How far can an exchange go without traditional market makers or the need to inject liquidity upfront, while still allowing traders to use up to 1,000x leverage? Just yesterday, Papertrade launched on HyperEVM. Thanks to its unique loss-mining mechanism, it generated hundreds of billions of dollars in notional trading volume in a short time, quickly attracting a swarm of high-frequency arbitrageurs. From the two founders’ financial experiment to the PAPER issuance curve and dynamic solvency model, this seemingly ingenious mechanism is being put to the test in real markets. Who bears the risk, and who can profit over the long term?
Do you remember how you felt the first time you were liquidated in a contract trade?
Perhaps you simply opened a 50x leveraged long position before bed, thinking you’d wake up to a handsome profit. But while you slept, the market reversed, and every last bit of margin in your account was liquidated. You stare at the glaring numbers on your screen.
You start to feel frustrated and regretful, even compulsively checking the candlestick chart again and again, imagining how different things might have turned out if you had used less leverage or set a stop-loss just a few minutes earlier.
In the past year alone, BTC contract liquidations may have reached tens of billions of dollars. According to verifiable historical Coinglass data, from July 31 to September 9, 2026, cumulative BTCUSDT long and short liquidations on Binance, Bybit, and OKX totaled approximately $1.973 billion.
An average of approximately $48.13 million per day, or about $17.57 billion in nominal liquidations over the year.
But you also know that there are no do-overs in financial markets. The money you lost won’t return to your account just because you regret it, and the liquidation record won’t be erased just because you say, “If only I’d known.” But what if someone told you that there really was a do-over for contract trading—one that might even compensate you more the more you lost? Would you be willing to try again?
More intriguing still, this do-over doesn’t require you to apply to the exchange for compensation or wait for the market to rise again. It comes from a new mechanism that turns trading losses directly into token rewards.
A decentralized exchange called Papertrade has brought this seemingly absurd idea to the Hyperliquid ecosystem. It lets traders use up to 1,000x leverage and introduces a loss-mining model called Loss-to-Mine. Whenever an eligible trading loss occurs, participants may receive PAPER tokens and, in turn, share in the protocol’s future earnings.
Of course, this do-over can’t guarantee that anyone will recover their losses, and future earnings are highly uncertain. But when real losses become an option for acquiring another kind of financial claim, an experiment in leverage, speculation, and human nature begins.
In traditional financial markets, the business model of an exchange is clear and easy to understand. Stock exchanges charge transaction fees for matching trades; futures exchanges earn revenue from trading, clearing, and related services; and market makers profit from bid-ask spreads and risk management.
Regardless of whether traders ultimately make or lose money, exchanges generally want markets to remain active, because trading volume means business revenue.
Papertrade, however, proposes a controversial design: traders’ losses themselves can serve as the foundation for bootstrapping protocol liquidity, issuing tokens, and distributing revenue to stakers.
Papertrade officially opened for trading on October 10. During the pre-deposit phase, around 11,465 addresses deposited a total of $138 million. After trading opened, the platform’s notional trading volume quickly exceeded $100 billion and then climbed further to several hundred billion dollars. Compared with this astonishing trading volume, the amount of capital actually committed to the system was much smaller.
Leverage of 1,000x theoretically means that one dollar of margin can control a $1,000 notional position. As a result, hundreds of billions of dollars in trading volume can be generated by repeatedly opening and closing positions with a relatively limited amount of capital. These figures are enough to show how active the market was.
So you might be wondering: what made so many traders willing to enter a high-leverage market where they could lose money rapidly?
First, we need to understand who the two people behind Papertrade are.
To answer that, we first need to meet Papertrade’s two founders, Jez and blurr.
Jez’s real name is Colin Hong, and he uses @izebel_eth on X. He graduated from the Massachusetts Institute of Technology in 2015 with a degree related to computer science, engineering, and management science. He later worked at Morgan Stanley for about four years before moving into crypto investing. Public biographical information from the World Poker Tour also lists him as a Venture Partner at Standard Crypto. Compared with many DeFi founders who started out in computer science, Jez’s background spans traditional finance, crypto trading, and competitive poker—experiences that clearly connect with Papertrade’s product philosophy.
The other founder, blurr, has the unmistakable profile of an independent developer.
In an interview lasting around 95 minutes, hosted by Thread Guy at the end of September, blurr revealed that he had almost no conventional work experience and had never adapted to a typical office job. As a young man, he spent much of every day playing RuneScape. Around 2017, he entered the crypto market and gradually shifted his interest in game mechanics, economic incentives, and programming toward on-chain finance.
CryptoKitties was one of his early formative experiences. He studied the dominant and recessive gene mechanics of the digital cats, using his understanding of on-chain data and breeding rules to find economic opportunities other participants had not yet discovered. Later, he also took part in the ecosystems of early protocols such as Compound, Uniswap, and Curve, with a particular knack for understanding the economic incentives and market behavior embedded in complex rules.
This background gave him an almost game designer-like attitude toward smart contracts. To blurr, a blockchain protocol is an economic system that can be observed, analyzed, and interacted with. Smart contracts set the rules, participants seek optimal strategies within those rules, and the true quality of a product ultimately depends on how the system evolves as those strategies compete.
Interestingly, this developer, who had long preferred to work alone, found a rare collaborator in Jez.
According to blurr, the two met in New York in the summer of 2021, initially through NFTs. Over time, they developed a steady working relationship. In an interview, blurr explained that he had disliked working with others in the past largely because explaining problems, dividing up work, and managing progress often took more time than doing the work alone.
In an interview, blurr revealed that the two may have written research papers on mechanisms that require no initial bookmaker capital as early as the summer of 2024, and tried applying them to games such as blackjack and coin flips, whose outcomes are relatively easy to model. These ideas later evolved into the Martingaler Protocol, the underlying economic framework currently used by Papertrade.
Traditional casinos need to have a certain amount of capital on hand to accept bets. Even if a casino has a mathematical advantage in the long run, a streak of big wins in the short term can leave it short of cash reserves. Martingaler’s idea is to handle temporarily unpayable winnings through a debt queue, replenish liquidity through future trading losses, and issue tokens that share protocol revenue to incentivize participants to keep contributing funds to the system.
This design makes it possible for a bookmaker model that would otherwise be difficult to launch to build a pool from zero. In blurr’s view, a large share of trading profits in crypto ultimately flows to market makers and exchanges. Papertrade aims to redistribute some of those economic benefits to participants, allowing people who have lost money trading to share in the trading system’s future earnings.
This took Papertrade in a highly controversial direction: launching so-called Loss-to-Mine, or loss mining.
Under the protocol’s design, traders can mint PAPER based on the amount of eligible realized losses when they close a position or are liquidated. When the liquidity pool tracked by the protocol is below $2 million, the initial minting rate is 100 PAPER for every $1 lost. As the pool continues to grow, the token issuance rate gradually declines along a preset curve.
For a simple example, a trader puts in $100, uses high leverage to trade ETH, and ultimately loses the entire margin. If the rate is still 100 PAPER per dollar lost and the loss meets the minting criteria, that loss could theoretically correspond to 10,000 PAPER.
How much are those 10,000 PAPER actually worth?
Under Papertrade’s initial mechanism, PAPER transfers are restricted, and holders primarily participate in the protocol by staking and unstaking. There is no fixed price that directly guarantees redemption value, nor any arrangement that guarantees you can recover one dollar in rewards from the protocol for every dollar you lose.
The economic value represented by PAPER comes primarily from the future distribution of protocol revenue. In other words, traders who lose real USDC today receive token rights that may generate an ongoing stream of cash flow in the future.
Papertrade has publicly disclosed PAPER’s basic issuance parameters, with its token supply starting at zero. In the initial phase, when the LP size tracked by the protocol is below $2 million, each dollar of eligible losses earns 100 PAPER. As the system reaches the corresponding thresholds, the number of newly issued tokens gradually declines.
The following results can be derived from the decay equation:
Assume that the initial $2 million pool is built entirely from eligible losses, and that the rate of 100 tokens per dollar always applies. In theory, this phase would generate around 200 million PAPER. If we also account for the possibility that the pool could fall back to a low level and re-enter a high issuance-rate range, the final supply could exceed what a single decay curve would suggest.
From an economic perspective, Papertrade has designed a token issuance mechanism tied to the health of its pool. The less money the protocol has, the greater the token incentive offered to new participants who incur losses. As the pool grows, new participants receive fewer tokens for the same amount of losses.
This issuance model naturally favors early participants. They can acquire more PAPER when the token supply is smaller and the number of tokens minted per unit of loss is higher, while later participants must bear a greater economic cost to obtain the same share of the supply.
The problem is that token scarcity and token economic value are still two independent variables. If protocol revenue falls significantly in the future, even a sustained decline in the issuance rate may not allow early holders to earn enough cash flow through staking to cover their original losses.
For this model to operate sustainably, we also need to understand who its counterparties are.
In traditional order-book trading environments such as Hyperliquid, the market typically relies on different traders and market makers to provide bids and asks. Papertrade uses a different design: traders’ contract positions are settled directly against the protocol’s internal USDC liquidity pool.
The protocol reads Hyperliquid’s best bid and ask prices through HyperEVM and uses their midpoint as the reference price. When a trader opens a position, the smart contract records the corresponding price; when the position is closed, profit or loss is calculated using the new reference price.
As a result, Papertrade itself does not need to find another trader willing to take the opposite side of every trade. Its liquidity pool acts as the common counterparty to all traders.
More unusually, this liquidity pool can be built from zero. The protocol’s design does not require traditional LPs to provide substantial market-making capital upfront, nor does it offer a direct liquidity deposit mechanism for ordinary investors. Early traders’ realized losses add to the pool’s balance, gradually enabling the protocol to pay out to profitable traders.
Suppose the platform initially has 100 traders, each putting in $100. If 90 of them eventually lose money, contributing a total of $9,000 to the liquidity pool, those funds become an important source for the protocol to pay out to other profitable traders. This example is only an illustration of the mechanism; actual settlements are also affected by factors such as profit deductions, fee distribution, and unsettled liabilities.
But this raises a new question: what happens if the profits owed to traders exceed what the liquidity pool can pay?
Papertrade’s solution is to introduce a settlement queue. When the liquidity pool is temporarily unable to cover all profits, unpaid earnings enter a waiting queue. As funds from subsequent losses flow in, the protocol gradually pays off the accumulated obligations according to its rules.
This effectively allows the protocol to incur a temporary shortfall in its ability to pay under certain conditions. It can reduce the need to halt trading outright when liquidity is insufficient, but it transfers some of the risk of cash-flow constraints to profitable traders.
For traders, there may be a delay between the profits shown in their accounts and the USDC they can withdraw immediately. For the protocol, its ability to maintain orderly settlements over the long term will depend heavily on the dynamic relationship between new losses, payout demand, liquidity reserves, and risk controls.
That is why Papertrade’s real test goes far beyond matching efficiency. It must prove that the protocol can continue to meet its settlement obligations over the long term, even as market participants constantly change.
This payment queue is, in essence, a dynamic balance sheet.
In an interview, blurr gave a straightforward example. Suppose the protocol has $5 million in liquidity and a trade generates $6 million in profits owed. The system can pay out $5 million upfront, record the remaining $1 million as an outstanding debt, and then use losses from future trades to replenish the funds.
From a systems-design perspective, this approach provides a buffer for a trading protocol that temporarily lacks cash. It preserves the possibility of making payments in the future, allowing the protocol to launch with zero initial liquidity. However, if debt accumulates faster than funds are replenished over an extended period, the waiting queue will continue to grow.
For the two founders, this experiment remains full of unknowns.
The market’s response is far more complex than mathematical simulations suggest. The exceptionally high trading volume at launch proved that speculative demand exists, while the unusual trades that followed reminded them that once economic rules are made public, every design becomes a target for participants seeking an edge.
For more than a decade, the crypto industry has repeatedly tried to combine trading, gaming, and financial incentives. From trade mining to liquidity mining, from casino tokens to today’s loss mining, the market has continually sought a mechanism that encourages participants to keep contributing funds while sharing in the economic benefits generated by the protocol.
Papertrade offers DeFi a new case study worth examining, and puts an age-old financial question back before all participants.
When a financial institution uses future traders’ losses to back today’s payment commitments, its fate ultimately depends on whether it can generate sufficient, sustainable, and realizable cash flow in the future.
Original: https://x.com/0xmediaco/status/2109645362730573824