PATTERN+ PREDICTIONS
Try Pro freeLog in
Updated

Research · every study re-run on our own database every night

Research notes

A page for each standing study - one question traders ask, tested on our own price history and re-run every night - a pattern explained and tested every day, and what our lab tested each week and what held. Every number comes from our own database and carries its date; nothing is typed by hand. Past patterns, not financial advice.

Latest 112 daily posts7 weeks of notes55 studies7,208,657 tests since RSS feed

Standing studies

one question each, re-run on the whole history every night

The cycle and the clock 17

Study1 revision

How long do crypto bear markets last? Every fall from Bitcoin's and Ether's all-time highs, and the way back

Bitcoin's bear markets since 2014: 847 days under the high in the middle case; today −52.7% from the high.

3 bear markets847 days under the high−77% the fall

first · updated

Study12 revisions

What if you missed Bitcoin's best days? Every day since 2014, and whether the best come right after the worst

Missing Bitcoin's 10 best days since 2014 cut +40% a year to +22%.

$78,092 held$13,263 10 best missed$766,287 10 worst missed

first · updated

Study3 revisions

DCA or lump sum? Bitcoin all at once or $100 a week, every first day since 2014, tested

All at once ended with more Bitcoin than $100 a week on 64% of first days since 2014, 19% more in the middle case.

64% all at once ahead+19% middle lead−22% deepest dip, weekly

first · updated

Study1 revision

The Mayer Multiple today: Bitcoin against its 200-day average, and what came after

Mayer Multiple 0.81 (under its average): under 0.80, Bitcoin did +5.6% over the next 90 days.

0.81 today$72,780 200-day average+5.6% under 0.80

first · updated

Study1 revision

Is the 200-week moving average Bitcoin's floor? Every close since 2017 against it, tested

0.87 times its 200-week average ($67,983): under it, Bitcoin did +116.6% over the next year.

0.87× today$67,983 200-week average+116.6% under it

first · updated

Study4 revisions

Is Bitcoin cheap under its 2-year moving average and dear at five times it? The 2-year MA multiplier, tested

0.66 times its 2-year average ($89,097): under it, Bitcoin did +136.1% over the next year.

0.66× today$89,097 2-year average+136.1% under it

first · updated

Study1 revision

Does the MVRV ratio call Bitcoin's tops and bottoms? Bitcoin against its realised price, tested

MVRV 1.10, Bitcoin 1 to 1.5 times its realised price: under it, Bitcoin did +116.0% over the next year.

1.10× mvrv$53,716 realised price+116.0% under it

first · updated

Study2 revisions

Does the Puell Multiple call Bitcoin's bottoms and tops? Under 0.5 and 4, tested

a Puell Multiple of 0.78: under 0.5, Bitcoin did +85.6% over the next year.

0.78 today$26.87M new coins a day+85.6% under 0.5

first · updated

Study1 revision

Does Bitcoin follow a power law? The line as each day could have drawn it, tested

0.47 times its power-law line ($125,920): 0.50 to 0.99 times it, Bitcoin did +168.8% over the next year.

0.47× today$125,920 the line5.31 slope

first · updated

Study7 revisions

Does Bitcoin follow the stock-to-flow model? Every close since PlanB published it

Bitcoin at 7% of the stock-to-flow model's price ($812,140); within a factor of 2 of it on 35% of days since it was published.

0.07× close over the model35% within 2× since 201913% over the model

first · updated

Study1 revision

Bitcoin's four-year cycle: where are we now? Every halving cycle since 2016, day by day

Day 944 since the halving: Bitcoin −52.7% from this cycle's high.

day 944 since the halving−52.7% from the cycle's highday 525 2016 cycle's top

first · updated

Study1 revision

Does the Pi Cycle Top indicator call Bitcoin's top? Every firing and every cycle, tested

the 111-day average 49% under twice the 350-day; within 30 days of the cycle's peak in 1 of 3 cycles.

1.02× today49% under to the line2 fired

first · updated

Study4 revisions

What's the best time of day to buy Bitcoin? Every hour since 2017, tested

Bitcoin's late US hours (21:00-24:00 UTC): +0.053% a day; the best hour 21:00 UTC.

21:00 UTC best hour23:00 UTC worst hour+0.053% late us hours

first · updated

Study2 revisions

Does the moon move Bitcoin? Every new and full moon since 2014, tested

New moon to full moon: +2.9% on average; full moon to new: +1.9% (158 lunar months).

+2.9% new to full moon+1.9% full to new moon+0.8 luck test

first · updated

Study6 revisions

Is Uptober real? Bitcoin and Ether month by month

Bitcoin rose in 9 of 13 Octobers, +14.5% on average.

9 of 13 octobers up+14.5% october's average54% any month up

first · updated

Study1 revision

Does Bitcoin dump before Chinese New Year? Tax day and the Santa rally too, every year tested

Before Chinese New Year: Bitcoin +6.9% on average, against +3.5% usually.

+6.9% before chinese new year+8.3% after chinese new year+0.4% before us tax day

first · updated

Study1 revision

Do weekdays, weekends and Fed weeks matter?

0 of 116 calendar and regime tests held.

116 tests0 first cut0 held

first · updated

Chart signals 11

Study16 revisions

What happens after a coin's 20% day?

A 20% crash with no news behind it did −6.1% against the market over the next week.

9,729 big days−1.1% pump, no news−6.1% crash, no news

first · updated

Study10 revisions

What happens after a coin hits a new all-time high?

A new all-time high: +1.7% against the market over the next week, −1.5% since October 2021.

1,760 new highs+1.7% the next week−1.5% since october 2021

first · updated

Study3 revisions

Do altcoins come back after an 80% crash? Every crash on our record, dead coins included, tested

15% of altcoins that fell 80% were back above their high within two years.

15% back from −80%8% back from −90%−14% the year after −80%

first · updated

Study3 revisions

How many altcoins make a new all-time high in the next cycle? Every coin at Bitcoin's top, followed to the next

Altcoins back over their old high next cycle: 16 of 214 (7%).

7% new high next cycle6% against bitcoin−92% at the low

first · updated

Study1 revision

What happened to each year's top 10 and top 100 coins? Every January on our record, a year on

Each year's top 10 a year on: 54% still in it, 25% ahead of Bitcoin.

54% still top 1025% beat bitcoin9% top 100 down 90%

first · updated

Study2 revisions

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

Bitcoin 31.9% under its 90-day high (a crash): after a deep dip, Bitcoin did +1.9% over the next 30 days.

−31.9% today$86,620 90-day high+1.9% after a deep dip

first · updated

Study1 revision

What happens after Bitcoin drops 10% in a day? Every 10% fall since 2014, and the day, week, month and three months after

After Bitcoin's 46 days down 10% or more since 2014, it did +11.5% over the next 30 days on average.

46 10% falls+11.5% after a 10% fall+5.1% after any day

first · updated

Study15 revisions

Golden cross, death cross: what came after? Bitcoin's crosses and the top 100 coins', tested

Bitcoin's 50-day average 7.1% over its 200-day: after a golden cross the top coins did −2.3% against the market over 90 days.

+7.1% todaygolden last cross−2.3% after a golden cross

first · updated

Study8 revisions

Does an oversold RSI bounce? Bitcoin's RSI and the top 100 coins', tested

Bitcoin's RSI at 17: after a fall under 30 the top coins did −0.6% against the market over 7 days.

17 bitcoin's rsi−0.6% after a fall under 30+1.7% after a rise over 70

first · updated

Study1 revision

Does calm come before the storm? Bitcoin's quietest weeks and the swing after, tested

Bitcoin's 30-day volatility 45% a year (busy): after the quietest days, the next 30 days' swing averaged 20.7%.

45% volatility25% calmer than20.7% swing after the quietest

first · updated

Study1 revision

Do CME gaps always fill? Bitcoin's weekend gaps since 2017, tested

CME gaps: 85% filled within a month, and 87% of the levels as far the other way.

85% filled in 30 days87% the other way42% gap's level first

first · updated

The crowd and leverage 11

Study2 revisions

Does buying Bitcoin at extreme fear work? The Fear & Greed Index, tested

Extreme fear: Bitcoin +4.3% over the next 30 days, against +3.7% on the other days.

12 today+4.3% extreme fear+3.7% other days

first · updated

Study1 revision

Is it altcoin season? Our altcoin season index, and what came after it

Altcoin season index 70 (neither): after an altcoin season the altcoins did −16.2% against Bitcoin over 90 days.

70 today35 of 50 beat bitcoin−16.2% after altcoin season

first · updated

Study3 revisions

Do altcoins follow Bitcoin? Their beta on its up days and down days, and whether the wildest pay, tested

The typical top-50 altcoin: beta 1.21 to Bitcoin, 1.21× its down days against 1.25× its up days.

1.21 beta0.47 correlation1.25× up days

first · updated

Study1 revision

Is Ether cheap against Bitcoin? ETH/BTC against its 200-day average, and what came after

ETH/BTC 0.0275, 0.94 times its 200-day average (under its average).

0.0275 today0.94 against its average+1.5% under 0.80

first · updated

Study2 revisions

Does hot funding mark the top? Bitcoin's funding rate since 2019, tested

Bitcoin's funding 1.9% a year (normal): after hot funding, Bitcoin did +1.0% over the next 30 days.

1.9% today3.5% last 7 days+1.0% after hot funding

first · updated

Study30 revisions

Do crowded shorts get squeezed? Altcoins' funding rates and the week after, tested

Crowded shorts: −4.6% against the other coins over the next week.

−4.6% crowded shorts+5.8% crowded longs8 · 0 today

first · updated

Study1 revision

Does the kimchi premium mark the top? Bitcoin's price in Korea against the world's, tested

Korean premium +3.0% (hot): at 10% or more, Bitcoin did −28.4% over the next 90 days.

+3.0% today1,337 won per dollar−28.4% 10% or more

first · updated

Study1 revision

Does the Coinbase premium lead Bitcoin? US demand against the world's, tested

Coinbase premium −0.000% (over its usual); after the year's highest premiums, Bitcoin did +5.8% over 30 days.

−0.000% coinbase premium+5.8% highest of the year−1.8% deepest discount

first · updated

Study1 revision

Do hedge funds' CME shorts sink Bitcoin? Every CFTC report since 2017, tested

Hedge funds on CME: net short 6% of open interest (the year's least short); after their year's most short, Bitcoin did +5.2% over 30 days.

−1,243 hedge funds, net6% of open interest+5.2% after the most short

first · updated

Study1 revision

Does Bitcoin move on options expiry? Every monthly and quarterly expiry since 2019, and the days around it

Bitcoin on 94 monthly options expiry days since 2019: 2.2% either way, +0.03% on average.

94 expiry days2.2% expiry day2.3% other fridays

first · updated

Study2 revisions

When everyone reads about Bitcoin, is the top in? Bitcoin's Wikipedia readers, tested

Bitcoin's Wikipedia readers at 1.4× the year's usual (usual): after a frenzy, Bitcoin did +14.1% over the next 90 days.

4,392 readers a day1.4× against the year+14.1% after a frenzy

first · updated

Money and the economy 8

Study3 revisions

Does new stablecoin money pump Bitcoin? The stablecoins' supply, tested

Stablecoin supply −6.0% in 30 days (shrinking): after a surge, Bitcoin did +17.6% over the next 90 days.

$283bn today−6.0% last 30 days+17.6% after a surge

first · updated

Study1 revision

Does a weaker dollar lift Bitcoin? The dollar index against Bitcoin, tested

The dollar +1.1% in 30 days (firming): after a falling dollar, Bitcoin did +43.0% over the next 90 days.

122.65 the dollar index+1.1% last 30 days+43.0% after a falling dollar

first · updated

Study2 revisions

Do Fed rate cuts pump Bitcoin? Every Fed decision since 2019, tested

11 Fed cuts since 2019: after a cut, Bitcoin did −1.6% over the next 30 days.

64 decisions−1.6% after a cut−5.1% after a hike

first · updated

Study2 revisions

Does Bitcoin go wild on CPI day? Every US inflation release since 2021, tested

a bigger move than its usual weekday on 57% of 70 CPI days.

70 cpi days57% bigger than usual51% closed up

first · updated

Study4 revisions

When the VIX spikes, does Bitcoin bounce? Wall Street's fear gauge against Bitcoin, tested

VIX 28.4 (nervous): at 30 or more, Bitcoin did +9.6% over the next 30 days.

28.4 the vix+9.6% after fear (30+)+7.2% after calm (under 15)

first · updated

Study1 revision

Does Bitcoin follow the stock market? Bitcoin against the S&P 500 and the Nasdaq, tested

Correlation with the S&P 500 0.48 over 90 trading days; after a Nasdaq bear market, Bitcoin did +8.3% over the next 90 days.

0.48 correlation, s&p 500−13.5% the nasdaq+8.3% after a bear market (20%+)

first · updated

Study2 revisions

Does money printing pump Bitcoin? The US money supply (M2) and the Fed's liquidity, tested

M2 +5.6% over a year (growing as usual): after fast money printing, Bitcoin did +28.5% over the next 90 days.

$23.4T us money supply+5.6% over a year+28.5% after fast printing

first · updated

Study2 revisions

Does miner capitulation mark the bottom? Bitcoin's hash ribbons, tested

Hash ribbons: no capitulation; after a recovery, Bitcoin did +10.2% over the next 90 days.

1,051 EH/s hash ratenormal ribbons+10.2% after a recovery

first · updated

Listings and unlocks 6

Study7 revisions

What happens to a coin after its Binance listing?

73% of 523 listings sat under their listing close six months later.

523 listings−47.8% after six months−30.1% against the market

first · updated

Study1 revision

Do new crypto coins go down after launch? Every new coin's first 90 days, and when it peaked

New coins under their first close 90 days on: 71% of 523.

71% under at day 90−36% middle coinday 8 peak

first · updated

Study18 revisions

Do token unlocks dump the price?

Token unlocks: −0.8% against the market over the 30 days before, +4.9% over the 30 after.

195 unlocks−0.8% 30 days before+4.9% 30 days after

first · updated

Study4 revisions

How many memecoins rug? Every Solana launch we recorded, tracked from birth

Solana memecoins: 87% ruined within 24 hours of launch.

6,748 launches87% ruined in a day52% alive at 1 hour

first · updated

Study6 revisions

What happens after an exchange lists or delists a coin that already trades?

After a Bybit spot listing, coins did −14.7% against the market over the next 30 days.

611 articles−14.7% after a bybit spot listing22% beat the market

first · updated

Study9 revisions

Do big caps move first?

Same day +0.74 to +0.89, a day later −0.10 to +0.03; 0 of 69 catch-up tests held.

+0.74 to +0.89 same day−0.10 to +0.03 a day later69 catch-up tests

first · updated

How the lab works 2

A pattern explained and tested, every day

the newest first · what happened after it, against ordinary days

Chart patterns

What happens to a micro-cap coin the day after a 90-day breakout?

−1.3% typical60% lower4,167 cases

2019-2026 · written Nov 21, 2026

Coin events

What happens to a small-cap coin in the 30 days after a usage surge?

+0.56% typical51% higher193 cases

2020-2026 · written Nov 21, 2026

Big market days

What do altcoins do the day after Bitcoin rises 3% or more?

+0.14% typical52% higher347 cases

2019-2026 · written Nov 20, 2026

Coin events

What happens to a micro-cap coin the day after a Korean discount?

−3.3% typical80% lower49 cases

2026 · written Nov 20, 2026

The economy and Wall Street

What do altcoins do the day after a US retail sales release?

−0.33% typical52% lower99 cases

2019-2026 · written Nov 19, 2026

Coin events

What happens to a micro-cap coin in the week after negative funding on a third day up?

−5.5% typical68% lower532 cases

2025-2026 · written Nov 19, 2026

One coin, one cause

What does XMR do the day after Ether falls 4% or more?

+1.3% typical67% higher207 cases

2019-2024 · written Nov 18, 2026

The economy and Wall Street

What does Ether do the day after a US retail sales release?

−0.58% typical57% lower100 cases

2019-2026 · written Nov 18, 2026

Coin events

What happens to a small-cap coin in the 30 days after a yearly high in Bybit longs while the price falls?

−10.3% typical68% lower462 cases

2021-2026 · written Nov 17, 2026

The economy and Wall Street

Big market days

Chart patterns

Coin events

One coin, one cause

Lab notes, week by week

what the lab ran each week, what it found and what stopped holding

This weekNov 16 – 22, 2026

Lab notes: the week of Nov 16 – 22, 2026

The lab has run 1,019,519 tests over 6 nights so far this week, the latest on Nov 21.

1,019,519 tests run1,304 new ideas247 holding15 first proven

updated every night · latest

Week 46Nov 9 – 15, 2026

Lab notes: the week of Nov 9 – 15, 2026

The lab ran 1,006,698 tests over 6 nights, Nov 9 to Nov 15.

1,006,698 tests run2,844 new ideas233 holding10 first proven

closed

Week 45Nov 2 – 8, 2026

Lab notes: the week of Nov 2 – 8, 2026

The lab ran 1,156,439 tests over 7 nights, Nov 2 to Nov 8.

1,156,439 tests run5,174 new ideas245 holding7 first proven

closed

Week 44Oct 26 – Nov 1, 2026

Lab notes: the week of Oct 26 – Nov 1, 2026

The lab ran 1,101,310 tests over 7 nights, Oct 26 to Nov 1.

1,101,310 tests run7,681 new ideas241 holding17 first proven

closed

Week 43Oct 19 – 25, 2026

Lab notes: the week of Oct 19 – 25, 2026

The lab ran 1,070,223 tests over 7 nights, Oct 19 to Oct 25.

1,070,223 tests run2,221 new ideas224 holding7 first proven

closed

Week 42Oct 12 – 18, 2026

Lab notes: the week of Oct 12 – 18, 2026

The lab ran 1,042,335 tests over 7 nights, Oct 12 to Oct 18.

1,042,335 tests run6,283 new ideas228 holding9 first proven

closed

Week 41Oct 5 – 11, 2026

Lab notes: the week of Oct 5 – 11, 2026

The lab ran 812,133 tests over 6 nights, Oct 6 to Oct 11.

812,133 tests run147,500 new ideas223 holding236 first proven

closed

The market, week by week

the weekly recap: our calls against what happened, the week's events and the movers

Weekly recapNov 9 – 15, 2026

The week of Nov 9 – 15, 2026

Each finished week, Monday to Sunday: our live calls against what happened, every dated event and what prices did after it, and the biggest movers. 15 weeks on record, a new one every Monday.

How these notes are written

Every night the build runs the lab: each coin against every kind of event, every collected series against Bitcoin, Ether and the altcoins, every trading rule the search builds, and the standing studies. Right after, it stores the night's numbers. These pages are those numbers in fixed sentences: a sentence whose number is missing is left out, never guessed.

Each night one more pattern gets its own post: a kind of event, a market day or a chart pattern, on a coin, on Bitcoin, Ether or the altcoins, with at least 30 measured cases. It says what happened after every past case against the same coins' ordinary days, why traders watch it, and what calls like it did out of sample. The families take turns, and a pattern is not written again within 180 days unless its numbers moved.

A week's notes are rewritten each night while the week runs and closed the first night after its Sunday; a closed week never changes. A study is re-run every night on the whole history, and each change to its headline numbers is listed with its date. Findings are proven only past a cut made for testing thousands of ideas at once; likely ones are leads, not facts.

More: The Lab: every model scored · What moves each coin · Which events move price · Methodology · Sources