Study · re-run every night · updated Oct 15, 2026

Does buying the dip work? Bitcoin's falls from its 90-day high, tested

Bitcoin closed at $81,682 on Oct 14, 5.7% under its highest close of the last 90 days ($86,620 on Sep 21): a dip. On the 1,384 days since 2014 it closed 15% to 30% under that high (a deep dip), Bitcoin bought at that day's close did +1.9% over the next 30 days on average, against +7.3% after the other days; 30% or more under it (a crash), +4.4%. No band is clear of luck (a luck test of 3 with both halves of the days agreeing): Bitcoin's big falls come in a few long stretches, and a few of them decide the averages. Past patterns, not financial advice.

Today

−5.7%

from its 90-day high

90-day high

$86,620

Sep 21, 2026

After a deep dip

+1.9%

Bitcoin, the next 30 days

After any day

+5.7%

Bitcoin, the next 30 days

First written Oct 15, 2026 · 1 revision

§ 01 · Today

Today's dip

as of Oct 15, 2026 · Bitcoin's daily closes, Mar 31, 2014 to Oct 14, 2026

"Buy the dip" may be the oldest advice in crypto: when Bitcoin falls hard, buy, because it has always come back. Here a dip is measured, not eyeballed: on each day, Bitcoin's close against its highest close of the 90 days up to it - within 5% of that high, a dip of 5% to 15%, a deep dip of 15% to 30%, a crash of 30% or more - and every such day is a buy of Bitcoin at that day's close, held 7, 30 or 90 days, set against buying on all the other days.

Oct 14, 2026: Bitcoin closed at $81,682, 5.7% under its highest close of the last 90 days ($86,620 on Sep 21, 2026): a dip.

Bitcoin under its 90-day higha deep dip, 15% to 30% undera crash, 30% or more under

§ 02 · After it

What came after it

as of Oct 15, 2026 · Bitcoin, from the day's close

Bitcoin's average next 30 daysevery day's averageunder each bar: its days

Under the high	Days	Stretches	Next 7 days	Next 30 days	Next 90 days
near its high within 5% of its high	1,130	51	+3.0%	+11.0%	+33.8%
a dip 5% to 15% under its high	1,250	93	+0.6%	+5.8%	+23.3%
a deep dip 15% to 30% under its high	1,384	65	+0.4%	+1.9%	+11.6%
a crash 30% or more under its high	787	27	+1.2%	+4.4%	+6.3%
every day	4,551	—	+1.2%	+5.7%	+19.4%

Days and stretches count the days with 30 days 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):

Under the high	Next 7 days	Next 30 days	Next 90 days
within 5% of its high	+2.3% t 2.9	+7.1% t 2.2	+19.2% t 2.0
5% to 15% under its high	−0.9% t −1.4	+0.2% t 0.1	+5.4% t 0.7
15% to 30% under its high	−1.2% t −1.9	−5.4% t −1.9	−11.2% t −1.3
30% or more under its high	−0.0% t −0.0	−1.5% t −0.4	−15.8% t −1.3

A deep dip (15% to 30% under its high) came on 1,384 days in 65 separate stretches, the last on Aug 18, 2026. Bitcoin bought on one of those days rose over the next 30 days 54% of the time (+1.3% in the middle case, +1.9% on average): −5.4% against the other days, −9.4% in the older half of the days and −1.3% in the newer. A luck test of −1.9, not past the bar of 3.0: 68 holds that share no day, and a few big swings decide the average.

A crash (30% or more under its high) came on 787 days in 27 separate stretches, the last on Apr 4, 2026. Bitcoin bought on one of those days rose over the next 30 days 57% of the time (+2.4% in the middle case, +4.4% on average): −1.5% against the other days, −0.5% in the older half of the days and −4.0% in the newer. A luck test of −0.4, not past the bar of 3.0: 34 holds that share no day, and a few big swings decide the average.

§ 03 · How it is tested

How it is tested

as of Oct 15, 2026 · 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.

Each day's high is the highest close of the 90 days up to and including it - only the past, so a day is never measured against a later top - and the day is a buy at its own close. Neighbouring days share most of their 7, 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. 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: 5%, 15% and 30% under the high. A fall is half of the story the saying leaves out: the dips that kept falling (2014, 2018, 2022) are in the same bands as the ones that bounced, and a crash's few long stretches can decide its average. Moves are Bitcoin's own price, before trading costs.

§ 04 · Revisions1 dated

Revisions

as of Oct 15, 2026 · every change to its headline numbers, dated

This study is re-run every night on the whole history: Bitcoin's daily closes from Jan 1, 2014 to Aug 16, 2017 are Bitstamp's, Europe's oldest exchange, and ours from the next day on (on the 3,346 days both have, the two closes differ by 0.06% in the middle case). 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.

Oct 15, 2026Bitcoin over the next 30 days against the other days: after a deep dip −5% · after a crash −2%

§ 05 · More

More research

as of Oct 15, 2026