Weekly Top Stories - 06/05/26

Welcome to Galaxy Research's Weekly Top Stories. Subscribe to get this newsletter delivered to your inbox every Friday morning.

In this week's edition, Alex Thorn analyzes the continued record outflows from spot bitcoin ETFs; Will Owens gives readers the tl;dr on the recent Polymarket imbroglio; and Thad Pinakiewicz revisits the debate about when stablecoin issuers should and shouldn’t freeze assets.

Got feedback on this newsletter? Email [email protected]. We’d love to hear from you.

Market Update

The total crypto market cap stands at $2.21tn, down 12.98% from last week (when it stood at $2.54tn). Bitcoin's network value is 4.02% of gold's market cap. Over the last seven days, BTC is down 15.01%, ETH is down 16.66%, and SOL is down 18.92%. Bitcoin dominance is 56.38%, down 255 basis points from last week.

Bitcoin ETPs See Record Outflows as BTC Slips Below $63k

U.S. spot Bitcoin exchange-traded products (ETPs) marked their 13th consecutive day of outflows on Wednesday, the longest such streak since they launched in January 2024.

Across those 13 days, the ETPs shed $4.33 billion (~60k BTC). On a 20-day rolling basis, the ETPs have unloaded $5.42 billion (~73k BTC), the biggest disposition across any 20-day period. This week is set to be the fourth consecutive week of net outflows.

At the time of writing, BTCUSD is trading below $64k for the first time since February and is now -49% from the Oct. 6, 2025, all-time high of $124,824. When compared to drawdowns that followed other prior all-time highs, Bitcoin’s price action in the last 240 days looks quite similar.

Galaxy Research’s proprietary Fear and Greed Index moved deep into “extreme fear” territory with a 13 print, among the lowest readings of the year. (The index combines onchain, derivatives, valuation, profit-taking, and ETF flows data).

Our Take

Bitcoin market sentiment is among the worst in years. After the November 2024 presidential election, longing bitcoin was the world’s most popular trade. Today, it’s nowhere near top of mind for the world’s investors – crowded out by a plethora of trades related to artificial intelligence.

Last year, many analysts said that BTC's famous “4-year cycle” would not repeat. This view was justified by a combination of declining “supply shocks” from the currency’s quadrennial halvings; an observed dampening of volatility; passive ETF flows; and the fact that bitcoin made a new all-time high before the halving for the first time.

But when we examine the data, 2024-2026 cycle again looks a lot like the prior cycles. All three prior cycles took 403-441 days from October of the halving year to reach all-time highs, and the 2024-2026 cycle took 370 days. The three prior cycles each bottomed a remarkably consistent 12 to 13 months after prior all-time high (354-406 days), and we’re currently at 240 days after all-time high. If this cycle plays out exactly as prior cycles, the eventual bottom will occur sometime in Q4 of this year.

Whatever the reason, the similarities are real and observable, and not just on date ranges. The shape of many metrics repeated across all three prior cycles and is also repeating in 2026: CVDD, realized price, realized cap, liveliness, vaultedness, net unrealized profit/loss (NUPL), reserve risk, market value to realized value (MVRV), spent output profit ratio (SOPR), and the Mayer and Puell multiples today all resemble their behavior in prior bull markets to post-ATH drawdowns. Here are some examples...

Bitcoin has been declared dead at least 472 times, but what is “dead” may never die. Even if BTC dropped to $30k, it would be double its prior bear-market low. The buyers at that level are the low-time preference, digital gold believers, and they grow in numbers and holdings every cycle. If bitcoin graduates from “risk asset” to “digital gold,” it will be in no small part thanks to them. -Alex Thorn

Polymarket’s Bad Call

Strategy (formerly MicroStrategy) sold BTC for the first time since 2022, and a $375m Polymarket contract asking whether it would do exactly that just resolved NO.

On June 1, Strategy filed an 8-K disclosing the sale of 32 BTC between May 26 and May 31. The sale was worth ~$2.5m and was the company’s first disclosed disposal since December 2022 (0.0038% of its 843,706 BTC treasury). The sale clearly closed inside the window. The prediction market’s question was whether Strategy “sells any of its Bitcoin” by May 31, with “information from MSTR and onchain data” named as the primary resolution source.

When the filing hit, YES shares spiked from 10% to ~80%. Then Polymarket posted “additional context” (at 1 P.M. ET Monday): “no MSTR filing, onchain data, or credible reporting had confirmed a sale within the market’s timeframe, and confirmation achieved outside that window doesn’t qualify.” YES shares collapsed below 1c within seconds. Two NO proposals, two disputes, and 48 hours of final review later, the contract resolved NO. The YES holders, who correctly predicted the event, were wiped out. The June 30 and December 31 contracts in the same series resolved YES at ~99.9% without dispute; the entire controversy is over one deadline sitting on the wrong side of a filing date.

Resolution runs through the UMA protocol’s optimistic oracle and, on dispute, its token weighted Data Verification Mechanism. This is the system Polymarket has leaned on for 1,150+ disputed markets in 2026, already past its full-year 2025 total. A May Wall Street Journal investigation found most disputed-market votes come from the 10 largest wallets, that a majority of active voters are linked to live Polymarket accounts, and that one in five disputes is decided by someone holding a position in it.

Our Take

We continue to view prediction markets as powerful tools for information discovery and hedging. Polymarket itself put the pollsters to shame in 2024. But this is a clean failure of the one thing these markets exist to do.

Strip away the oracle mechanics and one fact survives: Strategy sold Bitcoin before May 31. Everyone who bought YES predicted the future correctly and the market told them they were wrong. A prediction market is supposed to price what will happen; when resolution diverges from what actually happened, the product is merely pricing how the platform will read its own rules after the fact. That’s worthless.

What makes it worse is how quietly the lever works. Polymarket didn’t reach in and flip results; it posted “additional context,” and UMA’s voters fall in line every time. Call it an override or don’t; the outcome is a platform that authors the rulebook after kickoff. The solution is boring and well-understood: fix the resolution language before a contract lists, say up front whether you’re settling on the event or its announcement, and let verifiable outcomes resolve automatically against a predefined data source instead of going to a vote.

Hyperliquid's HIP-4 gestures at this. Its bitcoin price markets are the trivial case: a daily binary settles to 0 or 1 against BTC's mark price on HyperCore (the same feed that already settles Hyperliquid's perps). Resolution is a deterministic data lookup with nothing to vote on, clarify, or dispute. For markets tied to offchain events, where no onchain number exists, resolution relies on Hyperliquid's validator set rather than an external token-weighted oracle. Each market is also defined up front with its resolution source and an optional challenge window.

The lesson runs the other way for Polymarket. The MSTR market had a clean source – an SEC filing naming MSTR's own data as primary – and got resolved on discretion anyway.

And then there’s the part that should worry Polymarket most. This is a company that recently acquired a CFTC-regulated DCM through its $112m QCX purchase, and reportedly wants to bring its main, global market back onshore. A registered exchange answers to settlement and conflict rules that a token-vote oracle simply cannot satisfy. Especially when the people adjudicating can be holding the bet. Two systems, one regulated and one resolved on vibes, can’t coexist under the same roof for long.

A properly functioning decentralized oracle is a worthy long-term goal, but it's hard to argue this is it. For more on this debacle, read our recent report. - Will Owens

To Freeze, or Not to Freeze

Freeze, unfreeze, freeze again — the past weeks have turned stablecoin issuers into a case study in indecision.

Circle locked up a Zama smart contract holding ~$12mm in USDC, only for a U.S. court to pry it back open three days later. It walked back one of the 16 wallets it had frozen over a sealed civil case. And it sat on its hands as ~$230m in proceeds from the Drift hack slid through its bridge. Meanwhile, a fresh U.K. sanction against Huobi drew a collective shrug from Circle and Tether alike.

All of which raises a question: who actually decides which tokens get frozen?

Our Take

Governments announce sanctions against this or that entity — for sanctions evasion, money laundering violations, a whole bevy of reasons. But governments can't freeze tokens. They depend on the token issuer to do it. So, the real question isn't which government or agency (or onchain Twitter detective) declares an asset should be frozen. It's whether the issuer cares to listen, and whether it acts.

Lately, issuers haven't been listening consistently. The last few weeks have made that clear, with issuers both overreaching in murky cases and sitting on their hands during obvious ones. Circle, in particular, has earned the criticism.

On one side, Circle has frozen funds aggressively in cases where its obligation to act was far from clear. It recently froze 16 wallets tied to a sealed U.S. civil case, then rapidly unfroze one of them. Just this past week it froze a Zama smart contract holding roughly $12m in USDC, locking up not just one user's funds but every user of the contract. That freeze was also reversed by a U.S. court after a three-day lockout.

On the other side, Circle has been strikingly passive on known, active threats. It received notice of the Drift hack in April and declined to intervene, allowing a reported $230m in stolen funds to move through Circle's CCTP bridge. Tether, by contrast, moved fast on the same confirmed exploit and froze north of $300m in USDT.

By the raw numbers, Circle is the lighter hand: 638 addresses blacklisted holding $120m USDC (~0.1% of supply), against Tether's 2,962 blacklisted addresses holding $1.7b USDT (~0.9% of supply). Circle blacklists fewer wallets, yet it clearly isn't positioning itself as the hands-off operator whose token can be used for anything. The record is genuinely mixed.

Nothing illustrated this discretion better than last week's response (or lack of one) to the U.K.'s sanctioning of Huobi, which now does business as HTX. (HTX claims the designation applies only to its parent company and not the exchange, though the UK names HTX and HTX Exchange explicitly.) The sanction landed, and the issuers did nothing. Neither Tether nor Circle blacklisted the HTX wallets.

Users didn't wait around. HTX pairs all of its markets against USDT, and once the sanctions news hit, roughly 90% of the USDT on the platform fled, over $100m — leaving about $17m on the exchange, with the majority of remaining stablecoin balances now sitting in TRON's algostable, USDD.

And it wasn't just users. The compliance infrastructure moved too: Bybit warned that deposits coming from HTX would likely face heightened scrutiny, and TRM Labs added the exchange to its sanctions list. The point isn't that the designation is somehow hollow. Other exchanges and compliance shops clearly saw HTX as a real risk. It's that the stablecoin issuers, alone, have been slow to act on it.

Why the inaction? Do the issuers only really care about U.S. sanctions, and consider everyone else’s as optional? Was there genuine nuance in the parent-company-versus-subsidiary distinction that gave them reason to hold off? Or was it simply the stablecoin cabal extending its friends the benefit of time? Worth noting: Justin Sun — Huobi advisor and purported ultimate owner — also effectively controls the TRON blockchain through its super-validator set. TRON is the de facto home of large-scale USDT activity, holding more than 50% of total USDT supply, which gives it enormous leverage over issuers and exchanges alike. Did that leverage over the stablecoin ecosystem tilt the issuers’ decision to not blacklist the exchange?

The honest answer is probably that issuers' inconsistent blacklisting is based on incentives, not ideology. Over-freezing has a real, demonstrated cost: the Zama contract and the 16-wallet freeze both got reversed, which means lawsuits, headlines, and the look of a censor that overreached. Under-freezing on a foreign designation costs an issuer essentially nothing unless a U.S. court forces the issue. The rational move is to freeze only when compelled and slow-walk everything else — which is more or less the pattern we keep seeing. It also explains the Drift contrast: returning clearly stolen funds is reputationally clean and legally cheap, while acting on a contested U.K. sanction against a TRON-aligned exchange is neither.

The other half of the answer is leverage: who can hurt an issuer. Circle is a U.S.-listed public company chasing a national trust-bank charter under the GENIUS Act and operating under MiCA in Europe; its existential regulators are OFAC, the OCC, the NYDFS. Tether sits in El Salvador, opted out of MiCA, and spun up a separate U.S.-regulated token so it could keep USDT offshore and pointed at the markets G7 regulators can't reach. For neither company does a U.K. OFSI designation threaten the franchise — a U.S. designation would, but a U.K. one is more of a guideline than a rule. The issuers aren't snubbing London out of principle; Britain just has nothing to hold over them.

Which brings us back to where we started. When issuers freeze one wallet and ignore another, what rules or guidelines are they actually following? The law? Their commercial incentives? A geopolitical pecking order where only U.S. courts count? Or just their relationships with the people who control the rails their token rides on? Right now, it looks like some blend of all four.

And there may be a colder logic underneath: every discretionary freeze chips away at the promise of a programmable dollar that simply works, so honoring every jurisdiction's list turns an issuer into a global censorship desk and erodes the neutrality that makes the asset valuable in the first place.

A sanction, on its own, compels nothing; it waits on the issuer's discretion. And which requests get honored tells you exactly who holds power over the dollar's plumbing at the intersection of TradFi and DeFi. - Thad Pinakiewicz

Other News

🇮🇷U.S. Treasury’s OFAC sanctions four Iranian crypto exchanges

🤖 Trump EO ‘asks’ AI devs to give U.S. 30-day peek at frontier models pre-release

💰BitMine files for $300m preferred offering at 9.5%, adapting MSTR playbook

🔌Jane Street in talks for new data center as compute power runs scarce

🔐 Zcash patches critical bug that could have allowed double-spending

🤯Anthropic: Claude now writes 80% of our code

⚛️ Microsoft unveils quantum chip targeted for 2029 (Bitcoiners, take note)

🪙 Someone redeemed a 12-year-old “physical bitcoin” worth $1.78m

⁉️️Polymarket thinks Kalshi may be spying on its offices (Kalshi denies it)

🚦 CFTC greenlights perpetual futures for crypto on licensed exchanges

Charts of the Week: Solana's Struggles and HYPE's Heights

Solana's perpetuals volume and open interest market shares have fallen sharply from their November 2024 peaks, when the network captured roughly 16% of volume and 33% of open interest. Volumes have begun to recover in recent months as new exchanges came online, most notably Phoenix, lifting Solana's volume share back above 10%. Open interest has not followed and remains near 2%.

Solana has remained a leader in onchain activity, ranking first in DEX volumes for five consecutive quarters and falling to second in application fees last quarter after leading in that category from Q4 2024 through Q3 2025.

Diversifying onchain activity beyond speculative retail trading is a clear priority for the Solana Foundation and the broader ecosystem, with perpetuals a major focus. The breakout success of Hyperliquid and the wave of competitors that followed turned the perps market into one of the most contested in crypto, and Solana's general-purpose execution layer has been a difficult home for the product. Market makers need to cancel orders and requote prices in milliseconds and know where in a block their transactions will land.

The ecosystem is moving to close the gap. Jito's BAM Maker Priority plugin is testing deterministic top-of-batch execution for makers. Meanwhile, exchanges are expanding their offerings beyond crypto tokens to capture growing demand for commodity and equitiy derivatives that have helped keep Hyperliquid open interest on an upward trajectory. For more in-depth analysis of all Solana network activity, read our Q1 update, “Solana Q1 2026 Update: Defending Market Share in the Doldrums.”

Speaking of Hyperliquid, its native token HYPE is a standout, trading near all-time highs while other majors are in a slump. - Lucas Tcheyan and Thad Pinakiewicz