On-chain investigationHyperliquidZcash

Hyperliquid is Zcash's marketing department

Hyperliquid trades sixteen paper ZEC for every real coin that changes hands there. The candles, liquidations and price targets that market produces are the advertising Zcash never paid for. On the afternoon Grayscale listed the first spot Zcash ETF, four wallets opened accounts within three hours and thirty-nine minutes of each other, each with a bridge transfer of exactly 0.99 ZEC. Over six days they brought in 128,625 real ZEC, sold every last one into the crowd the campaign had drawn, and took out $103.2 million. The price went up while they did it.

At a glance
In the week of August 13, ZEC traded $340 million on Hyperliquid across 5,395 traders. Twelve days later Grayscale listed the first spot Zcash ETF in the world, and the same market did $6.46 billion across 22,244 traders. For every coin that actually changed hands on the venue, sixteen traded as paper that settles in dollars and never touches Zcash. On the afternoon of the listing, four wallets each received exactly 0.99 ZEC from the Unit bridge, then brought in 128,625 real coins and sold every one of them for $103.2 million. Every on-chain number here is linked to the query that produced it.
128,625
ZEC bridged in and sold by four wallets in six days
$103.2M
Proceeds, across 48,091 fills
0.99
Opening bridge transfer to each of the four, ETF day
67%
Of all UZEC the bridge minted that week, to 65 wallets

01 — The timelineRead the dates before the argument

Everything below is anchored to the day the ETF listed. Everyone informed was in position before it. Every exit ran after it. We have traced Hyperliquid's plumbing before, in the bridge ledger investigation.

Twelve days, one listing, and a $103M exitanchored to the ZCSH listing · UTC
12 days before
A quiet market
ZEC does $340M across 5,395 traders for the week on Hyperliquid. Nobody is watching it.
8 days before
The short squeeze begins
ZEC starts running from $486, with no headline attached to it.
6 days before
The venue gets its endorsement
Trump names Hyperliquid at a White House crypto summit and HYPE jumps 11%. The same day, the exchange's largest ZEC short starts building at ~$536.
4 days before
The crowd starts paying
Funding on the ZEC perp peaks at roughly 95% annualized, paid by longs to shorts.
2 days before
The hero image
ZEC prints $889.60, an eight-year high.
ETF day · 16:00–19:39
The exit is tested
Four wallets each receive a bridge transfer of exactly 0.99 UZEC, inside a three-hour-thirty-nine-minute window, hours before and after the bell.
ETF day
The bell rings
Grayscale's ZCSH lists on NYSE Arca. Hyperliquid records 1,220 short liquidations in a day, the worst for bears in the dataset.
1–6 days after
128,625 coins hit the book
The four take round tranches from the bridge and sell every coin they receive. ZEC rises from about $795 to about $845 while they do it.
7 days after
The crowd holds
Snapshot: 6,323 longs against 2,736 shorts, funding still positive, ZEC around $845.
The informed positions were taken before the listing. The distribution ran entirely after it.

02 — The paper market94% of ZEC traders never touch a coin

Two ZEC markets run side by side on Hyperliquid. The perpetual is a contract that tracks the price, settles in dollars and never touches a coin. The spot market is where real coins change hands. In August the perpetual carried almost all of the volume and almost all of the people; the spot market carried a rounding error. Nearly everyone trading ZEC on Hyperliquid never touches a market where ZEC moves. The table records both sides, and every figure in it opens the query that produced it in the Bitquery IDE.

ZEC perpetual — paperUZEC/USDC spot — real coins
12.75M ZEC traded, about $10B notional777K ZEC traded
29,622 traders across 5.96 million fills2,209 traders — 6.9% of the total
Settles in USDC, priced by an oracleCoins move; the bridge mints and burns them
Paper against coin — ZEC on Hyperliquid, August 2026ZEC perp (CoinRaw ZEC) vs UZEC/USDC spot (CoinRaw @272)
Volume — ZEC traded
Perp (paper)12.75M ZEC · 94.3%
Spot (real coins)777K ZEC · 5.7%
People — unique traders
Perp only29,622 · 93.1%
Touched spot2,209 · 6.9%
The perp settles in USDC and takes its price from an oracle. Roughly three quarters of Zcash's circulating supply traded as IOUs in one month, on one venue, while the coin itself barely moved. Sixteen paper coins for every real one.

None of those traders make shielded transactions. None of them run a Zcash node. What they produce is content. Every candle, liquidation and leaderboard position on Hyperliquid is public and machine-readable, and tracker accounts turn that feed into posts. @ZecStatsorg publishes each whale short as it is added, down to the entry and liquidation price. @zectrillionaire, three days after the listing: "ZEC short liquidations starting to roll in. Squeeze may hit soon." @alicharts, six days after: "Zcash $ZEC about to melt faces. First stop: $1,800." The tape is the ad copy, and in August it read well: ZEC up 76% in six days.

Hyperliquid markets the ticker rather than the technology behind it. It sells the chart to people who will never use the product. That is what made the next part possible.

03 — The campaignA 76% rally, and a landlord on the other side

The rally started eight days before the ETF listed, with no headline attached. Six days later ZEC had run from $486 to $854, up 76%. Part of that move was shorts being forced to close: every one of the last 3,000 ZEC liquidations on the exchange was a short position, and closing a short means buying, which pushes the price up again.

Hyperliquid publishes every position on the exchange, so we can see who took which side and when. 0xdd53c5297309130ab5fe5623dc905752e3342b13 began selling algorithmically six days before the listing at about $536, thousands of small fills, averaging up through $561, $672 and $790 as the price ran against it. A week later it held the largest ZEC position on the exchange, a $23.4M short at 3x.

Adding to a short through a 76% rally is a bad trade on price. This one was not being paid on price. The hourly fee crowded longs pay shorts annualized at roughly 95% at its peak four days before the listing, and the wallet collected about $170K just for holding. That is a landlord, collecting rent from the crowd on the other side.

The same day the short started building, the venue got its own endorsement. At a White House crypto summit, Trump said the CFTC chair was working to bring Hyperliquid into the United States "in a fully compliant and legal fashion". HYPE jumped 11%. That statement concerned Hyperliquid, not Zcash. But it put attention on the venue where ZEC's campaign was running, in the week it was running.

04 — The bellThe ETF lists and only bears get liquidated

On August 25, Grayscale's Zcash Trust converted into ZCSH on NYSE Arca, the first spot ZEC fund anywhere, holding about 387,000 ZEC. Its Head of Capital Markets went on Bloomberg to explain why Zcash is the next big investment after Bitcoin. The fund charges 2.50% a year, roughly ten times a Bitcoin ETF, and early fee revenue is earmarked for Zcash ecosystem development. The fund converts price appreciation into a development budget. Wall Street's product here is also marketing spend.

Nothing in that paragraph is alleged and nobody in it did anything wrong. An asset manager listed a product and promoted it. What follows is what happened on Hyperliquid the same day.

Listing day produced 1,220 short liquidations, the worst day for bears in our dataset. Of the last 3,000 liquidations on ZEC, every one was a short; zero longs have been liquidated in over a week. When only bears get hurt, longs keep arriving. A week after the listing, 6,323 accounts were long against 2,736 short, a crowd of small longs against a handful of very large shorts, paying those shorts rent every hour. Even the skeptics pointed the same way. @CryptoCharged, on listing day: "The time to build a Zcash spot bag is gone... Swing it or trade it on margin. Just do not start a bag here."

That is the audience the next section sold into.

05 — The exitFour wallets, one fingerprint, 128,625 coins

Real ZEC reaches Hyperliquid through Unit, a lock-and-mint bridge: deposit ZEC on Zcash, receive UZEC on Hyperliquid one for one. Unit's mint address, 0x54cd89623888e8010fdea1c62e86265a9c6da950, is published in Unit's own documentation and has never placed a trade. So when UZEC leaves that address for a wallet, real ZEC has just entered the exchange, and Hyperliquid's public API records it.

The bridge is the deposit ledgerUnit UZEC mint · Aug 24 – Sep 1, 2026
UZEC returned to the bridge
149,647
from 128 wallets (withdrawals)
0x54cd89623888e8010fdea1c62e86265a9c6da950
0 trades ever
UZEC minted out
192,450
to 65 wallets (deposits)
128,625
went to four wallets
67%
of everything minted that week
4 of 65
recipients
0
trades by the bridge itself
A mint out of this address means real ZEC was deposited by the receiving wallet. Two thirds of everything the bridge issued that week went to four accounts that had never held UZEC before.

On ETF day, between 16:00:14 and 19:39:09 UTC, the bridge minted

exactly 0.99 UZEC to four wallets that had never held it:

Wallet · first mint (UTC, ETF day)Bridged in · sold · proceeds
0x614e08cf0b6ab0248b040e0103be6ba61716b8e3
0.99 UZEC · 16:00:14
in 37,792.0 · sold 37,792.2 · $30.63M · 16,869 fills
0x75399cb6dbd484836e930e91a08bd29edf62460f
0.99 UZEC · 17:48:48
in 33,001.0 · sold 33,001.0 · $26.35M · 12,842 fills
0xa8178863d6f356623e07a429a94f4fd4c8f28b99
0.99 UZEC · 18:15:54
in 29,726.7 · sold 29,726.7 · $23.88M · 9,356 fills
0x7873f6df0e48980586834b62111b5d2566ec82b7
0.99 UZEC · 19:39:09
in 28,105.0 · sold 28,105.0 · $22.36M · 9,024 fills
Totalin 128,625 · sold 128,625 · $103.2M · 48,091 fills

Read the last two columns together. Each wallet sold, to within a rounding error, exactly what it brought in, and nothing was kept. The distribution ran across tens of thousands of individual fills at volume-weighted prices in the eight hundreds, and only one leg ever bought any back, a few hundred coins against tens of thousands sold.

Four legs, sized like one operationUZEC bridged in and sold, ETF day → +6 days
0x614e08cf…16b8e3first mint 16:00:14
37,792 · $30.63M
0x75399cb6…62460ffirst mint 17:48:48
33,001 · $26.35M
0xa8178863…f28b99first mint 18:15:54
29,727 · $23.88M
0x7873f6df…6ec82b7first mint 19:39:09
28,105 · $22.36M
Four accounts within a 3h39m window, each taking a comparable share. No single wallet carries the position; the size is split four ways and sold in parallel.

The identical opening transfer is the fingerprint. Small test deposits are not unusual on a bridge, so we checked. Of fourteen other wallets that took UZEC from the bridge in the same window, two opened with something similar, and those accounts handled a few hundred coins between them. Four wallets opening identically on the same afternoon, then each taking tens of thousands of coins, is not how independent depositors testing a bridge behave.

They are linked downstream too. As the proceeds left, the four routed assets through five relay wallets, each shared by two of the four legs:

0x5ee28b3ca856f9b1b6f600f965789fe7ba81316d
Shared by legs 0xa8178863 and 0x614e08cf. Zero trades in its entire history. Holds nothing.
0x212210befa7121eb7b40085258ff95806991f0b5
Shared by legs 0x614e08cf and 0x7873f6df. One trade ever; holds $0.001 of USDC.
0x1cc6ac689d6162710a934be77d161c2d9b74fcfd
Shared by legs 0x75399cb6 and 0x7873f6df. One trade ever; holds $0.0008 of USDC.
0x9541f53081a30f70309d537f2c569ba48f6eed51
Shared by legs 0x75399cb6 and 0x7873f6df. Two trades ever; holds $0.003 of USDC.
0xa026307ddfd12ff9684c78543c25d3f4a6427060
Shared by legs 0x75399cb6 and 0xa8178863. Zero trades. Received USOL from both legs on the same day.

Each has between zero and two trades in its entire history and holds dust today, under half a cent of USDC. They exist to receive and forward. The proceeds left as UBTC, USOL and HYPE. Tracing hops like these across chains is what our money flow tools and crypto investigation services are built for.

Between them, these four wallets took 67% of every UZEC the bridge minted that week, out of 65 recipients.

⬢ Why the bid was there
ZEC rose from ~$795 to ~$845 during the distribution
$103.2M
sold into the book in six days
+6%
ZEC price change across the same window
16:1
paper-to-coin ratio that made the bid deep enough

A sell that size needs a bid to match it, and ZEC's organic spot demand is nowhere near that deep. The bid existed because the campaign built it: crowded perp longs force market makers short, market makers hedge shorts by buying real coins, and borrowed size on the paper side becomes real buying on the coin side. The sellers sold into demand the advertising manufactured. And it held. Across the six days they were distributing, the price went up, as the panel above records. The price rising through the exit became the next advertisement.

06 — Nobody needed a planThe machine works without an operator

Advocacy pulls in margin buyers who want exposure rather than coins. Their crowding manufactures a spot bid through hedging, and pays large holders through funding to take the other side and wait. A holder with size sells into that bid and leaves without denting the price. The undented price becomes the next round of content, which recruits the next crowd, which rebuilds the bid for the next seller.

Grayscale listed a product. Hyperliquid ran a market. Unit ran a bridge. No coordination between them is required, and none of it needs any of them to be doing anything wrong. The only coordination we can see is among the four wallets, and that is visible because Hyperliquid publishes it.

One caveat. This structure has no loyalty. Three months earlier it destroyed the shorts, and the wallets currently short ZEC carry over $7M in realized losses from fighting this squeeze; the June tape shows ZEC falling 57% inside a week. The data here says distribution. It does not say top. What it does say is that on the day Zcash got its ETF, somebody with 128,625 coins already knew where the exit was.

How this was built

Every number in this piece is a query you can run

Trades, positions, liquidations and funding come from Bitquery's Hyperliquid data, with each figure linked to its saved query in the IDE. Bridge mints and account histories come from Hyperliquid's public info API. Market filters use Market.CoinRaw ZEC for the native perp, @272 for UZEC/USDC spot — never Symbol, which collides across HIP-3 and builder markets.

Hyperliquid TradesCurrentPositionsPerpLiquidationsPerpFundingsCoinRaw filtersHyperliquid info API
Open the Hyperliquid API docs Snapshot September 1, 2026, ~09:00 UTC
Scope, sourcing and what we do not claim

Trade, position, liquidation and funding data come from Bitquery's Hyperliquid API, snapshot September 1, 2026, around 09:00 UTC, with every figure linked to its saved query in the Bitquery IDE. Bridge mints, ledger transfers and account histories come from Hyperliquid's public info API (userNonFundingLedgerUpdates, userFills, spotClearinghouseState). Note that userFills truncates at 2,000 records, so all sale totals in this piece come from Bitquery, which returns full history.

Market filters use Market.CoinRaw (ZEC for the native perp, @272 for the UZEC/USDC spot pair) rather than Symbol, which collides across HIP-3 and builder markets. Unit's mint address is identified from Unit's published key addresses and confirmed by its complete absence of trading history. All relative dates are anchored to the ZCSH listing on August 25, 2026.

We make no claim about where these coins were before they reached the Unit bridge, or about who controls the four wallets. The behavioural links we describe are between those four wallets and nothing else, and they rest on four observable facts: the identical 0.99 opening transfer, the shared three-hour-thirty-nine-minute window, the sell-only pattern, and the five relay addresses each shared by two of the legs.

The title is a characterisation of market structure, not a description of any business relationship. Hyperliquid is a trading venue and Unit is a bridge; neither markets Zcash, and we are not aware of any agreement between them and the Zcash project, the Electric Coin Company or the Zcash Foundation. Nothing in this piece suggests Hyperliquid, Unit, Grayscale, NYSE Arca or any market maker knew of, facilitated or benefited from the wallet activity described, or did anything improper.

Grayscale listed and promoted a regulated exchange-traded product, which issuers routinely do; the promotional statements we describe are public and were made in the ordinary course of marketing a fund. The remark by President Trump is quoted from published reporting by The Block and CoinDesk, concerned Hyperliquid rather than Zcash, and is included only to date the attention on the venue. Nothing here suggests he, the CFTC or its chair had any knowledge of, or connection to, the trading described. Events are presented in sequence; sequence alone is not evidence that they are connected.

Social media posts are quoted as public commentary to show what the market was saying at the time. Quoting an account is not an allegation against it, and we make no claim that any quoted account is connected to the wallets in this piece.

One number we dispute: a widely-shared claim of $479M ZEC open interest on Hyperliquid does not match the position data, which sums to roughly $182M per side. This article is on-chain analysis for informational purposes and is not financial, investment or legal advice.

Run the next investigation in minutes

Ask any question of the blockchain. Bitquery's AI Investigations agent traces funds, attributes wallets, and writes auditable reports across every major chain.