Coindoo.