The market is celebrating Hyperliquid’s $3.6 billion RWA open interest as a breakthrough. I see a different number: the 48% decline in buyback volume. That’s the real story.
Let me rewind. I’ve been auditing smart contracts since 2017 in Cape Town, tracing liquidity flows that most people ignore. During DeFi Summer, I watched protocols pump APYs that were nothing but fiat debasement arbitrage. Now, I’m watching Hyperliquid’s HIP-3—a mechanism that lets anyone deploy a perpetual market by staking 500,000 HYPE—and the same pattern is playing out. Hype is just liquidity with a distorted memory.
Context: What HIP-3 Actually Does
Hyperliquid is a Layer 1 blockchain that hosts a decentralized perpetual exchange. HIP-3, introduced in early 2026, allows external entities to build markets on top of the protocol. They get 50% of the trading fees from those markets. The other 50% goes to Hyperliquid’s protocol, which then uses 99% of its revenue to buy back HYPE tokens via an Assistance Fund.
Kain Warwick, founder of Synthetix and Infinex, recently called this unsustainable. He’s been through the same game. Synthetix caps external builder fees at 30%. Warwick argued that Hyperliquid will eventually cut the split, because the protocol holds all the power—it can reduce fees or even absorb the builder’s market at any time.
But here’s the kicker: the data supports his skepticism. Hyperliquid’s total revenue fell from $357 million in Q3 2025 to $202 million in Q2 2026, a 43% drop. Buybacks fell from $290 million to $149 million, a 48% decline. Meanwhile, trading volume on the platform barely budged. The fees are just flowing to different people—specifically, to a single builder called trade.xyz, which controls over 90% of HIP-3 open interest.
Core: The Mechanical Chain of Value Destruction
Let me link the dots. This is where my macro lens kicks in. Total trading fees remain high. But 50% of those fees are now siphoned off to external builders. Protocol retained revenue drops. Buybacks drop. The deflationary narrative for HYPE weakens. The token price drops.
That’s the chain. It’s not a theory. It’s happening. The RWA perpetual market grew from 2% of total volume to 50% in a single quarter. That’s impressive, but it’s also a structural shift that concentrates risk. Trade.xyz alone accounts for $3.6 billion of the $36 billion RWA OI. If that builder sneezes, Hyperliquid catches pneumonia.
I’ve seen this before. In 2022, I analyzed the Terra/Luna collapse by focusing on the fragile tether of algorithmic stablecoins to global dollar liquidity. The same principle applies here: the protocol’s value capture is becoming dependent on a single counterparty that operates on a power-based relationship, not a smart contract guarantee.
Warwick pointed out that builders have no real alternative. Hyperliquid is the “mothership.” But that dependence is mutual. The protocol is now hooked on the volume that trade.xyz brings. And the 50% split? It’s a subsidy. Distraction is the tax we pay for novelty.
Contrarian: The Decoupling Thesis Nobody Wants to Hear
Here’s the counter-intuitive angle: The 50% split might actually be sustainable if it drives exponential growth in volume. But the data shows the opposite. Volume is flat. Revenue is down. The split is just redistributing a fixed pie.
Most market participants are still buying HYPE based on the buyback narrative. They see the $3.6 billion OI and think “growth.” They don’t see that the buyback pool is shrinking. The price has already dropped from $76.67 to $57.66, a 25% decline. But that might not fully reflect the halving of buybacks.
If Hyperliquid cuts the split to 30%—matching Synthetix—protocol revenue could jump by 40% immediately. That would restore buybacks. But trade.xyz might leave. Or it might not, because it has $28 million in HYPE staked as collateral. That’s sunk cost.
Warwick is right: the protocol can dictate terms. But the market is pricing HYPE as if the current split is permanent. It’s not. The real question is not whether the split will change, but when and how. That’s the blind spot.
From a macro perspective, we’re in a bull market. Hype masks technical flaws. But the flaws are structural. The concentration of counterparty risk, the declining buyback efficiency, the regulatory exposure from RWA perps—these are not temporary. They are the new normal.
Takeaway: The Window Is Closing
Hyperliquid is at a fork. It can either adjust the split preemptively to restore value capture, or let the market force the adjustment through price discovery. The latter will be more painful.
I’ve been through enough cycles to know that narrative decays faster than code. The HYPE token is not broken. The mechanism is. Fix the split, diversify the builder base, and the re-rating will follow. Ignore it, and the $57.66 level becomes a ceiling, not a floor.
The next iteration of DeFi derivatives will not be built on 50% splits. It will be built on sustainable value capture. Hyperliquid has a window to adjust. If it doesn’t, the market will adjust for it. And I’ll be watching the on-chain data, not the headlines.
