Does hot funding mark the top? Bitcoin's funding rate since 2019, tested
Bitcoin's funding on Binance came to −0.5% a year on Oct 14 (normal; the default, 0.01% every 8 hours, is about 11% a year), 1.8% over the last 7 days. On the 230 days since 2019 it ran hot (30% or more a year), Bitcoin bought at that day's close did +1.0% over the next 30 days on average, against +4.6% after the other days; on the 294 days under zero, +2.8% over the next week against +0.7%. No band is clear of luck (a luck test of 3 with both halves of the days agreeing): funding runs hot or under zero in a few long stretches, and a few of them decide the averages. Past patterns, not financial advice.
as of · Binance's BTCUSDT funding, Sep 11, 2019 to Oct 14, 2026
A perpetual future never expires, so the exchange keeps its price next to Bitcoin's with a payment between the two sides every 8 hours: when the contract trades over Bitcoin, the longs pay the shorts (funding over zero); under it, the shorts pay the longs. While the two prices sit close the payment stays at the default, 0.01% every 8 hours, about 11% a year; it runs higher when traders crowd into leveraged longs, and under zero when they crowd into shorts. "When funding runs hot, the top is in", the saying goes. Here every whole day of funding on Binance's BTCUSDT contract since Sep 11, 2019 is a buy of Bitcoin at that day's close, held a week, 30 or 90 days, each day's three payments added up as a year's rate.
Oct 14, 2026: −0.5% a year, normal; 1.8% over the last 7 days. Funding sat on the default on 25% of the 2,591 days.
each week's funding, a year's ratehot, 30% a year and overunder zero
Bitcoin's average next 30 daysevery day's averageunder each bar: its days
Funding, a year
Days
Stretches
Next 7 days
Next 30 days
Next 90 days
under zero −1% or lower
294
49
+2.8%
+8.5%
+20.9%
normal 0% to 11%
1,859
23
+0.6%
+4.3%
+14.2%
over the default 12% to 29%
201
29
+0.9%
+1.9%
+11.8%
hot 30% or more
230
19
+0.8%
+1.0%
+15.2%
every day
2,584
—
+0.9%
+4.3%
+14.9%
Days and stretches count the days with a week after them; a stretch ends after 7 days out of its band. Today's band is marked. Each band against all the other days over the same hold, with its luck test under it (✓ past the bar, both halves of the days agreeing; the halves on hover):
Funding, a year
Next 7 days
Next 30 days
Next 90 days
−1% or lower
+2.1% t 2.4
+4.7% t 1.5
+6.8% t 0.7
0% to 11%
−1.0% t −1.3
+0.0% t 0.0
−2.3% t −0.2
12% to 29%
−0.0% t −0.0
−2.6% t −0.7
−3.4% t −0.3
30% or more
−0.1% t −0.0
−3.6% t −0.6
+0.3% t 0.0
Hot funding (30% or more a year) came on 230 days in 19 separate stretches, the last on Dec 5, 2024. Bitcoin bought at the close of one of those days rose over the next 30 days 45% of the time (−2.0% in the middle case, +1.0% on average): −3.6% against the other days, −4.7% in the older half of the days and −3.4% in the newer. A luck test of −0.6, not past the bar of 3.0: 15 holds that share no day, and a few big swings decide the average.
Funding under zero (−1% or lower a year) came on 294 days in 49 separate stretches, the last on Oct 11, 2026. Bitcoin bought at the close of one of those days rose over the next week 60% of the time (+1.6% in the middle case, +2.8% on average): +2.1% against the other days, +3.5% in the older half of the days and +0.4% in the newer. A luck test of 2.4, not past the bar of 3.0: 84 holds that share no day, and a few big swings decide the average.
§ 03 · How it is tested
How it is tested
as of · methodology
A luck test (t) says how far a result sits from zero, in units of its own noise: about 2 is rarely luck for one test, but many tests at once need more, which is what the first cut is for.
Every whole day of funding is a buy of Bitcoin at that day's close: the day's three payments, at 00:00, 08:00 and 16:00 UTC, are all made before it, so nothing is read early. Neighbouring days share most of their week, 30 or 90 days, so the luck test does not count them as fresh evidence: each calendar day weighs in by how many of the band's holds, and of the other days', cover it, so a band counts as the holds that share no day, and each band is set against all the other days over the same hold, which takes out the years Bitcoin simply rose. The averages on this page use every day. With 12 tests (four bands, three holds), a result needs a luck test of 3.0 or more and both halves of the days on its side.
The bands were set before any number here was read: under zero, the default (11% a year, rounded to the whole percent), and 30% a year, almost three times it. Funding runs hot in manias and under zero after crashes, so its far ends come in a few long stretches: hundreds of days can be a handful of holds that share no day. One exchange's contract, Binance's, the largest; other venues' funding runs close to it. Moves are Bitcoin's own price, before trading costs.
§ 04 · Revisions1 dated
Revisions
as of · every change to its headline numbers, dated
This study is re-run every night on the whole history. It was first written on Oct 15, 2026; its numbers last moved on Oct 15, 2026. Each night its headline numbers moved is listed here, newest first.
Bitcoin over the next 30 days against the other days: after hot funding −4% · under zero +5%