Most crowded trade on Hyperliquid (cohort × coin heatmap)

One call to the heatmap returns every coin split by every cohort at the newest snapshot; rank the cells by net notional or by bias times share to find the most crowded trade.

GET /v1/hl-cohorts/heatmap answers it in one call: every coin with open positions and, inside each coin, one cell per cohort with its long and short notional, bias and share_of_symbol. A crowded trade is a cell that holds a large share of a coin and leans hard one way, so rank the cells by net notional (long minus short). The full parameter list is on Coin by cohort heatmap.

Quick answer

axis=size builds the board by account size.

cURL

# every coin by account-size cohort at the newest snapshot
curl -H "X-API-Key: $VIKE_API_KEY" \
  "https://data.vike.io/v1/hl-cohorts/heatmap?axis=size"

Example response

Captured from the live API on 2026-10-24 at 17:12 UTC, trimmed to 1 of 342 coins and 3 of its 12 cells; totals is complete. share_of_symbol is the cell's part of the coin's open notional, and bias runs from -1 (all short) to +1 (all long). BTC's 4x whale cell held a third of BTC and was net short $670M (bias -0.57). Ranked across all 342 coins, the largest net cells in this capture were ETH 4x whale (net short $676M, bias -0.89) and BTC 4x whale.

Example response

{
  "axis": "size",
  "opened_within": "all",
  "snapshot_ts": "2026-10-24T17:07:00Z",
  "source": "latest position snapshot",
  "totals": {
    "total_value": 15048887060.0,
    "total_long_value": 7520220245.0,
    "total_short_value": 7528666818.0,
    "bias": -0.0005612756098,
    "count": 423011,
    "n_symbols": 342,
    "n_wallets": 137770,
    "n_wallets_source": "precomputed totals"
  },
  "heatmap": [
    {
      "symbol": "BTC",
      "total_value": 3517530411.0,
      "total_long_value": 1755336015.0,
      "total_short_value": 1762194396.0,
      "count": 40907,
      "count_long": 28271,
      "count_short": 12636,
      "n_wallets": 40805,
      "bias": -0.001949771468,
      "segments": [
        {
          "cohort": "4x whale",
          "rank": 1,
          "total_value": 1166577215.0,
          "total_long_value": 248182815.0,
          "total_short_value": 918394400.5,
          "count": 22,
          "count_long": 7,
          "count_short": 15,
          "n_wallets": 22,
          "bias": -0.5745111224,
          "share_of_symbol": 0.33164666090751344
        },
        {
          "cohort": "3x whale",
          "rank": 2,
          "total_value": 314251727.9,
          "total_long_value": 213991225.1,
          "total_short_value": 100260502.8,
          "count": 27,
          "count_long": 17,
          "count_short": 10,
          "n_wallets": 27,
          "bias": 0.3619096165,
          "share_of_symbol": 0.08933873801513337
        },
        {
          "cohort": "2x whale",
          "rank": 3,
          "total_value": 227595594.5,
          "total_long_value": 136277749.2,
          "total_short_value": 91317845.39,
          "count": 42,
          "count_long": 20,
          "count_short": 22,
          "n_wallets": 42,
          "bias": 0.1975429439,
          "share_of_symbol": 0.06470323435641834
        }
      ]
    }
  ]
}

Variations

Rank every cell. Flatten the board and sort by the absolute net notional.

cURL — ranked

# the ten cells with the largest net long or short notional
curl -s -H "X-API-Key: $VIKE_API_KEY" \
  "https://data.vike.io/v1/hl-cohorts/heatmap?axis=size" \
  | jq '[.heatmap[] | .symbol as $s | .segments[] | {symbol: $s, cohort, bias, share_of_symbol, net: (.total_long_value - .total_short_value)}] | sort_by(if .net < 0 then .net else -.net end) | .[:10]'

By profitability. axis=pnl builds the same board by realized-PnL percentile. On this capture the largest net cell was BTC 3x smart, net short $322M.

cURL — pnl axis

# the board by realized-PnL cohort
curl -H "X-API-Key: $VIKE_API_KEY" \
  "https://data.vike.io/v1/hl-cohorts/heatmap?axis=pnl"

Only fresh positions. openedWithin=24h keeps positions first seen in the last 24 hours, so the board shows where new money is crowding.

cURL — last 24 h

# crowding among positions opened in the last day
curl -H "X-API-Key: $VIKE_API_KEY" \
  "https://data.vike.io/v1/hl-cohorts/heatmap?axis=size&openedWithin=24h"

Before you rely on it

A large response, and crowded against whom

The full board was 342 coins and about 1.2 MB in this capture, so cut it down client-side. A coin lists only the cohorts that hold it. Every long has a short, so each coin nets to about zero overall and a crowded cell is crowded against the other cohorts on that coin. The top cells are often a handful of accounts: check n_wallets before calling it a crowd.

Ask your agent

Connected to the Vike MCP server (https://vike.io/mcp) in Claude, ChatGPT or Cursor, ask it in plain words. The hl_segments tool answers (view heatmap), at 1 credit a call.

Ask your agent

What is the most crowded trade on Hyperliquid right now, by cohort and coin?

Most crowded trade on Hyperliquid (cohort × coin heatmap)

Quick answer

Example response

Variations

Before you rely on it

Ask your agent

Related