Large Positions

This page tracks the direction, size and changes of individual large accounts. On on-chain perpetual venues such data is read straight from the public ledger; on centralised exchanges it depends on voluntarily disclosed account-tier statistics.

The central trap in tracking large accounts is reading a position as an opinion. Big books routinely contain market-making hedges, cross-venue arbitrage legs and delta hedges against structured products — positions that look directionally committed but carry no price view. A sound filter excludes accounts simultaneously holding offsetting exposure and favours addresses with long holding periods and gradual scaling. On-chain, also account for a single entity splitting across many addresses.

Volume and market-structure metrics describe the size and distribution of trading activity. Methodologies differ sharply between venues (wash-trading filters, USD conversion), so confirm the accounting basis before comparing platforms side by side.

On-chain position tracking exposes fields that centralised venues never publish per account: entry price, leverage, margin mode, liquidation price and unrealised P&L, all recomputable from the ledger every block. The hard problem is entity resolution, because one trader operating five addresses looks like five traders, and clustering heuristics built on funding-source graphs, timing correlation and collateral movement recover only part of the picture. A practical watch-list discipline: track the same accounts across weeks rather than sampling whoever tops today's P&L board, note how their sizing responds to drawdowns, and treat sudden style changes, such as a swing account turning into a rapid-fire scalper, as a hint that the address changed hands or purpose. Aggregated per-account data also feeds the whale-positions and smart-money pages on this site, which apply exactly this longitudinal filter. What the ledger cannot show is off-chain context: an apparently reckless long may be one leg of a structure whose other legs live on a centralised venue.

The live chart for this metric is still being wired up. Global volume, venue share and perp-vs-spot pages already carry real data.