How our data works

Every number on RogueIQ comes from a named public source. This page shows where each one comes from and the checks it has to pass before you see it. RogueIQ shows where things stand and doesn't make calls.

Where each number comes from

EconIQ

  • Federal Reserve, BLS, BEA, Treasury, OFAC, DOJ, CFTC: each agency's own press releases and statements
  • SEC EDGAR: company 8-K filings (the company's own numbers)
  • Nasdaq Trader: trading halts
  • Release calendar: BLS, BEA and Federal Reserve schedules

MacroIQ

  • 10-year and 2-year yields, 2s10s spread: U.S. Treasury daily par yield curve
  • Brent crude: EIA daily spot (series RBRTE, pulled directly from EIA)
  • Broad dollar index: Federal Reserve H.10 (series JRXWTFB_N.B, pulled directly from the Fed Data Download Program)

BigMoneyIQ

  • CFTC Commitments of Traders: disaggregated and Traders in Financial Futures reports (positions as of Tuesday, released Friday)
  • Percentiles: our own math on CFTC history, recalculated on every refresh

HyperliquidIQ

  • Perp prices, funding, oracle: Hyperliquid's public API
  • Stock references: IEX last trade (one exchange, file published the next day)
  • Index and metal references: ETFs that track them (SPY, QQQ, GLD, SLV)
  • Oil references: EIA daily spot (WTI RWTC, Brent RBRTE, pulled directly from EIA)

Hyperliquid feed

  • Mark, oracle, funding, open interest: Hyperliquid's public API, refreshed every minute
  • Yearly funding: hourly rate × 24 × 365 (our own arithmetic)

Net worth by age

  • Federal Reserve 2025 Survey of Consumer Finances (SCF): Table 2 medians and means
  • Percentiles: our own math on the SCF public data file, using survey weights

We use agency data under each agency's public-domain or open-data terms and cite it on the tool page. Every series is pulled directly from the agency or exchange that publishes it.

How we check it

These checks run automatically on every refresh. If a check fails, the tool keeps showing the last value that passed and labels it with its date.

Sane range

Every new value has to fall inside a realistic band: yields between 0% and 20%, Brent between $5 and $400, and the broad dollar between 50 and 250.

Jump limit

If a value moves too far in one step, we hold it. That means more than 1 point for a yield, 40% for Brent or 5% for the dollar. We keep the last value that passed and say so on the page.

Freshness limit

Each series has a time limit: 4 days for yields, 10 for Brent, 12 for the dollar and 3 minutes for the Hyperliquid feed. CFTC data is due with each expected Friday release.

Arithmetic re-check

CFTC rows have to add up. Longs and shorts must fit within open interest, net must equal long minus short, and there can't be duplicate or future dates. We also recalculate each percentile ourselves.

Venue guardrails

We drop a Hyperliquid row if funding is above 4% per hour or the mark price is more than 20% away from the oracle.

Sample size

Weekend stats wait until there are 8 weekends of data. Usual-range bands wait for 7 days and 120 hourly snapshots. Thin CFTC and survey cells are marked.

Each number comes from one named source. We don't cross-check it against a second source yet.

Report a bad number

See a figure that doesn't match its source? DM @rogueiq on X with the page and the source link.

  1. We check it against the primary source.
  2. If it's wrong, we fix it.
  3. We note the fix and its date on the tool's page.