PATTERN+ PREDICTIONS
Try Pro freeLog in
Updated

Strategies · results · failures

The Lab

Every pattern we run, scored in the open: what works, what's on trial, what died.

Proven

0

On trial

4

Failed so far

11

Retired

0

Calls scored · 30d

282,760

Pattern scoreboard

walk-forward: every call made with only its own day's data · to Nov 15, 2026

Event Pattern

Past unlocks, listings and delistings lined up the same number of days out, and what price did after.

ON TRIAL
calls872
hit rate61.7%vs 62.0% always-down −0.3
skill+1.6%
luck test0.6σp 0.270
2025-11 · 27 calls · hit 77.8% vs 87.5% lazy call (−9.7 pts)2025-12 · 54 calls · hit 50.0% vs 55.0% lazy call (−5.0 pts)2026-01 · 66 calls · hit 100.0% vs 98.6% lazy call (+1.4 pts)2026-02 · 48 calls · hit 75.0% vs 74.5% lazy call (+0.5 pts)2026-03 · 89 calls · hit 58.4% vs 57.6% lazy call (+0.8 pts)2026-04 · 68 calls · hit 52.9% vs 61.8% lazy call (−8.9 pts)2026-05 · 91 calls · hit 75.8% vs 85.4% lazy call (−9.6 pts)2026-06 · 84 calls · hit 56.0% vs 67.0% lazy call (−11.0 pts)2026-07 · 92 calls · hit 65.2% vs 66.7% lazy call (−1.4 pts)2026-08 · 98 calls · hit 35.7% vs 76.9% lazy call (−41.2 pts)2026-09 · 91 calls · hit 41.8% vs 60.4% lazy call (−18.6 pts)2026-10 · 64 calls · hit 79.7% vs 91.9% lazy call (−12.2 pts)3 of 12 months beat the lazy call

Golden / Death Cross

The fast average crossing the slow one, golden or death cross, and what followed.

ON TRIAL
calls3,126
hit rate69.8%vs 69.8% always-down −0.0
skill+0.0%
luck test0.0σp 0.493
2025-11 · 112 calls · hit 83.0% vs 83.0% lazy call (+0.0 pts)2025-12 · 127 calls · hit 56.7% vs 56.7% lazy call (+0.0 pts)2026-01 · 123 calls · hit 83.7% vs 83.7% lazy call (+0.0 pts)2026-02 · 114 calls · hit 64.9% vs 64.9% lazy call (+0.0 pts)2026-03 · 188 calls · hit 57.4% vs 57.4% lazy call (+0.0 pts)2026-04 · 135 calls · hit 43.0% vs 57.0% lazy call (−14.1 pts)2026-05 · 318 calls · hit 76.4% vs 76.4% lazy call (+0.0 pts)2026-06 · 308 calls · hit 66.2% vs 66.2% lazy call (+0.0 pts)2026-07 · 283 calls · hit 69.6% vs 69.6% lazy call (−0.0 pts)2026-08 · 314 calls · hit 38.9% vs 61.1% lazy call (−22.3 pts)2026-09 · 480 calls · hit 65.4% vs 65.4% lazy call (+0.0 pts)2026-10 · 624 calls · hit 95.2% vs 95.2% lazy call (+0.0 pts)2 of 12 months beat the lazy call

Fresh All-Time High

Coins printing a fresh all-time high: continuation or top?

ON TRIAL
calls0
hit rate—no calls yet
skill—
luck test—
2026-03 · no calls2026-04 · no calls2026-05 · no calls2026-06 · no calls2026-07 · no calls2026-08 · no calls2026-09 · no calls2026-10 · no calls0 of 0 months beat the lazy call

Volume Spike

Volume explodes in an uptrend while Bitcoin holds up: what happened after every day like that.

ON TRIAL
calls0
hit rate—no calls yet
skill—
luck test—
2026-01 · no calls2026-03 · no calls2026-04 · no calls2026-05 · no calls2026-07 · no calls2026-08 · no calls2026-09 · no calls2026-10 · no calls0 of 0 months beat the lazy call

Breakout (30-day high / low)

Price breaking out above its recent range, or down below it, and whether the move kept going.

FAILED SO FAR
calls8,969
hit rate62.5%vs 63.1% always-down −0.6
skill0.0%
luck test—
2025-11 · 351 calls · hit 82.6% vs 81.6% lazy call (+1.0 pts)2025-12 · 641 calls · hit 37.3% vs 60.1% lazy call (−22.8 pts)2026-01 · 508 calls · hit 90.9% vs 90.5% lazy call (+0.4 pts)2026-02 · 950 calls · hit 75.7% vs 75.2% lazy call (+0.5 pts)2026-03 · 1423 calls · hit 43.1% vs 56.9% lazy call (−13.8 pts)2026-04 · 561 calls · hit 48.1% vs 51.9% lazy call (−3.7 pts)2026-05 · 1147 calls · hit 85.9% vs 85.9% lazy call (+0.0 pts)2026-06 · 536 calls · hit 68.8% vs 68.8% lazy call (+0.0 pts)2026-07 · 581 calls · hit 67.0% vs 67.0% lazy call (+0.0 pts)2026-08 · 907 calls · hit 33.1% vs 66.9% lazy call (−33.8 pts)2026-09 · 768 calls · hit 57.7% vs 57.7% lazy call (−0.0 pts)2026-10 · 596 calls · hit 88.8% vs 88.8% lazy call (+0.0 pts)4 of 12 months beat the lazy call

Capitulation Bounce

A brutal flush in a few days. Does the bounce come?

FAILED SO FAR
calls1,224
hit rate67.1%vs 67.1% always-down −0.0
skill0.0%
luck test—
2025-11 · 48 calls · hit 77.1% vs 77.1% lazy call (+0.0 pts)2025-12 · 55 calls · hit 45.5% vs 54.5% lazy call (−9.1 pts)2026-01 · 91 calls · hit 81.3% vs 81.3% lazy call (+0.0 pts)2026-02 · 192 calls · hit 63.5% vs 63.5% lazy call (−0.0 pts)2026-03 · 91 calls · hit 60.4% vs 60.4% lazy call (−0.0 pts)2026-04 · 72 calls · hit 45.8% vs 54.2% lazy call (−8.3 pts)2026-05 · 120 calls · hit 78.3% vs 78.3% lazy call (+0.0 pts)2026-06 · 178 calls · hit 67.4% vs 67.4% lazy call (+0.0 pts)2026-07 · 97 calls · hit 70.1% vs 70.1% lazy call (+0.0 pts)2026-08 · 114 calls · hit 57.9% vs 57.9% lazy call (+0.0 pts)2026-09 · 82 calls · hit 62.2% vs 62.2% lazy call (+0.0 pts)2026-10 · 84 calls · hit 90.5% vs 90.5% lazy call (+0.0 pts)0 of 12 months beat the lazy call

Green / Red Streak

Long runs of green or red daily closes in a row, and what the next leg did.

FAILED SO FAR
calls2,215
hit rate66.8%vs 65.1% always-down +1.7
skill0.0%
luck test—
2025-11 · 143 calls · hit 89.5% vs 88.3% lazy call (+1.2 pts)2025-12 · 157 calls · hit 42.0% vs 54.1% lazy call (−12.1 pts)2026-01 · 195 calls · hit 92.3% vs 90.0% lazy call (+2.3 pts)2026-02 · 397 calls · hit 80.4% vs 79.9% lazy call (+0.4 pts)2026-03 · 406 calls · hit 36.0% vs 57.6% lazy call (−21.7 pts)2026-04 · 86 calls · hit 44.2% vs 51.2% lazy call (−7.0 pts)2026-05 · 204 calls · hit 84.8% vs 82.6% lazy call (+2.2 pts)2026-06 · 135 calls · hit 71.1% vs 71.6% lazy call (−0.5 pts)2026-07 · 116 calls · hit 68.1% vs 65.7% lazy call (+2.4 pts)2026-08 · 121 calls · hit 37.2% vs 64.5% lazy call (−27.4 pts)2026-09 · 64 calls · hit 54.7% vs 52.8% lazy call (+1.9 pts)2026-10 · 191 calls · hit 91.6% vs 89.1% lazy call (+2.5 pts)7 of 12 months beat the lazy call

Trend Line (90-day average)

Above or below the long trend line, with the line rising or falling: what coins in that state did next.

FAILED SO FAR
calls47,316
hit rate62.7%vs 62.7% always-down +0.0
skill0.0%
luck test—
2025-11 · 1775 calls · hit 80.8% vs 80.8% lazy call (+0.0 pts)2025-12 · 3617 calls · hit 49.6% vs 50.4% lazy call (−0.9 pts)2026-01 · 3751 calls · hit 90.4% vs 90.4% lazy call (+0.0 pts)2026-02 · 3874 calls · hit 65.7% vs 65.7% lazy call (−0.0 pts)2026-03 · 4946 calls · hit 51.9% vs 51.9% lazy call (+0.0 pts)2026-04 · 3987 calls · hit 43.4% vs 56.6% lazy call (−13.2 pts)2026-05 · 5035 calls · hit 84.5% vs 84.5% lazy call (+0.0 pts)2026-06 · 4069 calls · hit 69.4% vs 69.4% lazy call (+0.0 pts)2026-07 · 4098 calls · hit 63.8% vs 63.8% lazy call (+0.0 pts)2026-08 · 5133 calls · hit 31.5% vs 68.5% lazy call (−37.0 pts)2026-09 · 4091 calls · hit 55.0% vs 55.0% lazy call (+0.0 pts)2026-10 · 2940 calls · hit 89.8% vs 89.8% lazy call (+0.0 pts)1 of 12 months beat the lazy call

Deep Drawdown

Coins sitting far below their high. Bottom-fishing, counted.

FAILED SO FAR
calls33,030
hit rate62.3%vs 62.3% always-down +0.0
skill−0.0%
luck test−1.0σp 0.830
2025-11 · 1027 calls · hit 83.7% vs 83.7% lazy call (+0.0 pts)2025-12 · 2188 calls · hit 45.1% vs 54.9% lazy call (−9.9 pts)2026-01 · 2227 calls · hit 92.9% vs 92.9% lazy call (+0.0 pts)2026-02 · 2715 calls · hit 67.6% vs 67.6% lazy call (+0.0 pts)2026-03 · 3516 calls · hit 51.9% vs 51.9% lazy call (+0.0 pts)2026-04 · 2834 calls · hit 43.0% vs 57.0% lazy call (−14.0 pts)2026-05 · 3515 calls · hit 87.6% vs 87.6% lazy call (+0.0 pts)2026-06 · 3013 calls · hit 71.8% vs 71.8% lazy call (+0.0 pts)2026-07 · 3058 calls · hit 65.4% vs 65.4% lazy call (+0.0 pts)2026-08 · 3878 calls · hit 27.2% vs 72.8% lazy call (−45.7 pts)2026-09 · 2972 calls · hit 52.9% vs 52.9% lazy call (−0.0 pts)2026-10 · 2087 calls · hit 91.8% vs 91.8% lazy call (+0.0 pts)0 of 12 months beat the lazy call

Rubber Band (30-day stretch)

How far price has stretched from its recent average, and what coins stretched the same way did next.

FAILED SO FAR
calls44,768
hit rate61.8%vs 62.7% always-down −0.9
skill−0.1%
luck test−1.4σp 0.914
2025-11 · 1746 calls · hit 80.8% vs 80.8% lazy call (+0.0 pts)2025-12 · 3528 calls · hit 49.2% vs 50.4% lazy call (−1.3 pts)2026-01 · 3379 calls · hit 90.6% vs 90.4% lazy call (+0.2 pts)2026-02 · 3818 calls · hit 65.6% vs 65.7% lazy call (−0.1 pts)2026-03 · 4701 calls · hit 51.6% vs 51.9% lazy call (−0.2 pts)2026-04 · 3581 calls · hit 41.6% vs 56.6% lazy call (−15.0 pts)2026-05 · 4414 calls · hit 84.5% vs 84.5% lazy call (+0.0 pts)2026-06 · 4068 calls · hit 69.4% vs 69.4% lazy call (+0.0 pts)2026-07 · 4098 calls · hit 63.8% vs 63.8% lazy call (+0.0 pts)2026-08 · 5126 calls · hit 31.4% vs 68.5% lazy call (−37.1 pts)2026-09 · 3699 calls · hit 51.6% vs 55.0% lazy call (−3.4 pts)2026-10 · 2610 calls · hit 90.2% vs 89.8% lazy call (+0.5 pts)6 of 12 months beat the lazy call

Last Year This Time

The same weeks in earlier years. Seasonality for fun, never counted in the vote.

FAILED SO FAR
calls25,503
hit rate56.4%vs 62.2% always-down −5.8
skill−0.2%
luck test−0.2σp 0.584
2025-11 · 761 calls · hit 71.9% vs 85.8% lazy call (−13.9 pts)2025-12 · 1617 calls · hit 46.3% vs 55.1% lazy call (−8.9 pts)2026-01 · 1396 calls · hit 80.7% vs 93.2% lazy call (−12.4 pts)2026-02 · 1272 calls · hit 60.9% vs 67.4% lazy call (−6.5 pts)2026-03 · 1777 calls · hit 50.4% vs 51.6% lazy call (−1.1 pts)2026-04 · 1550 calls · hit 54.1% vs 58.9% lazy call (−4.9 pts)2026-05 · 2807 calls · hit 81.2% vs 88.3% lazy call (−7.1 pts)2026-06 · 2731 calls · hit 48.1% vs 71.0% lazy call (−22.9 pts)2026-07 · 2803 calls · hit 48.5% vs 65.1% lazy call (−16.6 pts)2026-08 · 3700 calls · hit 36.1% vs 71.6% lazy call (−35.6 pts)2026-09 · 2963 calls · hit 57.1% vs 53.9% lazy call (+3.2 pts)2026-10 · 2126 calls · hit 69.9% vs 90.7% lazy call (−20.8 pts)1 of 12 months beat the lazy call

Bitcoin Shadow

Bitcoin's own prediction, scaled by this coin's beta: its average move for every percent Bitcoin moved.

FAILED SO FAR
calls32,810
hit rate59.4%vs 62.7% always-down −3.3
skill−0.4%
luck test−1.3σp 0.905
2025-11 · 1479 calls · hit 82.2% vs 80.8% lazy call (+1.4 pts)2025-12 · 2418 calls · hit 48.6% vs 50.4% lazy call (−1.8 pts)2026-01 · 1852 calls · hit 83.5% vs 90.4% lazy call (−7.0 pts)2026-02 · 3280 calls · hit 65.5% vs 65.7% lazy call (−0.3 pts)2026-03 · 3699 calls · hit 51.1% vs 51.9% lazy call (−0.8 pts)2026-04 · 2879 calls · hit 42.6% vs 56.6% lazy call (−14.0 pts)2026-05 · 3324 calls · hit 79.0% vs 84.5% lazy call (−5.5 pts)2026-06 · 3350 calls · hit 67.3% vs 69.5% lazy call (−2.2 pts)2026-07 · 2850 calls · hit 62.5% vs 63.8% lazy call (−1.4 pts)2026-08 · 3337 calls · hit 28.6% vs 68.5% lazy call (−39.9 pts)2026-09 · 2633 calls · hit 48.4% vs 55.0% lazy call (−6.6 pts)2026-10 · 1709 calls · hit 82.1% vs 89.8% lazy call (−7.7 pts)1 of 12 months beat the lazy call

Volatility Squeeze

The tightest range in a long while, and which way it broke.

FAILED SO FAR
calls8,834
hit rate59.1%vs 59.1% always-down −0.0
skill−0.5%
luck test−0.6σp 0.735
2025-11 · 65 calls · hit 76.9% vs 76.9% lazy call (+0.0 pts)2025-12 · 464 calls · hit 49.4% vs 50.6% lazy call (−1.3 pts)2026-01 · 917 calls · hit 92.1% vs 92.1% lazy call (+0.0 pts)2026-02 · 128 calls · hit 73.4% vs 73.4% lazy call (+0.0 pts)2026-03 · 1509 calls · hit 50.0% vs 50.0% lazy call (−0.0 pts)2026-04 · 1154 calls · hit 44.1% vs 55.9% lazy call (−11.8 pts)2026-05 · 1101 calls · hit 85.7% vs 85.7% lazy call (+0.0 pts)2026-06 · 219 calls · hit 66.7% vs 66.7% lazy call (−0.0 pts)2026-07 · 1169 calls · hit 57.9% vs 57.9% lazy call (−0.0 pts)2026-08 · 1155 calls · hit 29.8% vs 70.2% lazy call (−40.4 pts)2026-09 · 616 calls · hit 55.5% vs 55.5% lazy call (+0.0 pts)2026-10 · 337 calls · hit 86.1% vs 86.1% lazy call (+0.0 pts)1 of 12 months beat the lazy call

Market Breadth

How many coins trade above their own trend at once: the market's mood, read for every coin.

FAILED SO FAR
calls29,358
hit rate60.4%vs 62.8% always-down −2.4
skill−0.7%
luck test−1.5σp 0.928
2025-11 · 1775 calls · hit 80.8% vs 80.8% lazy call (+0.0 pts)2025-12 · 3617 calls · hit 49.6% vs 50.4% lazy call (−0.9 pts)2026-01 · 949 calls · hit 90.3% vs 90.3% lazy call (+0.0 pts)2026-02 · 3874 calls · hit 65.7% vs 65.7% lazy call (−0.0 pts)2026-03 · 3954 calls · hit 47.2% vs 52.8% lazy call (−5.5 pts)2026-04 · 1986 calls · hit 26.5% vs 73.5% lazy call (−46.9 pts)2026-05 · 3035 calls · hit 85.7% vs 86.3% lazy call (−0.5 pts)2026-06 · 4069 calls · hit 69.4% vs 69.4% lazy call (+0.0 pts)2026-07 · 2049 calls · hit 69.5% vs 69.5% lazy call (+0.0 pts)2026-08 · 2055 calls · hit 30.9% vs 69.1% lazy call (−38.3 pts)2026-09 · 1029 calls · hit 33.4% vs 58.5% lazy call (−25.1 pts)2026-10 · 966 calls · hit 90.8% vs 89.5% lazy call (+1.3 pts)1 of 12 months beat the lazy call

Lookalike Setups

Past moments when coins sat in the same spot (trend, distance from the high, Bitcoin's mood, unlocks ahead) and what came next. The main prediction.

FAILED SO FAR
calls44,735
hit rate61.7%vs 62.7% always-down −1.0
skill−1.2%
luck test−1.5σp 0.931
2025-11 · 1759 calls · hit 80.8% vs 80.8% lazy call (−0.0 pts)2025-12 · 3599 calls · hit 49.6% vs 50.4% lazy call (−0.8 pts)2026-01 · 2623 calls · hit 89.5% vs 90.4% lazy call (−1.0 pts)2026-02 · 3843 calls · hit 65.5% vs 65.7% lazy call (−0.1 pts)2026-03 · 4857 calls · hit 51.5% vs 51.9% lazy call (−0.3 pts)2026-04 · 3888 calls · hit 43.5% vs 56.6% lazy call (−13.1 pts)2026-05 · 4935 calls · hit 84.5% vs 84.5% lazy call (−0.1 pts)2026-06 · 4031 calls · hit 69.6% vs 69.4% lazy call (+0.1 pts)2026-07 · 4062 calls · hit 63.9% vs 63.8% lazy call (+0.1 pts)2026-08 · 5048 calls · hit 31.8% vs 68.5% lazy call (−36.7 pts)2026-09 · 3648 calls · hit 53.3% vs 55.0% lazy call (−1.7 pts)2026-10 · 2442 calls · hit 90.0% vs 89.8% lazy call (+0.2 pts)3 of 12 months beat the lazy call

The method · one picture

How a finding earns PROVEN

94,034TESTS LASTNIGHT4,049LIKELYLIKELYweaker evidence: watch themTRADINGRULES217DRIVERS AND SERIES14RULES14FADED RULESEVERY COIN ×EVERY KIND OF EVENTand every series we collectTESTED ONITS OWN HISTORYagainst its moves on any dayONE FALSE-DISCOVERYCUT, EVERY TEST10% for drivers and series,5% for trading rulesBOTH HALVES OF ITSHISTORY MUST AGREEthe older and the newer, and12 cases or moreTRADING RULES ONLY:YEARS IT NEVER SAWit must also earn on thenewest yearsPROVENpast every checkITS LIVE RECORDKEEPS COUNTING7 kinds of event movetonight's predictionsRETIREDwhen it stops holding94,034TESTS LAST NIGHTTRADING RULES14RULES14FADED RULES217 DRIVERS AND SERIESLIKELYweaker evidence: watch them4,049EVERY COIN ×EVERY KIND OF EVENTand every series we collectTESTED ONITS OWN HISTORYagainst its moves on any dayONE FALSE-DISCOVERYCUT, EVERY TEST10% for drivers and series, 5% for trading rulesBOTH HALVES OF ITSHISTORY MUST AGREEthe older and the newer, and 12 cases or moreTRADING RULES ONLY:YEARS IT NEVER SAWit must also earn on the newest yearsPROVENpast every checkITS LIVE RECORDKEEPS COUNTING7 kinds of event move tonight's predictionsRETIREDwhen it stops holding
The night of : 94,034 tests (84,283 coin-driver, 6,071 series, 3,680 trading rules); 217 proven, 4,049 likely; 14 trading rules active, 14 faded. What it found, coin by coin →

The ledger

every test the lab has run, night by night, kept by the lab itself

Tests run

6,868,025

since Oct 6, 2026, over 44 nights

Different ideas

172,455

each coin, series, rule and kind of event tested, counted once

Holding now

241

proven drivers and series, surviving rules, kinds of event moving the odds

Ever proven

378

218 stopped holding since

NightCoin driversSeriesTrading rulesOdds moversScoreboardTimingAll testsNew ideasStopped holding
159,922
216 held
6,071
1 held
3,680
14 survived
146
10 held
120
10 held
· 169,939 63 1
159,865
216 held
6,068
1 held
3,680
13 survived
146
10 held
120
10 held
· 169,879 220 6
159,645
218 held
6,069
1 held
3,680
13 survived
147
11 held
120
10 held
· 169,661 252 6
159,388
216 held
6,074
1 held
3,680
13 survived
146
11 held
120
11 held
· 169,408 217 3
159,175
212 held
6,070
1 held
3,680
11 survived
146
9 held
120
11 held
346 169,537 645 25
158,534
227 held
6,069
1 held
3,680
15 survived
146
9 held
120
11 held
· 168,549 884 2
157,655
226 held
6,066
1 held
3,680
15 survived
146
9 held
120
11 held
· 167,667 128 0
157,450
225 held
6,068
1 held
3,680
15 survived
146
9 held
120
11 held
· 167,464 559 3
156,903
224 held
6,061
1 held
3,680
18 survived
148
7 held
120
11 held
· 166,912 359 2
156,565
221 held
6,056
1 held
3,680
17 survived
148
7 held
120
11 held
· 166,569 269 1
156,295
220 held
6,052
1 held
3,680
17 survived
146
7 held
120
11 held
346 166,639 395 3
155,913
217 held
6,056
1 held
3,680
17 survived
146
9 held
120
12 held
· 165,915 553 5
155,444
217 held
6,051
1 held
3,680
17 survived
146
7 held
120
12 held
· 165,441 138 4
155,311
217 held
6,047
1 held
3,680
18 survived
146
7 held
120
12 held
· 165,304 515 5

A test is one idea checked against its history on one night: a coin after one kind of event over one horizon, a collected series against Bitcoin, Ether, the altcoins or its own coin, a whole trading rule, a kind of event moving the odds, a kind of event against ordinary days, a calendar or market-timing rule (Sundays). "Held" passed that night's bar.

The self-running lab

patterns tested live from the night they were found · promoted and retired by fixed rules, not by hand

7 patterns0 on trial7 testing live0 retired

PatternStatusLive tradesLive, a tradetOn history, a tradeLive since
Bitcoin beats the altcoins on Mondays (a 1-day hold) TESTING LIVE 5 −0.34% -0.6 +0.48% (436)
Bitcoin beats the altcoins in the six days before a Fed decision (a 6-day hold) TESTING LIVE 1 −1.73% · +2.59% (61)
The altcoins beat Bitcoin on Fridays and Saturdays (Thursday's close to Saturday's), after Bitcoin rose over 30 days (a 2-day hold) TESTING LIVE 1 −2.73% · +0.93% (237)
The altcoins beat Bitcoin on Fridays and Saturdays (Thursday's close to Saturday's) (a 2-day hold) TESTING LIVE 5 +0.38% +0.3 +0.59% (435)
Bitcoin: the side of its last 4 weeks, held a week TESTING LIVE 4 +4.25% +1.3 +1.24% (473)
Ether: the side of its last 4 weeks, held a week TESTING LIVE 4 +6.48% +2.5 +1.62% (473)
The altcoins against Bitcoin: the side of their last week, held a week TESTING LIVE 4 −1.69% -1.3 +0.65% (434)

The calls, live

CallsCallsDaysRightThat side, any coinEdge a calltVerdict
our calls · calls · 1d 8,36245 59%59% +0.00% +0.0 too few yet
our calls · lean · 1d 7,19945 58%59% −0.20% -1.0 too few yet
our calls · strong · 1d 1,16344 65%61% +1.67% +3.0 too few yet
our calls · calls · 30d 9,48417 92%95% +0.04% +0.1 too few yet
our calls · lean · 30d 5,36317 92%95% −0.52% -3.3 too few yet
our calls · strong · 30d 4,12117 92%95% +0.30% +0.7 too few yet
our calls · calls · 7d 23,57340 80%83% −0.49% -1.0 too few yet
our calls · lean · 7d 12,56540 79%81% −0.07% -0.1 too few yet
our calls · strong · 7d 11,00840 82%84% −1.55% -1.6 too few yet
all our data combined · abs1 calls · 1d 6,87135 63%61% +0.42% +1.9 too few yet
all our data combined · abs1 lean · 1d 5,44035 62%60% +0.19% +0.8 too few yet
all our data combined · abs1 strong · 1d 1,43135 65%61% +1.25% +2.2 too few yet
all our data combined · abs30 calls · 30d 5,55010 91%95% −28.69% -2.6 too few yet
all our data combined · abs30 lean · 30d 2,2757 93%95% −2.87% -21.3 too few yet
all our data combined · abs30 strong · 30d 3,27510 90%95% −28.21% -2.5 too few yet
all our data combined · abs7 calls · 7d 17,16531 81%83% −0.13% -0.2 too few yet
all our data combined · abs7 lean · 7d 13,87431 82%83% −0.09% -0.2 too few yet
all our data combined · abs7 strong · 7d 3,29130 77%83% +0.02% +0.0 too few yet
challenger · engine trend · 1d 1,30027 54%54% +0.72% +1.5 too few yet
challenger · reversal · 1d 71636 51%50% −0.69% -0.9 too few yet
challenger · crash · 7d 4621 67%84% +0.76% +0.2 too few yet
challenger · engine trend · 7d 15514 6%14% −0.56% -0.6 too few yet
challenger · interact · 7d 57929 57%50% +5.10% +6.9 too few yet
challenger · momentum · 7d 59230 52%50% −2.69% -1.6 too few yet
challenger · rsi · 7d 13025 66%64% −15.55% -1.7 too few yet

Each Sunday the lab searches every calendar and market-timing block (weekdays, the weekend, the month turn, Fed and expiry weeks, Bitcoin's trend, each alone and in each market) in one false-discovery cut, then holds what survives to its live record: on trial after 20 or more live trades whose average beat its cost (0.10% a trade, 0.20% for a pair) at a t of 1.645 or more; retired when it goes the other way, stops beating its cost or passes nothing in 40 trades. The calls: the last 90 days, each day counted once; a kind of call that lost live (100+ calls on 60+ days) is taken off the site, one that earned it live goes on only with the owner's OK. Live numbers are before costs.

Out-of-sample checks

older cases vs newer ones: does the edge survive data it wasn't found on?

Setups all setups →

Event patterns all patterns →

619 checked374 held245 flipped

share of cases higher the next day, older cases → newer cases · last column = cases

Strategy graveyard

ideas that failed or got replaced, public on purpose

R.I.P.FAILED SO FAR

Bitcoin Shadow

triedBitcoin's own prediction, scaled by this coin's beta: its average move for every percent Bitcoin moved.

died of1-day calls right 54.2% of 8,362 vs 51.3% for the lazy call, skill −0.1%; 7-day calls right 62.3% of 24,100 vs 60.8% for the lazy call, skill −0.3%; 30-day calls right 59.4% of 32,810 vs 62.7% for the lazy call, skill −0.4%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Breakout (30-day high / low)

triedPrice breaking out above its recent range, or down below it, and whether the move kept going.

died of7-day calls right 37.1% of 329 vs 65.4% for the lazy call, skill 0.0%; 30-day calls right 62.5% of 8,969 vs 63.1% for the lazy call, skill 0.0%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Capitulation Bounce

triedA brutal flush in a few days. Does the bounce come?

died of1-day calls right 63.3% of 412 vs 50.3% for the lazy call, skill −0.6%; 7-day calls right 32.6% of 408 vs 59.2% for the lazy call, skill −0.0%; 30-day calls right 67.1% of 1,224 vs 67.1% for the lazy call, skill 0.0%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Deep Drawdown

triedCoins sitting far below their high. Bottom-fishing, counted.

died of7-day calls right 59.4% of 27,329 vs 61.2% for the lazy call, skill 0.0%; 30-day calls right 62.3% of 33,030 vs 62.3% for the lazy call, skill −0.0%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Event Pattern

triedPast unlocks, listings and delistings lined up the same number of days out, and what price did after.

died of1-day calls right 50.2% of 767 vs 50.9% for the lazy call, skill −0.7%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Golden / Death Cross

triedThe fast average crossing the slow one, golden or death cross, and what followed.

died of7-day calls right 64.8% of 3,543 vs 64.8% for the lazy call, skill −0.2%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Green / Red Streak

triedLong runs of green or red daily closes in a row, and what the next leg did.

died of30-day calls right 66.8% of 2,215 vs 65.1% for the lazy call, skill 0.0%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Last Year This Time

triedThe same weeks in earlier years. Seasonality for fun, never counted in the vote.

died of1-day calls right 48.2% of 28,516 vs 52.0% for the lazy call, skill −0.6%; 7-day calls right 54.0% of 27,678 vs 61.6% for the lazy call, skill −0.3%; 30-day calls right 56.4% of 25,503 vs 62.2% for the lazy call, skill −0.2%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Lookalike Setups

triedPast moments when coins sat in the same spot (trend, distance from the high, Bitcoin's mood, unlocks ahead) and what came next. The main prediction.

died of1-day calls right 53.4% of 10,481 vs 51.2% for the lazy call, skill −0.1%; 7-day calls right 59.9% of 36,004 vs 60.8% for the lazy call, skill −0.3%; 30-day calls right 61.7% of 44,735 vs 62.7% for the lazy call, skill −1.2%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Market Breadth

triedHow many coins trade above their own trend at once: the market's mood, read for every coin.

died of7-day calls right 61.7% of 27,735 vs 63.6% for the lazy call, skill −0.1%; 30-day calls right 60.4% of 29,358 vs 62.8% for the lazy call, skill −0.7%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Rubber Band (30-day stretch)

triedHow far price has stretched from its recent average, and what coins stretched the same way did next.

died of7-day calls right 59.9% of 30,568 vs 60.8% for the lazy call, skill −0.2%; 30-day calls right 61.8% of 44,768 vs 62.7% for the lazy call, skill −0.1%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Trend Line (90-day average)

triedAbove or below the long trend line, with the line rising or falling: what coins in that state did next.

died of7-day calls right 60.1% of 42,645 vs 60.8% for the lazy call, skill −0.0%; 30-day calls right 62.7% of 47,316 vs 62.7% for the lazy call, skill 0.0%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED SO FAR

Volatility Squeeze

triedThe tightest range in a long while, and which way it broke.

died of7-day calls right 58.6% of 9,044 vs 58.6% for the lazy call, skill −0.0%; 30-day calls right 59.1% of 8,834 vs 59.1% for the lazy call, skill −0.5%.

nowstill scored every week in the open, never counted in a coin's pattern vote.

Nov 15, 2026real data

R.I.P.FAILED

Planted next-day breakout pop

triedTo give the new 1-day odds something real to find, add a push the day after a breakout in the demo generator: half a typical daily move up after a close above the 30-day high, down after a close below the 30-day low.

died ofAcross the demo's 2.5 years of prices the push showed: 65% of breakouts closed higher the next day and 36% of breakdowns, against 52% for all coin-days. In the scored year it did not: its 150 breakouts closed higher 52% of the time, the same as all coins, and the 1-day record sat 0.04 standard errors from zero. A smaller push (0.3 of a daily move) also knocked the 30-day records of Breakout, Market Breadth and Rubber Band below proven.

nowRemoved. The demo keeps its two planted effects (unlock drift and rubber band) with unchanged prices.

Oct 1, 2026synthetic test with planted effects

R.I.P.RETIRED

One score per coin (the Slate)

triedRank every coin by one composite score built from setup, risk, record, dilution and pulse.

died ofRetired as a separate board when the site went prediction-first.

nowThe old board redirects to Predictions, where coins are ranked by their chance of ending higher.

Oct 1, 2026product decision

R.I.P.REPLACED

Trade replay on the closest matches only

triedReplay a leveraged trade only on the few past cases closest to the coin's situation.

died ofIts chance of profit and the prediction above it disagreed by 10 points on JUP (59% vs 49%).

nowThe calculator replays the same weighted pool of cases the prediction uses, so the two agree (59.2% vs 59.1%).

Oct 1, 2026demo data

R.I.P.REPLACED

One shrink factor per pattern

triedPull each pattern's chance toward the all-coin up-rate by a single factor measured on its own past calls.

died ofIn a test market with a planted unlock effect (41% chance higher over 30 days) it flattened the unlock pattern to 51%: each event line rests on only about 13 cases, so the noise dragged the factor down.

nowEvent patterns are calibrated in groups (event kind and size); the same test now reads 43%.

Oct 1, 2026synthetic test with planted effects

R.I.P.REPLACED

Raw pattern chances

triedShow each pattern's raw chance of ending higher next to its line.

died ofThe 90-day trend pattern said 78% higher while its 30-day calls had been right 49.5% of the time.

nowEvery chance is calibrated against the pattern's own record, and a pattern without a proven record shows no number at all.

Oct 1, 2026demo data

Method log

every change to how we count, newest first

  1. patternsCHANGEDsynthetic test with planted effects

    Trust bar on a market with nothing planted

    tried
    Score all fifteen patterns at 1, 7 and 30 days (45 records) on a synthetic 600-coin market of random walks with a shared Bitcoin factor and market-wide drift spells, where no pattern effect was planted.
    result
    With one 2-standard-error test per record, 1 of the 45 passed by luck: Volatility Squeeze at 30 days (1,227 calls, 2.02 standard errors), and it then cast the only vote on 25 coins.
    now
    Proven now also has to survive a fluke filter across every check in the build (Benjamini-Hochberg at a 10% false discovery rate over all 45 records, with 2 standard errors kept as a floor): at most about 1 in 10 proven patterns is a fluke. On the 600-coin market nothing passes: a pattern with no edge posts a record as strong as Volatility Squeeze's 2.2% of the time, and the best of 45 checks has to be under 0.22%; all 600 coins read 'no proven pattern'. On the demo market 5 of 45 records stay proven (Lookalike, the event pattern and Rubber Band at 30 days, Rubber Band at 7 days, and Breakout at 30 days at 1.6% against a 1.8% cutoff); Market Breadth (2.0 and 2.05 standard errors) falls back to on trial.
  2. patternsKEPTdemo data

    Cases counted once per episode

    tried
    Count a pattern's past cases once per episode, on the first day its condition turns on, instead of once for every day it stays on.
    result
    On the demo market the ten market patterns were on for 32,236 coin-days in 4,221 separate episodes. A 50/200-day cross stays on for 20 days (1,462 coin-days, 77 crosses) and the 90-day trend state for weeks (14,750 coin-days, 865 episodes), so counting days would have counted the same move many times.
    now
    The cases behind each line and in the playground are episodes; a condition that switches off for less than 5 days and back on is the same episode. Each pattern keeps its 5,000 most recent cases.
  3. patternsCHANGEDdemo data

    One vote per horizon

    tried
    Let each pattern vote at 1, 7 and 30 days only where its own record at that horizon is proven, instead of one 30-day vote.
    result
    On the demo market 16 of 21 coins lean higher at 30 days and 5 come out even. At 7 days only Rubber Band has earned a vote: 1 coin leans higher, 20 come out even. At 1 day no pattern has, so all 21 read 'no proven pattern'.
    now
    The vote is stored per horizon. A pattern proven at 30 days has no say at 1 day until its 1-day record earns it.
  4. patternsFAILEDsynthetic test with planted effects

    Planted next-day breakout pop

    tried
    To give the new 1-day odds something real to find, add a push the day after a breakout in the demo generator: half a typical daily move up after a close above the 30-day high, down after a close below the 30-day low.
    result
    Across the demo's 2.5 years of prices the push showed: 65% of breakouts closed higher the next day and 36% of breakdowns, against 52% for all coin-days. In the scored year it did not: its 150 breakouts closed higher 52% of the time, the same as all coins, and the 1-day record sat 0.04 standard errors from zero. A smaller push (0.3 of a daily move) also knocked the 30-day records of Breakout, Market Breadth and Rubber Band below proven.
    now
    Removed. The demo keeps its two planted effects (unlock drift and rubber band) with unchanged prices.
  5. patternsKEPTdemo data

    Eight trader patterns added

    tried
    Add eight patterns traders already know: golden / death cross, breakout of the 30-day high or low, capitulation (down 25% in a week), green / red streaks, fresh all-time highs, deep drawdowns, volatility squeezes and market breadth. Nothing was planted for any of them in the demo.
    result
    Breakout earned proven at 30 days on the demo (168 calls, 70% right, 2.2 standard errors) and still clears the stricter fluke filter adopted the same day. It rides the demo's planted rubber band: breakouts above the 30-day high ended higher 30 days later only 34% of the time and breakdowns 75%, against 57% for all coins, and most breakouts are stretched prices. Market Breadth beat the up-rate too (248 calls, 2.0 standard errors; 2.05 at 7 days) but not by enough to survive 45 checks, so it is on trial. The other six stay on trial with 10 to 86 scored 30-day calls each, under the 100 needed.
    now
    Fifteen patterns are scored the same way at 1, 7 and 30 days. A market pattern draws a line only on the days its condition is on.
  6. patternsKEPTdemo data

    Next-day odds for every pattern

    tried
    Give every pattern a 1-day chance next to its 7- and 30-day ones, scored on its own 1-day calls and held to the same bar: beat the all-coin up-rate on calls it never saw, by more than luck.
    result
    No pattern cleared the bar at 1 day on the demo market. The closest stayed inside luck: Breakout (116 calls, 1.3 standard errors above zero), the event pattern (256 calls, 1.2) and Rubber Band (180 calls, 1.2); Lookalike made only 13 one-day calls. A single day's move is mostly the whole market's move.
    now
    Every line shows its 1-day chance and its own 1-day record. At 1 day every pattern is on trial, and the 1-day vote reads 'no proven pattern' on all 21 demo coins.
  7. scoringRETIREDproduct decision

    One score per coin (the Slate)

    tried
    Rank every coin by one composite score built from setup, risk, record, dilution and pulse.
    result
    Retired as a separate board when the site went prediction-first.
    now
    The old board redirects to Predictions, where coins are ranked by their chance of ending higher.
  8. calculatorREPLACEDdemo data

    Trade replay on the closest matches only

    tried
    Replay a leveraged trade only on the few past cases closest to the coin's situation.
    result
    Its chance of profit and the prediction above it disagreed by 10 points on JUP (59% vs 49%).
    now
    The calculator replays the same weighted pool of cases the prediction uses, so the two agree (59.2% vs 59.1%).
  9. patternsCHANGEDsynthetic test with planted effects

    Proven = beats the up-rate

    tried
    Mark a pattern proven once its calibrated chances beat the simple all-coin up-rate over 100 or more calls.
    result
    Two patterns with no real effect in the test market (Last Year This Time, Bitcoin Shadow) passed by luck: 20 coins that move together over one year give only about 12 independent months at 30 days.
    now
    Proven also needs the weekly gains to sit at least 2 standard errors above zero; both noise patterns now stay on trial.
  10. patternsREPLACEDsynthetic test with planted effects

    One shrink factor per pattern

    tried
    Pull each pattern's chance toward the all-coin up-rate by a single factor measured on its own past calls.
    result
    In a test market with a planted unlock effect (41% chance higher over 30 days) it flattened the unlock pattern to 51%: each event line rests on only about 13 cases, so the noise dragged the factor down.
    now
    Event patterns are calibrated in groups (event kind and size); the same test now reads 43%.
  11. patternsREPLACEDdemo data

    Raw pattern chances

    tried
    Show each pattern's raw chance of ending higher next to its line.
    result
    The 90-day trend pattern said 78% higher while its 30-day calls had been right 49.5% of the time.
    now
    Every chance is calibrated against the pattern's own record, and a pattern without a proven record shows no number at all.

Prediction from past patterns, not financial advice. Records are walk-forward backtests on our own price history; past patterns can stop working.

More: Research notes, week by week · Our track record · Predictions · Setups · Every event pattern · Crypto calendar · Methodology