"With a very heavy heart" — that's how BitMEX told users it would shut down on September 23 at 04:00 UTC, urging them to close positions and withdraw funds before the lights go out.
The exchange that invented the perpetual swap, reportedly once 57% of the global derivatives market at its 2019 peak and clearing $1 trillion a year, is now scheduling its own funeral and warning stragglers they'll be charged a $50 monthly fee for leaving money parked in a dead venue.
On the other side of the ledger, LMAX Group is reportedly working with Morgan Stanley and Stifel's KBW to evaluate strategic options at a valuation of up to $5 billion, weighing a sale, a SPAC merger, or a Nasdaq IPO.
Two exits, one week, opposite gravities. It's tempting to read both as symptoms of the same crypto downturn. That reading is wrong.
What's happening is a sorting. The industry is splitting its infrastructure into two piles: assets TradFi will underwrite, and relics the market winds down or auctions for scrap. The sorting mechanism is simple — can a bank build a deal around you.
The industry is splitting its infrastructure into two piles: assets TradFi will underwrite, and relics the market winds down or auctions for scrap.
Look at what LMAX actually sells. It runs institutional venues for foreign exchange and digital assets, regulated by the U.K.'s FCA, known for agency execution, transparent order books, and low-latency infrastructure — the plumbing banks, brokers and asset managers already recognize.
Its FX business is the tell. One person close to the process said LMAX is in no rush to list because that core FX franchise insulates it from the crypto downturn. The crypto exposure is reportedly upside; the FX cash flow is reportedly the floor.
That's why the number is $5 billion, up from the roughly $1 billion valuation implied when J.C. Flowers took a 30% stake for $300 million in 2021. And per E1, Ripple wrote a $150 million strategic check in January, tied to pushing RLUSD adoption through LMAX's settlement network. Regulated rails with real institutional flow command a premium.
Now look at what BitMEX sells. Or sold. A Seychelles-incorporated, offshore, up-to-100x-leverage perpetuals venue that pleaded guilty in 2024 to Bank Secrecy Act violations and paid $100 million in penalties.
Trump pardoned the founders in March 2025, wiping the criminal case. It didn't matter. Liquidity, market makers and whales had already migrated to venues with deeper books, more listings, and fewer legal hangovers.
By August 2023, CoinGecko ranked BitMEX ninth in derivatives with 0.9% of volume. By 2025 it had reportedly fallen out of the top ten entirely — while perpetual volume across those platforms climbed 47.4% to a record $86.2 trillion, per Cointelegraph. The product it invented thrived; the venue that invented it didn't.
That is the whole point. BitMEX's technology was foundational. Its exit multiple was zero. No bank underwrites a business whose defining trait is a decade of enforcement actions and a shrinking share of a market it no longer competes in.
And on the way out, the liabilities surface. BKX Services and David Namdar filed a proposed class action in the Southern District of New York on July 23, alleging BitMEX withheld collateral — 622.66 BTC, about $40.7 million — after liquidating positions while that collateral was still worth roughly twice the losses.
The complaint alleges BitMEX designed a system to retain customer collateral and route the balance to its insurance fund, and that an internal trading desk had access to private user data and could keep trading during server freezes that locked other users out. Those are precisely the traits that make a business unbuyable rather than un-fixable.
Roshan Dharia, CEO of Echo Base, called the shutdown "not an isolated event" but part of a wave of "structural corrections." The top five platforms now control an estimated 80% of global spot volume. His framing: "The headwinds are structural, not cyclical."
Structural is the right word. Regulatory compliance cost is no longer a footnote — it's a valuation input. The same period saw Coinbase, Kraken and Kalshi launch CFTC-regulated perpetual futures onshore. The offshore edge that made BitMEX has become the offshore liability that unmakes it.
TradFi is absorbing the parts of crypto it can regulate, license, and clear. Worth a single backward glance, because it's the same sorting logic playing out at scale.
Which brings us to the real story — the one that isn't a clean buy or a clean liquidation, but a stalemate at the price line.
B2C2, the London market maker 90% owned by SBI Holdings, has held takeover talks with several suitors over the past 18 months. The sticking point wasn't strategy or fit. It was valuation.
B2C2 wants north of $1 billion. One source said flatly that such a deal would be difficult to pull off in the current market. So it sits — a genuinely institutional business, spot, derivatives, structured products, OTC, 24/7 execution for banks and funds — unable to clear its own ask.
That gap is the tell for the entire sector. B2C2 isn't BitMEX; nobody's writing it off. But it isn't LMAX either, able to command a premium the market will pay right now. It's stuck in between, and the crypto downturn that suppresses trading volumes also suppresses the multiple.
Eighteen months of talks, more than a billion dollars in ask, and no deal — that's not a technology problem. It's a market that's pricing two different books. On one, regulated FX-backed infrastructure at $5 billion. On the other, a strong crypto-native market maker that can't find a buyer to meet its number.
And then there's the floor beneath the floor. Poolin, once nearly a fifth of Bitcoin's global hashrate, filed Chapter 11 in New Jersey on July 22 with roughly $173 million in debts against less than $10 million in assets.
The largest liability: about 11,700 wallet holders owed $163.7 million, frozen since September 2022 and papered over with IOU tokens that were never repaid. Their recovery now hinges on a $52 million stalking-horse bid from Thor CALAP LLC for two West Texas mining sites — sites that had already piled up $45.9 million in losses.
Do the arithmetic. A $52 million floor bid against $173 million owed, and that bid covers physical infrastructure only, not the frozen balances. The 11,700 users recover pennies, nearly four years after their withdrawals stopped.
That's the bottom of the sorting pile: assets so impaired they're auctioned for the hardware, users so far down the stack they get scrap. No premium, no buyer, no rescue — just a court-supervised liquidation setting a price.
Stack the three cases and the logic is unmistakable. LMAX gets underwritten at $5 billion because a bank can build a deal on FX cash flow and FCA regulation. B2C2 gets talks but no deal because its number and the market's number don't meet. BitMEX and Poolin get wound down and auctioned because there's nothing left to underwrite.
The technology was never the variable. All four built real things — perpetual swaps, mining pools, deep liquidity, order-book execution. Three of the four are being liquidated or stranded anyway.
The variable is reportedly the exit multiple, and the exit multiple is reportedly a function of one question: can this be absorbed by traditional finance on terms a bank will sign. That question now decides which pioneers survive as plumbing and which get sold for parts.
The market has already answered it. It's just publishing the results one press release at a time.