Research is a work in progress. Estimated release: 1 December 2026. Our goal is to make it a search engine for trading setups in recorded crypto markets.

edgedepthEARLY ACCESS
Menu
Terminal
Recent moves
ResearchResearch hubResearch WorkbenchLiquidation mapsPublished reportsEvent archiveSaved studiesLive alertsData coverage
LearnGuidesReading libraryWorked examplesGuided lessonsFAQ
DevelopersConnect an AI agentOpen source
← All readings
Browse related readings

Research Readings Price and volatility

Spot/perpetual close basis

Contract-unit-normalized perpetual close divided by spot close, minus one, for the exact same fully closed Binance one-minute trade bars.

Full definition and limits

Signed fraction. Only perpetuals with a qualified Binance spot pair carry it (410 accepted, not the whole recorded universe; 1000-unit contracts are scaled), each inside its own dated intervals (feature_scopes in the research contract). Missing or zero-trade bars are absent. Descriptive closes cannot establish executable spread, sub-minute lead/lag, flow causation or arbitrage profit.

Live monitoring
Unsupported: historical only
Observed
from matching closed one-minute perpetual and spot bars
Unit
Signed fraction (0.01 is 1%)
Valid range
-1 or more

Example scope: BTCUSDT, ETHUSDT, SOLUSDT; 2025-07-15T00:01:00Z to 2026-09-11T00:00:00Z; perpetual close at or above spot. Only the 410 perpetuals with a Binance spot pair carry it, each on its dates. Review this scope before running.

Review scoped example
Technical details

Field ID

feature.spot_perp_close_basis

Valid query operators

at least, at most, between

Window operators: lowest, highest, average, latest or change in, over a trailing window.

Editable query preset

Example: A recorded value of 0 signed_fraction can be used in the editable condition below.

Starting point only. Review and edit before running.

{
  "schema_version": "research_query.v2",
  "normalization_version": "archive_normalization.v1",
  "feature_version": "feature_defs.v1",
  "target": "record_occurrences",
  "where": {
    "all": [
      [
        "feature.spot_perp_close_basis",
        "gte",
        0
      ],
      [
        "identity.symbol",
        "in",
        [
          "btcusdt",
          "ethusdt",
          "solusdt"
        ]
      ],
      [
        "times.anchor_time",
        "between",
        [
          "2025-07-15T00:01:00Z",
          "2026-09-11T00:00:00Z"
        ]
      ]
    ]
  },
  "sort": [
    "times.anchor_time",
    "desc"
  ],
  "page": {
    "limit": 5,
    "cursor": null
  }
}

The search reports how often the record held this condition and what followed. It does not decide whether an idea works.

Read alongside