The Only Two Profitable DATs: A Structural Anomaly or a Mirage?
Hook
CoinTelegraph reported yesterday that Hyperion and Hyperliquid are the only two digital asset trading platforms (DATs) currently showing positive unrealized PnL. In a sector where most platforms operate at a structural loss—subsidizing user incentives with token dilution or relying on volatile fee income—this data point stands out. The immediate reaction? Optimism. A rare green flag in a sea of red. But before we celebrate, we must ask: what exactly is being measured, and more importantly, what does it hide?
Context
DeFi derivatives have been the battleground for capturing CeFi market share. Platforms like dYdX, GMX, and SynFutures compete on liquidity depth, fee structures, and risk management. Yet most operate at negative unrealized PnL when accounting for their own market-making inventory or liquidity provider incentives. Why? Because the cost of attracting and retaining liquidity often exceeds the fees generated in a low-volatility environment. The only way to flip that equation is to either scale volume dramatically (which requires market share) or to take asymmetric risk positions that work in your favor. Hyperion and Hyperliquid appear to have cracked the code—or have they?
Core
The critical question revolves around the definition of “unrealized PnL.” A protocol’s unrealized PnL can originate from three distinct sources: (1) its own trading inventory (the protocol acting as a market maker on its own platform), (2) yield earned on deposited collateral that is not yet withdrawn, or (3) price appreciation of native tokens held in treasury. Each source carries a different risk profile. Source (1) is the most sustainable if the protocol possesses genuine alpha in market-making—like a perfected execution algorithm. Source (2) is largely passive and can vanish if the underlying lending market suffers a contraction. Source (3) is the most dangerous: it is circular, self-referential, and can lead to a death spiral if the native token price drops.
Based on my 2020 liquidity stress-testing experience at a quantitative fund, I built models that separated these revenue streams. The key metric is not just the sign of the PnL but its composition. If a protocol’s unrealized PnL is heavily weighted toward source (3), then the reported “profitability” is merely a reflection of its own token performance—a hallow victory. I have audited over 400 smart contracts during the ICO boom, and I can confirm that most projects that boasted of positive treasury PnL were masking capital inefficiency. We do not predict the wave; we engineer the hull.
Let’s examine the available numbers. Hyperliquid, for instance, has a native token HYPE that has appreciated 15% in the past month. If the protocol treasury holds a significant HYPE position, that appreciation alone could have turned the overall PnL positive—even if the core trading business is losing money. Hyperion is more opaque; little is known about its tokenomics. The reported data likely comes from on-chain aggregators like Token Terminal or Dune, but the methodology is not disclosed. The aggregate PnL may include unrealized gains from liquidity provision in external pools, which are notoriously volatile.
Contrarian Angle
Now, the contrarian viewpoint: positive unrealized PnL might actually be a warning signal, not a validation. In a sideways market, consistent positive PnL suggests that the protocol is taking on excessive directional risk—betting that a particular asset will rise. This is exactly the opposite of what a neutral market-making protocol should do. The role of a DAT is to provide liquidity and capture the bid-ask spread, not to speculate. If Hyperliquid and Hyperion are making money from directional bets, they are effectively acting as hedge funds, not exchanges. That exposes them to wipeout risk if the market turns. I recall the 2022 Terra collapse: many “profitable” protocols were actually making money from lending out UST at high yields, which turned out to be an unsustainable arbitrage.
Furthermore, the exclusivity of this “only two” narrative may itself be a trap. It creates a scarcity effect that drives capital toward these two platforms, potentially inflating their TVL and token prices beyond fundamental value. The risk is a double dip: first, when the inevitable correction hits and the unrealized gains become realized losses, and second, when the narrative breaks and capital flees. We do not predict the wave; we engineer the hull. This is a call to examine the engineering, not to ride the narrative.
Takeaway
So, what is the actionable takeaway for a market that is trending sideways? Do not treat positive unrealized PnL as a buy signal. Instead, ask for a breakdown: what is the source? What is the time-weighted average capital at risk? How does the PnL correlate with market direction? Only then can you assess whether these two DATs are outliers because of genuine structural efficiency or because of luck and hidden leverage. My recommendation: if you must position, wait for a quarterly report that separates realized vs. unrealized and breaks down the sources. The market often punishes those who chase headlines. We do not predict the wave; we engineer the hull. Let the data guide your next move, not the headline.
