Weekly Perp News ReviewWeeklyWeek of August 24-30, 202613 items~14 min read
Perpetual Futures Weekly News Review - Week of August 24-30, 2026
Thirteen items from inside the week of 24-30 August 2026. Every item is attributed to a named source with a date. Where sources disagreed, both figures are printed. Anything that broke after the window closed is held for next week's review.
The dominant story of the week was not a price move. Bitcoin closed the week at $78,543, down -0.5%, and ether at $2,466, down -0.6%. Both were unremarkable. What was remarkable is that leverage kept building while the cost of holding it stayed at its own average - and that the regulatory perimeter around on-chain perp venues moved further in seven days than it had in the previous seven months.
Perpetual market snapshot, week of 24-30 August 2026
Metric
Reading
Week on week
Perp DEX 30-day volume
$533.63bn
trailing 30d measure
Perp DEX open interest
$22.44bn
-12.74%
Bitcoin perp open interest
$20.77bn
z +2.49
Ether perp open interest
$12.19bn
z +2.01
Bitcoin funding, 8h settlement
0.00906%
9.92% annualised
OI-weighted funding, HL cohort
12.24% annualised
on $9.49bn notional
24h liquidations
$438m
$298m long / $140m short
Stablecoin collateral base
$289.6bn
flat on the week
1. Regulation moved from principle to structure
Two filings inside four days did most of the work. On 24 August the Hyperliquid Policy Center asked the SEC and CFTC to recognise qualifying cash-settled equity perpetuals as security futures. On 27 August five former agency principals told the same two agencies to stop layering duplicate requirements on identical risk. Neither filing is binding. Both matter because they frame the question regulators will actually have to answer: whether a perp is classified by what it references, or by how it is built and traded.
The number that gives the second letter its force is the estimate that offshore perpetual volume exceeded $90 trillion in 2025 against $28 trillion in 2023. That figure comes from Kalshi, a market participant, and is quoted in a comment letter - it is not an official regulator statistic, and it should be read that way. But the direction is not controversial, and it is why onshoring has become policy.
2. Leverage stayed on; carry stayed cheap
Six of the eight largest perps sat above a +2 open interest z-score at the weekly close. Bitcoin carried $20.77bn at a z-score of +2.49. Zcash was at the 99th percentile of its own history. Solana added 10.2% of open interest in a single week to reach $2.51bn - while spot fell 4.01%, meaning new positions were being opened against the move rather than old ones closed.
Funding did not follow. Bitcoin's annualised funding printed 4.95% against a z-score of +0.01, which is its own average to two decimal places. Ether printed 5.36% at +0.38. Only NEAR, at 24.31% annualised, sat beyond two standard deviations among perps with more than $150m of open interest. A market holding this much position at this price of carry is not charging for risk - which is exactly the setup that turns a routine move into a cascade if the level goes.
3. The on-chain category grew turnover and shrank commitment
Perp DEX venues cleared $533.63bn over the trailing 30 days while holding $22.44bn of open positions, and open interest fell -12.74% on the week. Hyperliquid alone reported $106m of August fees, up 23% month on month, on roughly $400.00bn of volume. Fee revenue grew while committed capital across the category contracted. That is not a contradiction, but it does change what the revenue is: turnover-driven fees are more sensitive to incentive programmes than fees earned on standing risk.
4. Alt positioning finally unwound
The alt leverage flagged in the prior week's data did come off. XRP shed 11.8% of open interest with price down 6.9%. Dogecoin shed 13.5% with price down 8.1%. TRUMP flipped from the board's crowded long to its crowded short, funding at -16.73% annualised. The unwind landed precisely where positioning had been most one-sided, which is what a positioning unwind is supposed to look like.
5. Volatility front collapsed, back end did not
Bitcoin's seven-day realised volatility fell from 60.99 to 27.46; ether's from 114.96 to 35.61. Thirty-day implied followed more slowly to 35.43 and 49.30, leaving variance risk premium at -13.13 and -29.75 respectively - both below the tenth percentile of their own years. The twenty-day windows, at the 81st and 90th percentiles, still carry the August rally. The front of the curve looks calm because the week was calm, not because the regime changed.
6. Collateral was not the constraint
Total stablecoin market capitalisation held at $289.6bn, with USDT at $183.3bn and USDC at $73.5bn - together about 89% of the total. The base was flat on the week. Since most perp margin is stablecoin-denominated, that flatness tells us the deleveraging visible in open interest was a positioning decision rather than a funding constraint.
What the desk would push back on
The most common reading of this week is that high open interest plus neutral funding is constructive. The desk's own reading is closer to neutral-with-a-tail. Leverage at a +2.5 z-score with carry at its own average means nobody is being paid to hold the crowded side, which removes the early-warning signal that expensive funding normally provides. The warning is not that a cascade is imminent; it is that the usual indicator would not give much notice.
The week, item by item
1. Hyperliquid Policy Center asks the SEC and CFTC to treat qualifying equity perpetuals as security futures Regulation
The policy centre argued that regulators should examine how a derivative is structured and traded before using its underlying asset to divide jurisdiction. Under the proposal, futures-like perpetuals referencing single stocks would fall under the security futures framework that the SEC and CFTC administer jointly. The submission put a number on the category it is asking regulators to accommodate: HIP-3 builder markets processed more than $480bn in cumulative notional volume in their first 10 months. This is the most concrete attempt yet to fit a permissionless perp venue inside an existing US product category rather than invent a new one.
The interesting part is not the ask, it is the landing spot. This does not petition for a new category, it argues for an existing one - security futures - which is a far easier thing for a regulator to accept than inventing a framework. The question to watch is not whether the classification argument lands, but whether it arrives with the same front-end conditions a designated contract market carries. That is where the actual barrier sits.
2. Five former SEC and CFTC officials tell the agencies to stop duplicating oversight of the same risk Regulation
Responding to the joint SEC-CFTC request for comment on the definition of swap and security-based swap, the group advanced one principle: similar risks should receive comparable treatment, and overlapping requirements aimed at the same risk should be coordinated rather than duplicated. They were explicit that this is not a deregulation argument - the letter states that neither more nor less regulation is an end in itself, and that the target should be regulation proportionate to risk. For perp markets the practical question is whether a single contract ends up answering to two agencies with two rulebooks.
Former principals carry weight precisely because they know how the two rulebooks collide. But note the standing: several now advise or sit with crypto firms. The argument holds on its own terms - duplicate oversight of identical risk is inefficient - yet this is not a disinterested source, and it should be read with that in mind.
3. Offshore perpetual volume put above $90 trillion for 2025, against roughly $28 trillion in 2023 Market structure
The letter uses the figure to argue that a very large market has already formed outside the US regulatory perimeter: $90 trillion in 2025 against $28 trillion in 2023, more than a tripling in two years. It closes with the line that regulation can relocate activity without eliminating either the demand or the risk. Treat the number as an estimate rather than an official statistic - it comes from a market participant, not a regulator - but the direction of travel is not in dispute and it explains why onshoring has become a policy priority.
Worth repeating that this figure is a market participant's estimate quoted inside a comment letter, not a regulator statistic. What survives that caveat is the order of magnitude and the direction: offshore perp volume is now large enough that ignoring it is not a policy option, which is the reason onshoring became a live question at all. Do not use it as a precise number.
4. HYPE prints an all-time high near $86.71 before settling back Hyperliquid
The token reached roughly $86.71 on 27 August and traded back near $84 into month-end, still up more than 60% since the start of August on the figures cited. The move is a clean illustration of how perp venues now trade as listed proxies: the token responded to regulatory and commercial news about the venue, not to anything that changed in the order book that day. Anyone using HYPE perp open interest as a positioning signal should note that the underlying is itself a claim on venue economics.
Token price and venue fundamentals pulled apart this week - fees set a monthly high while open interest across the category contracted. Anyone reading HYPE as a proxy for perp throughput got contradictory signals. We would rather read the relationship between fee revenue and committed capital than the token.
5. Hyperliquid records $106m of fees in August, up 23% month on month Hyperliquid
August fees of $106m, on roughly $400.00bn of perpetual volume, imply an annualised revenue run rate near $1.251bn and cumulative protocol revenue of $512.08m. The monthly gain of 23% came while the broader perp DEX category was shedding open interest, which is the more interesting part: revenue grew even as committed capital across the category fell -12.74% on the week. Fee growth driven by turnover rather than by standing risk is a different quality of revenue, and it is more exposed to incentive cycles.
Fees rising while open interest falls means the revenue is increasingly turnover-driven rather than earned on standing risk. Turnover-driven fees are more sensitive to incentive programmes and less stable through a quiet month. The test is the next month: if volume slips, does the fee line follow, or does it hold on the strength of the book?
6. AQA v2 directs roughly 90% of USDC reserve yield to HYPE buybacks Hyperliquid
The upgrade routes about 90% of cost-adjusted reserve yield from USDC collateral into HYPE buybacks. Structurally this converts a portion of the margin stablecoin balance into protocol-level demand for the token. It also tightens the link between two things this desk tracks separately: how much stablecoin collateral sits on the venue, and what the venue's token does. Anyone modelling one should model the other.
This ties fee revenue more tightly to token value and adds an interest-rate dependency to the revenue mix. Reserve yield is not trading income; it moves with the rate environment rather than with usage. Credit where due for writing the rule down explicitly, but anyone modelling the revenue should separate what the venue earns from trading against what it earns on the float.
7. Bitcoin perp open interest sits at a +2.49 z-score while funding prints its own average Open interest
Bitcoin perp open interest of $20.77bn carried a z-score of +2.49, and six of the eight largest perps sat above +2. Ethereum registered $12.19bn at +2.01, Solana $2.51bn at +2.33, Hyperliquid $2.70bn at +2.14, Zcash $1.15bn at +2.52 - the last at the 99th percentile of its own history. Yet bitcoin's annualised funding printed 4.95% with a z-score of +0.01, effectively its own average to two decimal places. Leverage is elevated; the price of carrying it is not.
This is the week's most important divergence. Positioning sits at +2.49, a historical extreme, while the cost of holding it prints its own average. Normally a crowded side gets charged for the privilege. It is not being charged this week, which means the funding spike that usually warns of crowded longs simply will not fire. The risk is not that a cascade is imminent; it is that the usual early warning is muted.
8. Solana perp open interest adds 10.2% of book in a single week Open interest
Solana added 10.2% of open interest in one week to reach $2.51bn. That is the fastest build in the large-cap cohort and it happened while SOL spot fell 4.01% on the week. Open interest rising into a falling price usually means new positions are being established against the move rather than old ones being closed, which matters for how the book behaves if the level breaks.
Price fell while open interest added 10.2% of book in a week. That is new positions being opened against the move, not old ones being closed. The pattern reads as disagreement rather than trend confirmation, and the tell is what these positions do next: add, or capitulate.
9. Alt leverage unwinds: XRP sheds 11.8% of open interest, Dogecoin 13.5% Open interest
XRP lost 11.8% of open interest with price down 6.9%; Dogecoin lost 13.5% with price down 8.1%. TRUMP moved from the board's crowded long to its crowded short, with funding at -16.73% annualised. This is the deleveraging that was flagged the prior week finally arriving, and it is concentrated precisely where positioning had been most one-sided.
The unwind landed exactly where positioning had been most one-sided, which is what a healthy deleveraging looks like. It clears extremes rather than accumulating them. The version worth worrying about is the inverse - price down, open interest flat.
10. NEAR is the only perp above $150m of open interest with funding beyond two standard deviations Funding rate
NEAR carried funding of 24.31% annualised, the single outlier in an otherwise unremarkable funding tape. For everyone else the cost of carry normalised: bitcoin at 4.95% annualised (z +0.01) and ether at 5.36% (z +0.38) both sit essentially at their own averages. When open interest is at a +2.5 z-score and funding is at zero z, the market is holding a lot of position without paying for the privilege.
One outlier among the large books is a local story, not a systemic one. Single-asset extremes this far out are usually driven by an event or a listing rather than by market-wide leverage. The right response is to go and find the catalyst, not to read it as a signal about perp markets as a whole.
11. Realised volatility collapses faster than implied, and the variance risk premium inverts Volatility
Bitcoin's seven-day realised volatility fell from 60.99 to 27.46; ether's from 114.96 to 35.61. Thirty-day implied followed more slowly, bitcoin to 35.43 and ether to 49.30, leaving headline variance risk premium at -13.13 for bitcoin and -29.75 for ether, both below the tenth percentile of their own years. The twenty-day windows still carry the August rally at the 81st and 90th percentiles, which is why the front looks calm and the back end does not.
The front looks calm because the week was calm, not because the regime changed - the longer windows are still carrying the summer rally near the top of their annual range. Reading short realised volatility as risk having passed is the easiest mistake available this week.
The five assets where Hyperliquid, Binance and Bybit disagreed most on annualised funding included SKR at -4,079.6% on Hyperliquid against -4,376.8% on Binance and +11.0% on Bybit - a spread of 4,387.7 percentage points. Spreads of this size are not free money. They usually mean thin books, aggressive mark-price mechanisms, or a listing that one venue is pricing and another is not.
Spreads measured in thousands of percentage points annualised are not an arbitrage. They are a symptom: thin books, different mark-price mechanisms, or one venue pricing a listing another has not. When dispersion reaches that level the correct move is to find out why, not to trade it.
13. 21Shares lists a Hyperliquid exchange-traded product on SIX Swiss Exchange Hyperliquid
The Swiss listing follows the Bitwise spot HYPE ETF on NYSE Arca in May 2026. Regulated wrappers now exist for the venue's token in two jurisdictions while the venue itself remains geofenced out of the United States. That asymmetry - regulated exposure to an unregulated venue - is likely to be one of the more awkward questions for supervisors over the next year.
A European listing and a US pathway are different things and should not be read as one. This gives traditional brokerage accounts a regulated wrapper, which is real, but it does not resolve US access. It evidences offshore demand; it does not evidence a US opening.
Sources
Hyperliquid Policy Center asks the SEC and CFTC to treat qualifying equity perpetuals as security futures — Hyperliquid Policy Center, 24 August 2026. Source
Five former SEC and CFTC officials tell the agencies to stop duplicating oversight of the same risk — Chris Giancarlo, Sharon Brown-Hruska, Brian Quintenz, Steven Wallman, Chester Spatt, 27 August 2026. Source
Offshore perpetual volume put above $90 trillion for 2025, against roughly $28 trillion in 2023 — Kalshi estimate, cited by the former-officials comment letter, 27 August 2026. Source
HYPE prints an all-time high near $86.71 before settling back — Market data cited by CryptoBriefing, 27 August 2026. Source
Hyperliquid records $106m of fees in August, up 23% month on month — AInvest, citing DefiLlama, August 2026 (monthly). Source
AQA v2 directs roughly 90% of USDC reserve yield to HYPE buybacks — Hyperliquid, August 2026. Source
Bitcoin perp open interest sits at a +2.49 z-score while funding prints its own average — TradingRiot Market Digest, Week ending 30 August 2026. Source
Solana perp open interest adds 10.2% of book in a single week — TradingRiot Market Digest, Week ending 30 August 2026. Source
Alt leverage unwinds: XRP sheds 11.8% of open interest, Dogecoin 13.5% — TradingRiot Market Digest, Week ending 30 August 2026. Source
NEAR is the only perp above $150m of open interest with funding beyond two standard deviations — TradingRiot Market Digest, Week ending 30 August 2026. Source
Realised volatility collapses faster than implied, and the variance risk premium inverts — TradingRiot Market Digest, Week ending 30 August 2026. Source
Cross-venue funding dispersion on small-cap perps reaches absurd levels — Perps.com briefing, Week ending 30 August 2026. Source
21Shares lists a Hyperliquid exchange-traded product on SIX Swiss Exchange — 21Shares, August 2026. Source
Perp DEX category and venue data — DefiLlama, 31 August 2026. Source
Liquidations and futures aggregates — CoinGlass, 31 August 2026, 23:23 UTC. Source
Per-market funding and open interest — Hyperliquid public API, 1 September 2026, 14:02 UTC. Source
Bitcoin funding and coin-denominated open interest — CryptoQuant, 1 September 2026. Source
Stablecoin and price data — CoinGecko, 1 September 2026. Source
Volatility and z-score readings — TradingRiot Market Digest, 1 September 2026. Source
Swap and security-based swap definitions — SEC public filings, 27 August 2026. Source
Cross-venue funding dispersion — Perps.com briefing, 1 September 2026. Source
Hyperliquid-Payward talks — Bloomberg, reported by crypto.news, 31 August 2026. Source
Frequently asked questions
What moved perpetual futures markets in the week of 24-30 August 2026?
Regulatory structure moved more than price. Two comment letters to the SEC and CFTC - one from the Hyperliquid Policy Center on 24 August, one from five former agency principals on 27 August - reframed how perpetuals might be classified in the United States. On the market side, open interest built to +2 z-scores across six of the eight largest perps while funding stayed at its own average.
How large is the offshore perpetual futures market?
A joint comment letter filed with the SEC and CFTC on 27 August 2026 cited a Kalshi estimate of more than $90 trillion of offshore perpetual futures volume in 2025, against roughly $28 trillion in 2023. The figure is a market participant's estimate quoted in a comment letter, not an official regulator statistic.
Were funding rates elevated this week?
Broadly no. Bitcoin's annualised funding printed 4.95% with a z-score of +0.01 and ether 5.36% at +0.38, both essentially at their own averages. NEAR was the only perpetual above $150m of open interest with funding beyond two standard deviations, at 24.31% annualised.
Did on-chain perp DEXs gain or lose share?
The category lost committed capital while gaining turnover. Perp DEX venues cleared $533.63bn over 30 days while open interest fell -12.74% on the week to $22.44bn. Hyperliquid's fees still rose 23% month on month to $106m.
Where can the underlying numbers be checked?
Every figure in this review links to a primary or whitelisted source - exchange announcements, protocol documentation, regulator filings, DefiLlama, CoinGlass, CoinGecko, or the named market-data providers. The full whitelist is published on the Sources page.