The $255 million Ether squeeze was a positioning reset, not a re-rate — here's how to read it

Open the Ether perpetuals tape the way you'd open a crime scene. On September 11, a hotter-than-expected U.S. inflation print set off a cascade that wiped out more than $255 million in Ether short positions within 24 hours — about $188 million of it in a single hour — and shoved ETH to $2,600, an intra-session gain of more than 8%. This is not the story of Ethereum suddenly becoming valuable. It is the story of a crowded short book getting force-fed its own risk, and it tells you more about positioning than about the network.

The detail worth stopping on first is the inversion. Core CPI rose 0.3% month over month against a 0.2% consensus — inflation ran hot. Risk assets usually sell into that. Instead, the tape ripped. That is the tell: the market was already positioned for the bad outcome, so every leveraged bear became the fuel. Hot data did not create demand; it created forced buying. The exact same print, hitting a neutral tape, would have done nothing.

Why the squeeze landed on ETH, not BTC

The mechanics of a squeeze are simple once you see the ledger. A trader who borrows and shorts Ether via perpetual futures is on the hook to buy it back. As price pushes against them, the funding rate — the fee paid between longs and shorts — turns punitive, and sooner or later the exchange force-closes the position. That forced buyback drives price higher, which triggers the next forced buyback. It is a chain reaction with a fuse, and the $255 million in ETH shorts was the powder.

Ether ate the explosion more violently than Bitcoin for a reason that is visible in the same numbers. ETH short liquidations ran to about $255 million against roughly $172 million for BTC, and ETH gained more than 8% while Bitcoin managed under 4%. The comparative data says the bearish book was more concentrated, and more aggressive, in Ether. The squeeze hit where the crowded trades lived. That is the observable input, and it gets checked before any narrative gets a vote.

What this move was not: no protocol upgrade, no spot-ETF approval, no product milestone announced into the candle. When an 8% day has no fundamental catalyst attached, the default reading is not "adoption." It is "unwind." Classify it that way until the wallet tape — not the thread — overrides it.

The wallet tape was already moving the other way

This is where the squeeze stops being noise and becomes a market structure worth tracking. Across the weeks before the pop, holders were pulling Ether off exchanges: 1.4 million coins left exchange wallets between June and late August, cutting the exchange-held supply by roughly 18% down to about 6.28 million. Over the same stretch, U.S. spot Ether ETFs collected about $1.63 billion in net inflows while Bitcoin funds netted a comparatively thin $173 million. The signaling tape and the headline tape ran in opposite directions — the headlines read "ETH lags for a cycle," while the coins quietly moved into self-custody and funds.

So the September 11 squeeze did not happen on an empty war chest. It happened on an exchange supply that had already thinned. That is the difference between a one-day pop and a leg that can extend.

The two readings, and the data that separates them

The $255 million number supports two readings, and you keep both until the tape votes. Reading one: the shorts capitulated, the borrowed fuel is burned, and if real demand shows up, there is open room above. Reading two: it was a single-session cascade with no structural marginal buyer, and price fades back toward where it started. Note that the fade already began intraday — by the U.S. close ETH sat near $2,516, up only a hair on the day, not 8%.

Watch three checkable inputs, and let them decide. First, funding: once it renormalizes toward zero, the cheap trade everyone was crowding is gone, and the easiest money in it evaporates. Second, does the exchange-supply drain continue or reverse — coins leaving exchanges is the difference between "demand came" and "leverage left." Third, does ETH hold above its own structure, the 50-day near $2,175 and the 200-day near $2,057, with the RSI near 63 rather than rolling over. One caveat on conviction: retail sentiment on Stocktwits stayed bearish through the rally, which is seasoning, not proof.

Set the expiry before you set the entry. This squeeze's edge expires the session the short book resets and funding normalizes — the forced buyer is a one-time resource, and an 8% candle will not print twice off the same positions. The move only becomes a trend if fresh spot demand, visible as continued exchange outflows and ETF intake, replaces the borrowed fuel; the macro floor for that story is "investors got overly hawkish," an argument that weakens the moment hotter core inflation reads as a longer-than-expected higher-for-longer Fed.

The number that made this session real is not 8%. It is the $188 million forced back within a single hour. That fuel is spent, and the squeeze cannot refund anyone else's bearish conviction. Run the screen again only when you can name the next marginal buyer — not the last forced cover — because that is the only buyer who prints a second candle that is not already priced.