Bitcoin Shorts Bear Bulk of $194 Million Liquidated Over 24 Hours

Shorts Bear Nearly $194 Million

A rolling 24-hour count of perpetual futures liquidations, taken at 04:53 UTC on Friday, October 2, records about $194.4 million in forced closures across the three largest markets, and shorts carried nearly all of it. Bitcoin (BTC) liquidations reached $135.2 million in the window, with 91.13% of that total coming from short positions. Ethereum (ETH) followed at $44.7 million, 72.7% of it shorts, and Solana (SOL) posted $14.5 million at an 81.25% short share. Better than seven in ten closed positions belonged to traders who had bet on lower prices on every one of the three assets, the arithmetic signature of a market moving the other way. The direction is unambiguous: Bitcoin price is up 2.1% over the same 24 hours and trades near $86,000 as of writing, about 0.7% below the level it held when the liquidation snapshot was captured. Ethereum sits near $2,722, up 0.2% on the day, and carried the highest long share of the three at 27.3%, still a minority. A liquidation is the forced closure of a leveraged position once its margin can no longer cover potential loss, and perpetual contracts, typically margined in major coins and stablecoins, close automatically at that threshold. Each forced buy-in of a short adds upward pressure, which can trip the next tier of stops, so a cluster like this tends to mark an accelerated move rather than a reversal. The count covers perpetual futures only and describes a full day of trading, not a single hourly spike.

Galaxy Counts Polymarket Retail Losses

A Galaxy Research study of Polymarket trader behavior found that 69.2% of retail accounts on the platform finished below breakeven, against cumulative losses of $338.9 million. The dataset covers 2.9 million wallets drawn from Polymarket's complete on-chain settlement record, with figures harmonized by the oracle network Stork. Polymarket settles outcomes on-chain through oracle resolution rather than the order-book mechanics of a typical DEX, which is what makes the full trading history auditable. The study takes in only the international platform, which operates separately from the company's US app, where traders complete standard KYC checks.

To separate humans from software, the report used a judgment-based cutoff of 50 orders per active day, since the order rate shows no natural break. The filter removed 125,429 accounts, which had executed 80.8% of all orders and finished ahead by $246.8 million. Among the retail wallets that remained, the average account ended about $3 lower, while the bottom 1% lost at least $4,804. Losing also changed behavior: within 30 days of a loss, 15.2% of accounts never returned to trading, versus 6.1% after a profit, though the report cautions it cannot link multiple wallets to one person, so some apparent quitters may have resurfaced under new addresses.

The findings line up with a Yale study showing that 3% of Polymarket traders captured 27% of profits. Galaxy classifies 44.1% of traders as specialists, meaning more than 60% of their markets sit in one topic, and sports makes up 47% of that group. Sports specialists were the weakest cohort: only 25.1% finished profitable, against 41.2% for tech and science specialists and 30.4% for generalists. The platform's recent US marketing push targets sports-focused traders, precisely the cohort with the worst results in the international data. Galaxy argues that nothing in the report undermines prediction markets as truth-seeking tools, and that a losing majority may be a structural necessity, since informed traders need uninformed flow to trade against.

Taken together, the two datasets describe the same retail dollar from two sides. The $194.4 million in liquidations over the 24 hours to Friday, October 2 is leverage being repriced mechanically between the accounts on the losing side and the ones on the winning side of the same move. The $338.9 million in Polymarket losses is direction picked badly, spread across 2.9 million wallets with the average surviving retail account down about $3. On our own reading, the settlement-level record Galaxy published is the primary document that matters here: it states plainly that a majority of participants can lose while the market still prices information correctly. Both totals now carry dates a reader can check.