Bitcoin Held. The Market Split.

Bitcoin was calm. The rest of crypto was choosing sides.

Bitcoin rose less than 1% over the fixed window. Ether did almost the same. Those two numbers make the week look uneventful. The rest of the market says otherwise.

Cardano and Ethena rose about 18.6%. Pump.fun gained 17.2%, and Zcash gained 8.8%. At the other end, XRP fell 4.2% and Solana slipped 0.8%. Hyperliquid rose 1.8% on the nearest available public range. These were not small differences around one common trade. Capital was selecting specific stories.

Now test whether performance was broad. AlphaRank removed stablecoins, tokenized cash and credit funds, and gold wrappers from CoinGecko's top 100. Of the 69 crypto risk assets left, 33 rose and 34 fell. Two were flat. The middle asset went nowhere.

Traders call this market participation breadth. This week had leadership without breadth. Rotational money chased a few strong narratives, but most of the market did not rise. That can be profitable, but it does not prove that the entire market has a new source of demand.

Cash entered through two doors while borrowed exposure stepped back.

Four different measures answer four different questions. Fund flows track demand through traditional brokerage accounts. Stablecoin supply estimates dollar-like cash already inside crypto. Exchange volume tracks activity in trading applications. Open interest counts futures positions that are still active.

Exchange-traded funds, often shortened to ETFs, let traditional brokerage accounts buy regulated fund shares linked to an asset. These particular funds hold bitcoin or ether rather than a futures contract. Across the five complete U.S. sessions from July 31 through August 6, bitcoin ETFs took in a net $498.2 million and ether ETFs took in $203.1 million. That is clear demand through regulated products. It did not produce a broad rally, so it was support, not proof that every part of crypto had turned higher.

Stablecoin supply rose from $305.51 billion to $306.09 billion. A $578 million increase is small beside the total, but direction matters because supply had shrunk the week before. Crypto's internal cash balance stopped moving backward.

A decentralized exchange, or DEX, lets wallets trade through software rather than a traditional broker. DefiLlama's protocol table recorded $45.53 billion over seven days. PumpSwap handled $4.48 billion and grew 14.31%. The provider's surfaces can use different comparison windows, so AlphaRank does not make a market-wide growth claim from this total. The useful conclusion is narrower: there was substantial activity behind the applications.

Borrowed exposure moved the other way. On OKX, the number of open BTC futures contracts fell 2.85% from July 31 to August 6. ETH fell 8.31%, and SOL fell 2.70%. Funding, the recurring fee that keeps these no-expiry futures close to spot prices, stayed positive but modest. Traders were not paying extreme prices to stay long.

The combined read is constructive and limited: net fund inflows, a little more cash inside crypto, substantial application use, and less borrowed exposure. That is sturdier than a speculative squeeze. It was still not enough to lift most large crypto assets.

The strongest proof came from things people used.

PUMP's token gained attention, but the more useful evidence was behind it: PumpSwap processed $4.48 billion of trades. Circle supplied a second kind of proof. Its digital dollars moved through the economy at a scale large enough to support a real public company.

Circle reported $73.3 billion of USD Coin, or USDC, in circulation at quarter end, up 19% from a year earlier. It also reported $14.8 trillion of transactions recorded on blockchains during the quarter, up 151%. Transaction volume is the value of transfers, not the money Circle kept. It shows use, not profit.

The business result was still substantial. Revenue plus income earned on reserves reached $701 million. After distribution and other direct costs, $289 million remained. Circle's preferred adjusted-profit measure was $143 million. In plain English, the digital dollars produced income, but the companies that distribute them took a large share.

Circle was not the whole stablecoin business. Tether reported $1.5 billion of operating profit for the quarter and a $4.11 billion reserve cushion above the tokens it owed customers. It also said secured lending fell by $2.38 billion, or 15%. These are company-reported point-in-time figures, not a full audit. They still show that stablecoin economics extend beyond one issuer and that reserve quality matters alongside growth.

The week's crypto-AI story belongs here because it touched actual payments. Circle said more than 900 paid services used its Agent Stack and 99.3% of payment volume on a small agent-payment network called x402 settled in USDC. Those are company-defined ecosystem measures, not proof that autonomous software is already a mainstream customer. They are early evidence that a stablecoin can become the cash register for software.

Ethereum also opened an institutional front door for companies exploring payments, tokenized assets, and settlement. That positioning matters, but it is not revenue. Circle's filing and PumpSwap's volume remain the stronger proof because they measure what customers actually did.

Strategy proved institutional adoption can create a seller.

A corporate bitcoin treasury is still a corporate treasury. It has dividends, buybacks, cash reserves, and financing costs. Those ordinary obligations can turn a long-term holder into a short-term source of bitcoin supply.

Strategy sold 1,638 bitcoin for $104.73 million at an average price of $63,957 during the filing period. It used about $52.4 million for preferred-stock dividends and $52.3 million to repurchase one class of preferred shares.

The sale did not end Strategy's bitcoin thesis. The company still held 842,138 bitcoin. Its total purchase cost was $63.51 billion, or an average $75,419 per coin, and it maintained a $4 billion U.S.-dollar reserve. The scale makes the lesson more important, not less.

What happened: a major corporate holder sold bitcoin to meet cash obligations. What it means: institutional ownership can produce supply as well as demand. Why it matters: investors can no longer assume every dollar raised by a bitcoin-treasury company becomes a new bitcoin purchase.

The research network agreed on resilience and use. It split on breadth.

AlphaRank compared 6,645 ready long-form records with 479,800 collected X posts across 21 active long-form sources and 77 active X sources. Exact identities and private excerpts remain in the audit record. Every public factual claim was checked against publishable evidence.

The strongest agreement was simple. Bitcoin absorbed the prior selloff. Stablecoins and trading applications kept working. The most convincing projects were the ones connected to payments, settlement, trading, or another visible customer action.

The disagreement was about what the selective gains meant. The optimistic case says rising stablecoin supply, steady exchange activity, lower futures exposure, and several double-digit winners are the first stage of a broader recovery. The cautious case says half the large risk assets still failed to rise, the middle return was zero, and Strategy became a seller.

A second week can resolve more than another round of opinions. Stablecoin supply needs to keep growing. A majority of large crypto assets need to rise. Application activity needs to persist after the token excitement cools.

This is a source-blind synthesis of AlphaRank’s private research network. Exact identities, timestamps, and provenance remain in the internal audit ledger.

Do not mistake narrative rotation for a broad market. Watch whether proof broadens.

A market label compresses too much. The useful question is whether more cash, more users, and more assets join the same move.

ADA, ENA, PUMP, and ZEC led this week for different reasons. That is selection, not one shared cycle. Bitcoin and Ether held. SOL and XRP did not. The next durable leader should have at least one advantage that survives a quieter market: customers, repeat activity, growing cash flow, or a policy change that opens distribution.

Macro matters only where it changes the price of money. The weak jobs report reduced pressure for higher rates. The next inflation reports will show whether that relief can last. If inflation stays hot, easier-money hopes can fade even while crypto-specific use improves.

The proof calendar

The strongest version of this market is not every token rising. It is more real money entering, more useful products being used, and more assets earning their place in the move.