by Sarah ConnorThree CEXs shut in a week: is the model dying, or is this just a cull?
BitMart announced it is shutting down, days after BitMEX and AscendEx did the same. Is centralized crypto exchange structurally on its way out — or is this ordinary bear-market attrition that is killing DeFi protocols at the same rate?
The chain of the argument
Wu Blockchain broke the news: BitMart suspended new registrations, deposits and new orders on July 26 and will cease all trading by August 26 — the second centralized exchange shutdown in a week, after BitMEX, and the third named platform this year counting AscendEx. Wu Blockchain's own thread supplied the backstory: BitMart raised at a $300M valuation in 2021 on the strength of altcoin listings, then lost nearly $200M to a hack at the end of that year and never fully recovered operationally. Simon Dedic read the pattern as terminal for the whole CEX category: an "extraction model" that depends on a constant supply of new users to exploit, now running dry in a bear market. Ruslan Khairullin pushed back on the framing rather than the fact — volume isn't rotating cleanly from CEX to DEX, it's consolidating to the top five exchanges of either kind, with everything mid-tier getting squeezed out regardless of architecture. Ash then posted the fuller ledger: alongside the CEX closures, a dozen DeFi protocols wound down in the same stretch, which undercuts any story where "onchain" is automatically the safe side of this. Justin Wu framed it as cycle, not architecture — every exchange looks unstoppable in a bull market, and it's only the downturn that tests which businesses were actually solid.
Volume is going straight to top 5 CEXes or onchain perp DEXes. Nothing in between makes it out.
named CEXs have shut down in 2026 — BitMEX, BitMart, AscendEx
Wu BlockchainDeFi protocols also wound down this year, per Ash's list — Zapper, Odos, Radiant, Goldfinch and more
@ahboyashthe 2021 hack that crippled BitMart and set up this shutdown
Wu BlockchainFigures as reported in the linked threads; TT has not independently verified BitMart's internal financials or the completeness of the DeFi shutdown list.
The two sides
For — the CEX model is structurally done
Dedic and the onchain camp
The extraction model has a fatal flaw: it needs a steady supply of victims. When those dry up, so does the business. A brutal bear market is actually healing the market.
@sjdedicProjects only list on a centralized exchange because they pay for it. When the exchange goes down, the paid listing — and any funds sitting on it — goes down too. Don't underestimate the power of DEX.
@KirjakulovEverything is tradable on a Solana DEX now. There's no need for CEXs.
@CryptoCurbMost exchanges aren't even necessary, and people are realizing that very quickly.
@TheArsonDragonAgainst — it's a cull, not a CEX death spiral
Khairullin, Ash & the skeptics
Volume is going straight to the top five CEXes or onchain perp DEXes — nothing in between makes it out. More tier-10 exchanges will shut before Q4. It's a barbell, not a rotation to DEX.
@Rus_KhairullinAlongside the CEX closures, a dozen DeFi protocols shut down this year too — Zapper, Odos, Radiant, Goldfinch, Step Finance and more. "Onchain" hasn't been immune to the same shakeout.
@ahboyashEvery exchange looks unstoppable in a bull market — volume booming, listings everywhere, users arriving. You only find out how strong the business really is when the cycle turns. BitMart just reached that point.
@hackapreneurFewer exchanges people can actually trust is the real fix here — crypto doesn't need more exchanges, of either kind, it needs fewer bad ones.
@FityethThe exchange, in their words
Why this is an investment question, not gossip
Strip away the headline and there's a real structural question: does a wave of CEX shutdowns mean capital should rotate toward DEXs, or does it mean sub-scale platforms of any kind — centralized or onchain — are the actual thing to avoid right now? The "CEX is dying" read is the easier story to tell, but it doesn't survive contact with Ash's list: DeFi protocols with real onchain distribution shut down in the same window, for the same underlying reason — weak unit economics once free-spending users left. The variable that predicted survival wasn't CEX versus DEX. It was scale, and whether the platform had a business model that worked without a constant inflow of new retail.
The TT desk thoughts
Treat this as a scale story, not an architecture story. Concentrate exchange exposure in the handful of platforms with real depth on each side — the top-tier CEXs and the leading onchain perp DEXs — and treat any balance sitting on a tier-two or tier-three venue as withdrawal risk right now, regardless of whether it's centralized or onchain. "It's a DEX" is not a safety property by itself; Zapper, Odos and Radiant prove that. If you're holding funds on a platform outside the top handful by volume on its own side of the market, the base rate from this week says move them, not wait for a headline to confirm it.
Keep reading
How perp DEXs actually work · DeBank alternatives for tracking exposure · All notes