Bitcoin is at $86,035 this morning, up 6.7% in 24 hours. The total crypto market cap is back above $3 trillion for the first time since January. ETH is up 5%. The Nasdaq hit a record high. The 10-year Treasury yield fell back below 5%. Everything ran at once.
The move is real. The question is what is holding it up.
How This Started
Bitcoin’s initial breakout was fueled by short covering. Schwab’s head of crypto research Jim Ferraioli described it directly: “Bitcoin up 5% this morning due to short perpetual futures contracts being liquidated.” Short liquidations made up 96% of total Bitcoin liquidations on September 21. The largest single liquidation event was $290 million on September 21.
That is the mechanical part of the move. A crowded short position gets squeezed, forced buybacks amplify the rally, and price overshoots in the short term. It has happened before and it is not by itself a reason for concern.
What matters is what happened next.
The Leverage That Followed
Futures open interest increased to $61.55 billion, up 11.33%, or $6.26 billion, over two days. Across all tokens, open interest in perpetual futures has climbed to nearly $160 billion, the highest since late October last year, according to Coinglass data.
Since the breakout, about $2 billion in new leveraged exposure has been added, according to Coinalyze data, suggesting traders are placing fresh bets even as shorts got wiped out.
That is the part worth paying attention to. The shorts got liquidated. And then new leveraged longs immediately took their place. Open interest rising faster than price is the signal that the move is being extended by borrowed money rather than organic spot demand.
The funding rate sits at 0.0075% per eight hours, equivalent to an annualized 8.18%, while remaining below levels associated with excessive long-side crowding. That is the one piece of data that cuts against the concern. Funding is elevated but not extreme. The market is not yet in the territory where funding rates alone trigger a cascade.
What the Spot Side Actually Looks Like
Glassnode identified that exchange spot taker flow flipped from net selling to net buying as volume increased, while the monthly change in realized capitalization moved above its high band. That combination matters because it shows the rebound had buyers beyond those forced to close bearish positions.
US spot Bitcoin ETFs recorded approximately $433 million in net inflows on September 18, led by Fidelity’s FBTC with about $311 million and BlackRock’s IBIT with roughly $108 million.
But the same data set also shows ETFs saw a combined $746 million of outflows on Tuesday and Wednesday as the Clarity Act cloture vote failed in the Senate and the Fed hiked rates. Two days of strong inflows followed by two days of $746 million in outflows in the same week is not a clean institutional floor reasserting itself. It is a market trying to decide which signal to price.
The Three Things to Watch
According to Wintermute’s De Maere, traders should be watching ETF flows over the next few days, signs of excess in perpetual futures through inflated open interest or funding rates, and Friday’s options expiry. “So far this rally is looking pretty healthy,” he said.
The honest version of that is: the rally looks healthy until the leverage unwinds. Futures open interest, funding and realized profit-taking all sat above Glassnode’s bands, which is the data version of saying the market has moved fast enough that it has built a new vulnerability into the structure.
The macro catalyst that started all of this, the US Treasury’s announcement last month that it would increase buybacks of long-dated bonds, is real and structural. The market has added more than $740 billion in value since that announcement, according to CoinGecko data. That is not nothing. Lower long-term yields reduce the opportunity cost of holding Bitcoin. The thesis is intact.
What is also intact is the leverage sitting on top of it. $86,000 held by spot buyers is a different price than $86,000 held by perpetual futures traders with liquidation levels sitting somewhere below. The number is the same. The stability is not.
