Funding Rate History

How to read funding history

Each point is a funding settlement on the venue's own schedule. We convert every reading to an 8h-equivalent percentage so Hyperliquid's hourly prints sit on the same axis as OKX's eight-hour swaps. The sign tells you who paid: positive means longs paid shorts that period.

APR on this page is simple annualisation: average 8h rate × 3 × 365, without compounding. It answers "how big was carry if this pace lasted a year," not a forecast.

Why funding history matters

Sustained positive funding often means crowded longs; sustained negative funding, crowded shorts. History reveals carry cost for holding a direction, regime shifts around news, and whether extremes mean-reverted. Compare venues to see when Hyperliquid and OKX diverged.

Hyperliquid funding history

Core perps settle every hour. The API returns the hourly funding rate (and premium field); we multiply by eight for the chart. Hyperliquid caps the hourly rate at 4% per hour on core markets. Coins such as GOLD that live only on HIP-3 are omitted from Hyperliquid history here.

OKX funding history

USDT swaps usually settle every eight hours; we measure the gap between timestamps to catch four-hour or hourly instruments. Pagination walks backward with the after cursor until the window start.

CSV download

The CSV columns are: timestamp_utc, venue, symbol, funding_rate_raw, interval_hours, rate_8h_equivalent, apr_simple. Import into a spreadsheet to build your own charts or join with price data.

Methodology and limits

Data is fetched on request from public venue APIs, cached for about an hour, and capped at 90 days. Failed venues show a note but do not break the page. Our seven-day all-venue median line comes from history-v1.json in KV (hourly snapshots from this site's cron). See the methodology page and live spreads on the arbitrage scanner.

Using history for trade planning

History is descriptive, not predictive. A week of strongly positive BTC funding tells you longs paid heavily recently; it does not guarantee the next hour will repeat. Still, pairing history with live tables helps you avoid entering a crowded side at the worst moment, size carry into earnings or token unlocks, and sanity-check whether a venue's current rate is an outlier versus its own past month.

When you compare Hyperliquid and OKX on the same chart, look for persistent gaps (structural differences in liquidity) versus brief spikes (one venue's book moved first). The daily table smooths noise so you can spot weeks where both venues agreed on direction but disagreed on magnitude.

Interval normalisation in detail

Hyperliquid stores an hourly decimal funding rate. OKX stores the rate for the interval that just ended—usually eight hours. We divide OKX's raw rate by the measured hours between settlements (from timestamps) before multiplying by eight for the chart. If two OKX points are four hours apart, we treat each print as a four-hour rate, not an eight-hour one. That keeps the OKX line from looking artificially flat or steep.

APR on the stats cards is always derived from the average 8h-equivalent rate, not from compounding each period. That matches how APR is shown on the rest of the site.

CSV and spreadsheets

Each CSV row is one settlement event. Join on timestamp_utc with price candles from your data vendor, or pivot by venue to build custom charts. The funding_rate_raw column is what the venue published; multiply by your notional to estimate cash flow for that interval.

If you only care about one venue, set the venue filter to Hyperliquid or OKX so the chart and CSV omit the other exchange. The canonical link for SEO always stays on the base path without filters, while your bookmark can keep the query string you prefer.