Market Update: Hot-wallet outflows, 30 October 2026

Friday 30 October 2026, 15:00 AEDT. Neo-X desk. Covers 07:24 to 14:00 AEDT. Wholesale clients only. All figures are observations from the desk's market-data pipeline and are illustrative, not quotes.

At a glance

Australian spot traded in two waves after suspicious outflows from a global exchange's hot wallets began at 07:24 AEDT. Bitcoin was 4.7% below its pre-outflow level at the 13:04 low and 3.7% below at 14:00. Neo-X desk flow ran at 3.8 times a normal session for the same hours. Counterparties bought stablecoins and sold bitcoin and smaller tokens. Every Australian venue we track thinned, none stopped quoting, and most books were back near normal by midday. Neo-X was not affected: trading, settlement and custody ran as usual throughout.

Neo-X operations

Operating as usual. Neo-X was not affected by the outflows. Trading, pricing, onboarding and client support ran as usual throughout the session, and the desk quoted OTC and on Australian order books from the first outflow onward.

No exposure. Neo-X had no exposure to the affected exchange: it held no balances there and routed no settlement through it.

Settlement times unchanged. Settlement ran on its usual timetable. Crypto deliveries and AUD payouts on the session's tickets went out the same day through the Global Settlement Network, which settles same day across 27 currencies, with banking through Cuscal and Banking Circle AU.

Custody. Customer funds under custody are held with Copper, an independent custodian, not on exchange hot wallets.

Why diversified liquidity matters. Neo-X prices from about 25 venues plus its own inventory, as principal, rather than from a single exchange. When one venue is under stress, pricing and hedging continue through the others, and client settlement does not depend on that venue's wallets. Today one of the largest offshore venues was under stress and the desk kept quoting at its usual hours. Separating execution, liquidity and custody across independent providers limits how far one venue's failure reaches.

What happened

The outflow. On-chain data shows 167,464 ETH leaving one of a global exchange's hot wallets at 20:24 UTC on 29 October (07:24 AEDT, 30 October), with further transfers from a second exchange wallet in the minutes after (CoinDesk, 30 Oct). Blockchain-security firm Blockaid put suspicious transfers from the exchange's wallets at about US$3.3 billion across Ethereum, Arbitrum and Avalanche (Cointelegraph, 30 Oct). Chainalysis estimates about US$5.9 billion was taken (Cointelegraph, 30 Oct). The exchange has not yet published a loss figure or a post-mortem.

The response. The exchange paused withdrawals and then deposits, said it will cover all losses, and at 11:52 AEDT put the verified frozen total at US$3 billion (Cointelegraph, 30 Oct). Tether said it had frozen more than US$1.567 billion of USDT linked to the transfers, and Circle more than US$800 million of USDC (Cointelegraph, 30 Oct). At the time of writing the exchange's withdrawals had not resumed.

Price

Two waves. Offshore spot fell 2-4% in the 25 minutes after the first outflow. A second wave hit between 11:00 and 11:30 AEDT: bitcoin went from 3.1% to 4.2% down on the morning, SOL to 6.4% and LTC to 7.1%. Prices below are Bybit spot, one-minute closes, measured from the last minute before 07:24 AEDT.

Lows and 14:00 levels. Bitcoin's low was -4.7% at 13:04 AEDT. Ether's low was -5.9%, SOL's -6.5%, LTC's -7.2% and XRP's -5.4%. At 14:00 AEDT bitcoin was -3.7%, ether -4.6%, XRP -4.0%, SOL -3.9% and LTC -3.7%.

Local against offshore. After 11:00 AEDT local AUD books traded slightly rich to offshore. On one venue the BTC mid moved from about 4bps under offshore fair to 5bps over it. A second venue's AUD book stayed at fair throughout, so the move was local, not the AUD/USD leg.

Desk flow

Volume. Desk flow is Neo-X's OTC tickets and order-book fills combined, measured from the counterparty's side. From 07:30 to 14:00 AEDT desk flow ran at 3.8 times the 14-day average for the same hours, on about eight times the usual number of trades. The desk was quiet in the three hours before the news.

Timing. Selling came first: the first half hour after 07:30 AEDT ran at about seven times a normal half hour, and two-thirds of it was selling. The largest half hour was 11:30 to 12:00 AEDT, when buying alone ran at 67 times a normal half hour, driven by stablecoins.

Stablecoins. Stablecoins were 58% of desk flow, all of it counterparty buying, at 2.7 times normal. Including stablecoins, counterparty buys were 74% of desk flow.

Excluding stablecoins. Excluding stablecoins the picture reverses. Crypto flow ran at 8.8 times normal and 62% of it was counterparty selling.

Bitcoin. Bitcoin was 19% of desk flow at 6.6 times normal, and 80% of it was selling. The selling came as a steady run of small and mid-sized tickets through the session rather than one block.

Rotation. Selling in smaller tokens (26 times normal, 84% sells) was matched by SOL buying (34 times normal, 97% buys). Ether was 3.5 times normal and slightly net sold. XRP was 5.0 times normal and mostly bought.

These are observations of desk activity, not a view on direction.

Australian order books

Four venues. The table compares each venue's books from 07:24 to 13:00 AEDT with its usual level for the same hours (median, 15-28 October). Venues are lettered. Every venue thinned near the price, and none stopped quoting.

Where books held. Venue A's BTC spread averaged 9bps against 8 normal, its BTC offers stayed at normal and its BTC bids were back to normal by 12:00 AEDT. Venue C's BTC spread tripled in the first wave and recovered by 10:00 AEDT.

Where books stayed thin. XRP and SOL bids on Venue A were still at about half their usual size at midday. Venue C's XRP offers were at 5% of normal. Venue D's BTC spread ran at about twice normal, though on a very tight base, and its LTC book was empty within 0.25% of the price from the event onward.

Cost to trade A$25k. Bitcoin cost about 1bp more than usual to trade on Venues A, C and D. XRP and SOL were where costs rose: selling A$25k of XRP cost 9bps on Venue A against 6 usual, 19bps on Venue D against 13, and 23bps on Venue C against 13. Buying A$25k of SOL cost about double its usual level on Venues A and C. Every A$25k BTC, ETH, XRP and SOL order could still be filled in full on Venues A, C and D.

Stablecoin books. USDT was bought on Venue C at a 16-35bps premium to the deepest AUD stablecoin book for about ninety minutes after the first wave. That deepest book, at about 1.4bps wide, did not move. One venue's USDT/AUD quoter pulled both sides three times between 08:20 and 09:30 AEDT, then returned at its usual width.

What we are watching

Exchange statements, ongoing. A confirmed loss figure, the post-mortem and the timing of withdrawals reopening. These set how long stablecoin demand and the bitcoin selling persist.

Sunday 1 November. The exchange's next proof-of-reserves snapshot (CoinDesk, 30 Oct). It will show whether the affected ETH and token balances have been replaced.

Issuer freezes, ongoing. Further Tether and Circle freezes, and any release of frozen balances. These affect how much of the outflow can be moved or sold.

Tuesday 3 November, 14:30 AEDT. RBA cash rate decision (RBA meeting calendar). A local rate move on top of an offshore credit event would add to AUD stablecoin demand.

Data and method

Sources. Offshore prices: Bybit spot one-minute closes. Australian books and prints: the desk's market-data pipeline (level-2 books and public prints) and venue public APIs. Desk flow: Neo-X OTC tickets and order-book fills.

Normal. Normal is the 14-day average for the same clock window for desk flow, and the median of the same hours of day over 15-28 October for books. Flow multiples are against a baseline that includes weekends, so they are if anything conservative for a weekday morning.

Desk flow. Desk flow counts trades where a counterparty dealt with Neo-X: OTC tickets, and order-book fills where a counterparty crossed a Neo-X quote. The desk's own hedging trades are excluded. Fills on one venue whose data had not loaded for the session are excluded from both the session and the 14-day baseline, so the multiples compare like with like. Buys and sells are the counterparty's side; Neo-X took the other side as principal. n/a in the book table means the venue's feed or usual book is not usable for that pair.

—

Neo-X Financial Pty Ltd (ACN 674 294 127, AFSL 558430). Services are provided by Neochain Pty Ltd (CAR 001320366), a corporate authorised representative of Neo-X Financial Pty Ltd.

Wholesale clients only. This update is for wholesale clients within the meaning of ss 761G and 761GA of the Corporations Act 2001 (Cth), and sophisticated or professional investors under s 708. It is not for retail clients.

Principal disclosure. Neo-X acts as principal, makes markets in digital assets, and may hold positions in the assets discussed. Desk flow describes Neo-X's own trading activity in aggregate and does not identify any client.

Not advice. This update is general market commentary, not financial product advice. It does not take into account any person's objectives, financial situation or needs. Figures are observations from the stated sources, are illustrative and are not quotes. Digital asset prices can move sharply and liquidity can fall without notice.

Contact. sales@neox.au · compliance@neox.au · (02) 9136 0264 · Suite 3.03, 80 Mount Street, North Sydney NSW 2060 · Desk 8am–10pm AEST, 24-hour support.