The funding fee on 21 coins across five venues — per coin, not just bitcoin. Our own five-second capture, sitting on top of each venue's own funding archive back to 2019. Above zero the longs are paying.
+0.001% BTC · funding by exchange · Binance on Oct-22 2026, the last day in this build, measured and published by askthetape.com.
Holding a perpetual is not free: every few hours one side pays the other, and the size of that payment is the cleanest measure of who is desperate. The live premium page shows bitcoin; this one shows all the coins we record, on five exchanges. 🩸 read each venue against ITSELF, not against the others: the number is percent per that venue's own funding interval, and the intervals differ — measured on this build, Binance's rate is 4.40 times Hyperliquid's on the median coin, which is the interval, not the market. A coin paying far more than the rest OF ITS OWN VENUE is a crowded trade with a bill attached.
| leg | `funding_ticks` — own capture every 5 s (~17,209 ticks per venue-coin-day), daily mean · before our first day, each venue's own funding archive: Binance 2019-09-11 → · Bybit 2020-03-26 → · Hyperliquid 2023-05-12 → |
|---|---|
| cadence | 5 s, daily mean · archive = every settlement the venue published, averaged over the same UTC day |
| since | 2019-09-11 (the whole page) · our own capture 2026-04-11 (BTC, ETH, MON, SOL) · 2026-06-07 (the other 17) |
| 🩸 the archive is measured against our own capture, never trusted | 79,977 rows of venue history were spliced in, pulled 2026-09-14. Two things were checked before a single point went on the chart, on the days where the archive and our own capture overlap. THE LABEL: every settlement was tried on three alignments — the settlement's own UTC day, the day before, the day after — and the settlement's own day won on every venue: Binance 0.00087 percentage points on the settlement's own day against 0.00241 and 0.00238 for the two shifts; Bybit 0.00187 percentage points on the settlement's own day against 0.00349 and 0.00405 for the two shifts; Hyperliquid 0.00005 percentage points on the settlement's own day against 0.00034 and 0.00038 for the two shifts. THE SIZE, recomputed on every build and read as a SLOPE, not as a ratio of one coin: Binance back to 2019-09-11 on 21 coins (98 overlapping days per coin): slope of our tick on their settlement 0.99, coin to coin 0.96-1.00, furthest apart PENDLE 0.82, ARB 0.91, OP 0.92; Bybit back to 2020-03-26 on 21 coins (98 overlapping days per coin): slope of our tick on their settlement 0.89, coin to coin 0.81-0.94, furthest apart OP 0.70, AAVE 0.75, APT 0.75; Hyperliquid back to 2023-05-12 on 21 coins (98 overlapping days per coin): slope of our tick on their settlement 0.99, coin to coin 0.95-1.00, furthest apart ATOM 0.87, SEI 0.90, ETC 0.92. 🩸 Bybit's settled payments run about 11 % below our running tick across 21 coins, and the gap is not uniform by coin; on Binance and Hyperliquid it is under 2 %. That is what it says: our capture is the mean of the RUNNING rate the venue quoted all day, the archive is what it actually charged at settlement. The archive is placed strictly BEFORE our own capture, so the two are never mixed on one day — on every overlapping day this page shows OUR number. |
| 🩸 the interval is not constant over six years — the legs and the dates, counted | the number is percent per that venue's OWN funding interval, and a venue can change that interval for a coin without notice: the same market then prints a rate two to four times smaller from one day to the next. This build walks every archived leg, finds its long-lived billing regimes and NAMES the switches: **ONDO on 2025-01-16 (Bybit, 3 → 6 payments a day); TAO on 2025-01-16 (Bybit, 3 → 6 payments a day); TIA on 2025-01-16 (Bybit, 3 → 6 payments a day)**. On those coins the drop on that date is the exchange, not the market, and the chart marks the day with a dashed line. Background rate of odd days: Binance 0.0 % of archive days carry a different number of settlements than its usual 3 per day · Bybit 4.0 % of archive days carry a different number of settlements than its usual 3 per day · Hyperliquid 1.1 % of archive days carry a different number of settlements than its usual 24 per day. 🩸 61 archive days out of 80,038 are not drawn, and only these: 46 incomplete · 15 irregular. A day is drawn whenever its number of settlements equals what the leg was paying AT THAT TIME (its local mode over a 30-day window), so a coin that used to bill every 8 hours and now bills every 4 keeps ITS WHOLE HISTORY — the switch is marked, not deleted. What is dropped is a day that does not match its own era: the listing day with a single payment, or 46 of them where the venue's tape has a hole. |
| 🩸 OKX and Bitget start where we did | not laziness, and not a choice: checked live on 2026-09-14, OKX's `funding-rate-history` answers for about three months and Bitget's `history-fund-rate` for about 500 records back to 2026-04-13. There is no free archive to splice, so those two lines begin on our first day of capture and say so. |
| 🩸 the day the archive hands over | the dashed vertical on the chart is the first day of OUR capture for this coin (2026-04-11 for the four we started with, 2026-06-07 for the rest): everything left of it is the venue's own archive, everything right of it is our five-second tape. The first and the last day of every archived leg are dropped, because a listing day carries one or two payments instead of a full day of them and its average is a single fee, not a daily rate — the same guard our own leg has always had on its first and its running day. |
| 🩸 what the chart draws before our first day | the full daily archive is what this page computes on — every record, every average, every rank. What travels to your browser for the years BEFORE our capture is thinned to keep the page light (the step is in the data as `bucket`), and each bucket keeps its highest and lowest day, so the envelope of the line is real even where the dots are further apart. In the 3-year view and shorter, our own capture is drawn day by day; the all-time view is thinned END TO END to about 160 points (the Bybit open-interest lines: about 110), so there even the last months are buckets, not single days. |
| 🩸 open interest: one side, same as everywhere else | 🩸 ONE SIDE (Bybit's own singleOpenInterest) EVERYWHERE — one convention across the whole site, not per page. Bybit and Hyperliquid count BOTH sides of every contract in their public field, so both are halved. Bybit: checked again on the live handles 2026-09-12, openInterest / singleOpenInterest = 2.000000 on all seven symbols probed (BTC, SOL, HYPE, UNI and the TradFi books XAUT, SPX, TSLA — the exchange's convention, not one market's). Hyperliquid reports two-sided open interest — its own note of 2026-09-23: «Two-sided OI is calculated as the sum of long and short positions» — and our own record of its positions agrees: on no market do the tracked longs reach much past half of the field. We halve it to one side, like the other venues, while Binance 103,753.375, OKX 27,373.52 and Bitget 35,260.41 already publish one side and are left alone. Our own capture of the same hour sits within 0.1-0.2% of each one-sided handle and at 2.0014x of Bybit's single side. The correction runs over the WHOLE history — neither field has ever been one-sided, so there is no seam to date; until this correction, Hyperliquid was printed here at both sides. On THIS feed the correction was measured directly: over the days bitcoin appears on both, this capture and our 5-minute open-interest snapshots carry the same Bybit figure (median ratio 1.00002, n = 149, 5th-95th percentile 0.9989-1.0010), and that figure is the two-sided one — so the same halving applies here. What changed on 2026-09-12, last closed day 2026-09-11: Bybit BTC $4.068B → $2.034B, ETH $1.895B → $0.948B, SOL $0.633B → $0.316B; Bitget and Hyperliquid did not move. No total on this page adds the three venues together, so nothing was re-summed — the lines simply became comparable. |
| 🩸 Bybit open interest before our first day: the exchange's public archive | exchange's public archive → own capture since 2026-04-11 (BTC, ETH, MON, SOL) · 2026-06-07 (the other 17): left of the dashed line the Bybit line is the exchange's own daily open-interest history (`v5/market/open-interest`, its one-sided field), priced at the exchange's own mark close of the same UTC day; from the line on it is our five-second capture, halved to one side as above. 28,244 days of archive on 21 coins, back to 2020-08-04 (bitcoin). The size is re-measured on this build on the days both exist (111–162 days per coin): k = 1.0000 on each coin, r ≥ 1.00. The label was measured when the archive was pulled (2026-09-26): the bar the exchange stamps 00:00 UTC carries the open interest at the END of the day before, and that reading beat the stamped day and the day after on 21 of 21 coins, in the first 45 and the last 45 days of our capture. Days the exchange published a count but no price of its own are not drawn, because a dollar figure needs the venue's own price: ETH 145 days before 2021-03-15, SOL 108 days before 2021-10-15, UNI 257 days before 2021-11-30. Bitget and Hyperliquid start where our capture did. |
| policy | 🩸 open interest is only reported by Bitget, Bybit, Hyperliquid on this feed — Binance and OKX send none at all — so the open-interest view is explicitly three venues and never a five-venue total. The funding number is the mean of the day's ticks, in percent per funding interval, as each venue reports it — NOT normalised to a common interval. · export: from 2026-04-11 (rows before that day are the exchange's own archive and stay here) |
| 🩸 the lines are not rankable against each other | each venue publishes its fee per its own funding interval and this page prints it raw, so the axis mixes intervals. Measured on this build across 21 coins, the median ratio of |rate| is Binance : Hyperliquid = 4.40× — close to what the interval difference alone would give, before any market story. Compare a coin to the other coins ON THE SAME VENUE, or a venue to its own history; do not read one venue sitting below another as a cheaper market. Nothing is rescaled here because the interval is not in the capture: printing an assumed 8-hour conversion would be a number we did not measure. |
| gaps | measured on this leg, not assumed: the line has no point on 2023-07-02 … 2023-07-02 (1 day) on Hyperliquid; 2023-08-23 … 2023-08-23 (1 day) on Hyperliquid; 2024-08-15 … 2024-08-15 (1 day) on Hyperliquid; 2026-05-01 … 2026-05-06 (6 days) on Binance, Bybit, OKX, Bitget — it BREAKS there and nothing is drawn across it · the same week is missing from every exchange-tape leg we keep (order books, funding ticks, basis, liquidations), so it is one collector outage, not a property of this market |
| hourly window | prod funding_ticks — our own five-second capture on all five venues, hourly mean of the live rate (the same quantity the daily bar averages, in % per that venue's own funding interval; nothing is rescaled) · 4,531 hours actually captured inside a 4,685-hour span by data, 720 of them inside the last 720 hours · 105 coin-and-venue rows ride this one window, so how many of those hours it DRAWS is decided by the size cap on the window file, not by the carrier — the window's own header prints what it drew, and the daily chart under it is untouched · captured 2026-10-23T21:21:56Z |
| this build | data to Oct-22 2026 |
| policy | no backfill · read it here · export: from 2026-04-11 (rows before that day are the exchange's own archive and stay here) |
The chart itself is drawn in the browser: turn JavaScript on to hover a day, pick a window and draw your line. The numbers above come from the last build, Oct-23 2026 21:24Z.
What the tape says today · what's new · all 111 metrics