Bearish sentiment around HYPE has risen sharply in recent weeks, and the market conversation has centered on three issues: the supply overhang from ongoing team unlocks, direct selling pressure tied to institutional unstaking and ETF outflows, and Hyperliquid’s growing dependence on a single deployer.
On the surface, those concerns look separate. The ChainCatcher article argues they point to the same root cause. Team tokens are unlocking, which increases the need for support through assistance fund buybacks. Those buybacks are funded by protocol fees. A large share of those fees comes from HIP-3 trading activity. And for now, that HIP-3 activity is concentrated almost entirely in trade.xyz.
That leaves a meaningful part of HYPE’s value support resting on the continued operation of one deployer.
The market has also been discussing reports that trade.xyz is seeking funding at a valuation of about $1.5 billion. That has not been officially confirmed, but the signal alone has pushed investors to revisit the longer-term valuation logic for HYPE. The core questions are whether HIP-3 has already moved into a structural bottleneck, and what such a concentrated setup could mean for HIP-4 and HYPE over time.
What happened to HYPE
Price is the most visible starting point. HYPE peaked around mid-June at nearly $77, then moved steadily lower in a choppy decline. It now sits around $56, down roughly 25% from the top.
Signs of selling pressure showed up earlier. On June 4, on-chain monitoring indicated that Arthur Hayes sold about 247,000 HYPE for roughly $18.02 million, nearly wiping out his position. That came only days after he had publicly said he was betting HYPE would outperform the top 10 crypto assets by market capitalization by year-end.
Institutional activity picked up in July. Multicoin Capital unstaked about 1.97 million HYPE near the end of the month, worth about $108 million, and also moved tokens to several exchanges over the same period. Selini and Bitwise were also seen sending tokens to exchanges.
Selini founder Jordi Alexander said those transfers were not sales. He said several parts of his firm’s operating stack use HYPE, including fee staking, wallet functions, market making and arbitrage on HyperEVM, and HYPE auctions. He also said he does not believe Paradigm and Multicoin are selling their positions.
Still, fund flows have changed direction. According to SoSoValue, the HYPE spot ETF shifted from consistent net inflows since launch to three straight weeks of net outflows in July, making it the only crypto ETF product to post net outflows that month.
Short positioning has built up as well. By early August, trader Loracle had raised combined short positions in HYPE and ETH to more than $46 million, with the HYPE short opened around $52.7.
During this adjustment, sentiment around HYPE has diverged noticeably from the broader market. That has sharpened the question: is this a normal cyclical pullback, or is the structure behind HYPE starting to show strain?
Concentration in HIP-3 is no longer hypothetical
DefiLlama data cited in the article shows Hyperliquid generated about $200.7 billion in perpetuals volume over the past 30 days, with about $10.7 billion in open interest and annualized protocol fees of roughly $1.82 billion. That still places it among the largest on-chain perpetuals venues.
A sizable portion of that activity now comes from HIP-3. A Dune dashboard shows cumulative HIP-3 volume has topped $480 billion since launch in October 2025.
According to a Q2 report from Hyperliquid Research Collective, HIP-3’s share of platform volume rose from 1.8% last year to 20.7% in the first quarter of this year, then to 32.2% in the second quarter.
Data from hl.eco shows the figure has moved even higher recently, reaching about 64.6% on a seven-day smoothed basis. In other words, close to half or more of Hyperliquid trading activity is no longer coming from native markets. It is coming from the open deployment framework.
The concentration looks even more extreme inside HIP-3 itself. As of August 2026, TradeXYZ had deployed 103 markets, with 88 of them actively traded. Those markets span commodities, foreign exchange, U.S. and Asian equities and indices, and pre-IPO products, including Cerebras (CBRS), SpaceX (SPCX), and CXMT.
TradeXYZ is posting a 30-day average daily volume of $3.7 billion, cumulative volume above $440 billion, and open interest of $3.5 billion. Since July 17, its seven-day volume has exceeded Hyperliquid’s native crypto perpetuals volume.
The Q2 report shows trade.xyz’s share of HIP-3 volume climbed from 85% in March to 97% in June and approached 100% in July. On a cumulative basis, trade.xyz has accounted for about 93% of HIP-3 volume. The second-place deployer, dreamcash, stands at just 4.2%, while all other deployers combined, including Kinetiq, Felix, and Paragon, account for less than 3%.
An Arrakis on-chain study from the first quarter found that about 63% of trade.xyz volume came from roughly 360 market-making wallets, including professional firms such as Jump Crypto, Selini Capital, and Wintermute. By wallet count, about 43% of addresses came from a sybil farm built by a single Polymarket operator, but those addresses contributed less than 1% of real trading volume.
Others are leaving, and newcomers are struggling to matter
The article argues this market structure took shape while some early participants exited and new ones found it hard to gain traction.
In June, early deployer Felix formally shut down its HIP-3 perpetuals markets. Co-founder Charlie said in a postmortem that Felix had generated about $3 billion in volume with first-mover products tied to crude oil, gold, and silver, but was eventually overtaken after trade.xyz launched the same markets quoted in USDC. The firm chose to exit.
The same report says Ventuals and Felix both left on the USDH settlement date in mid-June, Dreamcash stalled in early July, and Kinetiq chose to migrate, launching its first USDC pairs on July 1.
There are still attempts to take share. Paragon, for example, is said to have won several tickers since mid-July, spending about 6,328 HYPE. It focused on strong narrative assets such as AI supply chain names, humanoid robotics, and Reddit-related products, while deliberately avoiding the large-cap mainstream equity markets where trade.xyz is strongest. Its weekly volume rose more than tenfold in a short period, but its cumulative share remains negligible.
The structure of HIP-3 points to winner-take-most outcomes
HIP-3 was designed as a permissionless framework. Anyone willing to stake enough HYPE can deploy a perpetual market. In practice, the article says, the mechanism has produced something much closer to winner-take-most.
The first reason is the entry threshold. Launching a HIP-3 market requires staking 500,000 HYPE. At recent prices, that works out to roughly $20 million to $30 million. That excludes most teams from the start and leaves only a small set of well-capitalized players.
The second reason is the interaction between auctions and first-mover advantage. Each deployer gets the first three markets for free. After that, additional listings have to be won in a shared Dutch auction that starts at 500 HYPE, and the auctioned tokens are burned. Late entrants face higher listing costs while also competing against liquidity that has already been pulled into existing markets. Once an early deployer establishes depth and user mindshare, cold-start conditions become much harder for everyone else.
The third issue is payback economics. Based on estimates from Blockworks Research analyst Shaunda Devens, most HIP-3 deployers other than trade.xyz are earning annualized returns on staked HYPE that are close to, or even below, 1%. Across 136 paid-listed markets in the sample, only 44 recovered their auction costs. For non-trade.xyz markets, the median payback period was four years.
The same pattern may be carrying into HIP-4
The article says this logic is now showing up in HIP-4 as well.
On July 20, Hyperliquid said HIP-4 would allow permissionless deployment. HIP-4 is the platform’s move into prediction markets, or binary outcome markets. The path is more constrained than HIP-3: templates must be approved through validator voting, and validator-deployed markets are capped at fewer than 10 per year. But the stake requirement did not change. It is still 500,000 HYPE.
Arrakis, tracking the first two weeks after launch, found that one frontend, Outcome.xyz, was routing more than 10 times the volume of the second-largest frontend.
Algorithmic wallets represented only 6% of wallet count but generated nearly half of HIP-4 trading volume. Retail wallets made up the majority of addresses and also recorded the highest peak open interest, yet they contributed less than one-third of HIP-4 volume.
Over the same period, HIP-4 matched Polymarket in BTC binary options volume, but the gap between HIP-4 pricing and Deribit implied probabilities was four to five times larger than the gap seen on Polymarket or Kalshi.
Thin liquidity has also been visible in execution quality. Arrakis measured the maximum trade size that markets could absorb within ±2% slippage. It found that after launch-day excitement initially pulled in liquidity, execution conditions deteriorated as depth remained insufficient through most trading periods. In the worst case described in the article, a $1,000 trade would expose a trader to 2% slippage.
Arrakis said that is a bottleneck for users and an opportunity for market makers. In a market with little trading activity, the first market maker willing to commit meaningful capital is likely to face very little competition and can profit from the inefficiencies left by weak order-book depth.
From this perspective, HIP-4 is showing early signs of the same pattern: thin liquidity and high concentration. The article frames that not as an accident, but as a default outcome of the mechanism, one that naturally favors a small number of players with capital, resources, and an early lead.
What that means for HYPE
If the mechanism issue is already a settled fact, the next question is how long HYPE’s value support can hold under this market structure.
In the short term, the article argues, a large part of HYPE’s price floor is still being supported by trade.xyz. HIP-3 contributes a substantial share of platform trading activity, and protocol fee income feeds into the assistance fund used for buybacks.
Data cited from on-chain analyst MLM shows that since team tokens began unlocking in December 2025, about 4.93 million HYPE has entered team wallets. Of that, around 4.33 million was sold or transferred over the counter, generating about $165 million. Over the same period, the assistance fund bought back about 9.8 million HYPE for around $364 million, more than twice the pace of team sales.
On that reading, unlocks themselves have not yet created the main source of selling pressure. The article says the larger issue weighing on price is the change in demand. Recent data also shows Hyperliquid has cumulatively burned 47.53 million HYPE worth about $2.68 billion, equal to 4.75% of the 1 billion maximum supply.
Medium-term variables are more complicated. Reports that trade.xyz may be seeking funding at a valuation of about $1.5 billion remain unconfirmed, but the market is treating the signal seriously because it could weaken the tight alignment between protocol and deployer. If an independent valuation is established, trade.xyz could gain more bargaining power in future negotiations over revenue sharing, terms, and strategic direction.
That said, the article does not present a split as the base case. Blockworks analysts cited there argue that fears of trade.xyz leaving because revenue sharing is too expensive are among the weakest concerns on the table. The two sides are deeply tied on reputation, economics, and architecture, and neither has a clear reason to walk away.
If trade.xyz were to leave, it would have to rebuild the exchange layer on its own, which the article describes as the hardest part of the stack, while also giving up nearly its entire trader base. If Hyperliquid instead tried to internalize RWA markets, it would send a message to every future deployer that success only increases the odds of replacement. That would be a major reputational hit as well.
The article also notes that a 50% revenue split is not Hyperliquid’s only monetization channel. The platform can earn from write priority fees and read fees paid by market makers. An even larger source, it says, comes from second-order effects: traders bring in USDC to go long RWA products, increasing on-chain balances. Hyperliquid keeps 90% of that revenue stream, which is estimated at about $30 million per month, already more than the entire HIP-3 perpetual fee pool that it splits with trade.xyz.
Another variable is HIP-4. Because the prediction-market framework still uses a high staking threshold, the article says it will most likely end up dominated by a small number of players again. If that happens, the open-access narrative becomes harder to defend, and HYPE’s long-term platform-token premium may come under pressure.
Valuation remains contested
The article’s conclusion is that the growth brought in by trade.xyz is real. For Hyperliquid, the key test is whether that growth continues to settle at the protocol layer. If trade.xyz’s expansion still translates into value capture for HYPE, then concentration may not automatically be a negative.
On valuation, a recent Grayscale report said HYPE looks cheaper than traditional fintech companies on a cash-flow basis. Using an assumption of about $1 billion in protocol revenue in 2027, roughly 20% above 2025, the report estimates earnings per share of around $3.25 to $3.75. At current prices, that implies a forward price-to-earnings multiple of about 15x to 18x.