Bitcoin spot ETFs pulled in about $930 million in net inflows over the first week of August 2026, while whale addresses added roughly $1.2 billion on-chain during the same stretch. Even with that apparent $2.1 billion in buying power on paper, BTC failed to hold $65,000 and briefly fell below $63,000, according to a MarsBit analysis that argues the market is seeing inflows without the kind of directional demand that can lift price.
The article contrasts the current setup with January 2024, when Grayscale’s GBTC began its ETF conversion and saw more than $6 billion in outflows over two months. Panic spread at the time, yet Bitcoin still rose from $42,000 to $73,000 because, in MarsBit’s telling, the money leaving GBTC was absorbed by real buying from newly launched ETFs.
ETF inflows are not the same as outright bullish demand
MarsBit says the "institutional bull market" narrative that followed approvals for spot Bitcoin ETFs from firms including BlackRock, Fidelity, and Bitwise now faces a serious test. If nearly $1 billion returning to ETFs still cannot move price higher, then ETF demand alone may no longer be enough to support the idea that institutional adoption automatically means a rising market.
The article points to the holder structure of IBIT and says a meaningful share of the money is tied to hedge fund arbitrage positions. The trade is straightforward: buy IBIT on one side and short an equivalent amount of CME futures on the other. As long as a futures premium exists, the trader can lock both legs and capture what MarsBit describes as annualized returns of 5% to 15%, with the spread at times rising above 40% in extreme conditions.
That is the classic cash-and-carry trade. Capital enters the ETF, and Bitcoin is indeed bought, but an equal-sized short is created in derivatives at the same time. The result is little or no net directional exposure. MarsBit also says quantitative analysis shows that even using the implied forward price from IBIT options against CME futures still leaves a 2.58% annualized spread after fees. As long as that window remains open, the article says arbitrage capital has an incentive to keep flowing in.
On that view, ETF inflow numbers may look strong on the surface while doing very little for spot price.
Authorized participants may also be dampening upside
MarsBit then turns to authorized participants, or APs, describing them as another, less visible force. These large banks and market makers are tasked with capturing small dislocations between an ETF’s market price and its net asset value.
When money enters the fund, they do buy Bitcoin in the spot market. But each purchase is paired with dynamic hedging. If price starts moving sharply, risk models can trigger trades in the opposite direction. MarsBit argues that a mechanism designed to smooth price gaps can become an amplifier when liquidity is thin, turning an ordinary pullback of a few percentage points into a sharper local drop.
The article says the risk is higher on weekends and in after-hours trading. AP risk systems do not distinguish between orderly adjustment and panic selling once a threshold is hit; execution becomes mechanical.
Miner selling and AI conversion costs are adding supply
On the sell side, MarsBit places miners at the front of the queue. After the April 2024 halving, block rewards were cut in half, yet total network hash rate kept rising, pushing hash price down to historic lows.
Citing CoinShares data, the article says the weighted average cash cost for listed miners to produce one Bitcoin has climbed to $76,000 to $80,000. With Bitcoin trading around $65,000, many miners are operating in a position where each newly mined coin is produced at a loss.
MarsBit says miners have little room to hold back supply. They still need to pay power bills, service debt, and fund capital spending tied to a shift into AI data centers. The article says on-chain data shows that whenever BTC rebounds above $60,000, miners’ one-day transfers to exchanges jump to more than 10,000 coins.
It also names Core Scientific and TeraWulf as examples of leading mining companies converting mining sites into AI data centers and signing computing contracts worth billions of dollars. Upgrading cooling and power infrastructure requires large sums of capital, leaving Bitcoin reserves as a ready source of cash. In MarsBit’s framing, this type of selling is tied less to market conditions than to construction timelines.
MicroStrategy’s debt profile is framed as a tail risk
The article also singles out MicroStrategy. It says the company holds more than 840,000 BTC, or about 4% of total supply, but also carries more than $8 billion in convertible debt maturing between 2028 and 2032.
MarsBit describes the funding model as one built on a high premium in the company’s stock relative to the net asset value of its Bitcoin holdings. The company has repeatedly issued zero-coupon convertible notes to raise cash and buy more Bitcoin. For that flywheel to keep turning, the article says Bitcoin needs to keep rising and the equity premium needs to stay elevated.
If price stalls for a long period and that premium narrows, access to funding could tighten. If maturities then collide with a period of weak liquidity, forced Bitcoin sales to repay debt cannot be ruled out, according to the article. MarsBit adds that market makers are already pricing this tail risk.
Holder cost basis is creating overhead resistance
Beyond institutions and miners, the article says on-chain positioning remains harsh. Short-term holders have an average cost basis of $69,000 and are broadly underwater. Each time price rebounds toward that level, break-even selling tends to appear.
Long-term holders are in less urgent shape, but MarsBit says they are still selling around 12,800 BTC per week to lock in gains. That leaves the market facing two layers of overhead supply at once: holders eager to exit near break-even and long-term holders steadily taking profit.
Derivatives hedging is absorbing spot momentum
MarsBit also lists a series of signals pointing to weak spot demand and stronger influence from derivatives desks. First, the Coinbase premium index has stayed negative for nearly 80 straight days, the longest such streak on record. The metric tracks active buying from U.S. institutions and high-net-worth clients, and the article reads the nearly three-month negative run as a sign that real U.S. spot demand has dropped to a cycle low even while ETFs continue to show inflows.
Second, Binance, described in the piece as the world’s largest offshore liquidity pool, is seeing a continued decline in cumulative spot volume delta. MarsBit says high-frequency deposit and withdrawal activity from large holders points to a stronger willingness to distribute than to build positions.
Third, the macro backdrop remains difficult. The article says U.S. inflation has proved far stickier than expected, shifting the market’s attention from when the Federal Reserve might cut rates to how long high rates may need to remain in place. With risk-free yields elevated, Bitcoin’s appeal in institutional portfolios is reduced because it does not generate interest income. MarsBit also says JPMorgan puts the probability of a global recession in 2026 at 35%, with risk aversion pushing hedge funds to cut crypto exposure.
In options, gamma positioning is described as defining the market’s trading boundaries. Between $62,000 and $65,000, market makers are said to hold large positive gamma exposure. Each time price pushes higher into that zone, they mechanically sell spot to keep delta-neutral books.
The article compares that high-frequency hedging behavior to a sponge soaking up what little bullish momentum is left in the market. Below that, in the $57,000 to $58,000 range, MarsBit says large negative gamma exposure has built up. If price falls into that area, hedging flows would reverse and become momentum-selling, increasing the risk of a faster decline.
The piece also says 25-delta option skew continues to tilt in a bearish direction. In plain terms, even investors already holding spot are willing to pay more for downside protection. MarsBit’s conclusion is that the spot market has surrendered a meaningful part of price discovery, with price increasingly constrained by derivatives market makers’ hedging algorithms.
MarsBit’s conclusion: Bitcoin lacks directional capital, not inflows
MarsBit closes by describing the nearly $1 billion in ETF inflows as a "perfect mask." Behind it, the article says, basis traders are neutralizing bullish momentum through financial engineering, miners are selling into strength, and MicroStrategy’s debt maturity wall starts looming from 2028. At the same time, the $69,000 short-term holder cost basis and options market gamma positioning are pinning Bitcoin inside a range.
The article argues the more likely outcome is neither a clean breakout nor an outright collapse, but a prolonged period of attrition. In its view, a material macro turn would be needed to change that, such as an unexpected drop in inflation that resets rate-cut expectations, or a spot move strong enough to push through the $69,000 cost basis and flip underwater holders back into profit. Until then, MarsBit says the $60,000 to $70,000 range remains Bitcoin’s cage.
Its final line is simple: Bitcoin does not lack inflows. It lacks money willing to make a real directional bet.