Lab notes · week 41 of 2026 · closed
Lab notes: the week of Oct 5 – 11, 2026
The lab ran 812,133 tests over 6 nights, Oct 6 to Oct 11. 147,500 of them were ideas it had never tested before. 223 findings held at the end of the week: proven coin drivers and series, surviving trading rules and kinds of event that move the odds. 236 were proven for the first time (the ledger's first week: every finding was new to it), 207 still at its end. 27 that had been proven stopped holding. Bitcoin fell 3.2% over the week; 72 of the top 217 coins closed it higher. Past patterns, not financial advice.
Tests run
812,133
over 6 nights
New ideas
147,500
tested for the first time
Holding
223
at the week's end
First proven
236
this week
Stopped holding
27
proven before, not now
§ 01 · Night by night6 nights
Night by night
as of · the lab's ledger, kept by each stage itself
tests run that night, the latest litblue under a day: findings holding after it· no run that night
| Night | Coin drivers | Series | Trading rules | Odds movers | Scoreboard | All tests | New ideas | Stopped holding |
|---|---|---|---|---|---|---|---|---|
| 103,205 221 held | 5,857 1 held | 3,144 8 survived | 115 6 held | 105 13 held | 112,426 | 113,495 | 0 | |
| 111,248 223 held | 5,857 1 held | 3,336 7 survived | 124 7 held | 105 13 held | 120,670 | 8,260 | 8 | |
| 134,440 202 held | 5,848 1 held | 3,336 8 survived | 145 6 held | 120 15 held | 143,889 | 23,251 | 39 | |
| 134,927 206 held | 5,875 1 held | 3,336 7 survived | 145 6 held | 120 15 held | 144,403 | 514 | 3 | |
| 135,508 209 held | 5,899 1 held | 3,336 7 survived | 145 7 held | 120 15 held | 145,008 | 627 | 2 | |
| 136,209 207 held | 5,925 1 held | 3,338 9 survived | 145 6 held | 120 15 held | 145,737 | 1,353 | 5 |
By the end of the week, the ledger held 812,133 tests over 6 nights since Oct 6, 2026, 147,500 different ideas among them.
Source: the lab's ledger, every idea each stage tested, night by night · stored
§ 02 · First proven236 new
First proven this week
as of · coin drivers and whole-market findings
236 findings passed the bar for the first time this week: the ledger's first week, so every finding it holds is new to it. The strongest 12:
-
PROVEN
XRP
When Ether fell 4% or more in a day, XRP ended higher the next day more often than on any day
345 cases · earned out of sample: right 54% of 5,799 calls since 2019, against 49% for that side on any day · first proven · its own pageShare on Xcopy link
-
PROVEN
XRP
When its sector fell 5% on average in a day, XRP ended higher the next day more often than on any day
288 cases · earned out of sample: right 57% of 1,678 calls since 2019, against 49% for that side on any day · first proven · its own pageShare on Xcopy link
-
PROVEN
ETC
When Ether fell 4% or more in a day, ETC ended higher the next day more often than on any day
337 cases · earned out of sample: right 54% of 5,799 calls since 2019, against 49% for that side on any day · first proven · its own pageShare on Xcopy link
-
PROVEN
BTT
When Bitcoin fell 3% or more in a day, BTT ended higher the next day more often than on any day
145 cases · no clear edge out of sample · first proven · its own pageShare on Xcopy link
-
PROVEN
HBAR
When Bitcoin fell 3% or more in a day, HBAR ended higher the next day more often than on any day
265 cases · no clear edge out of sample · first proven · its own pageShare on Xcopy link
-
PROVEN
IOTA
When its sector fell 5% on average in a day, IOTA ended higher the next day more often than on any day
271 cases · earned out of sample: right 57% of 1,678 calls since 2019, against 49% for that side on any day · first proven · its own pageShare on Xcopy link
-
PROVEN
HBAR
When its sector fell 5% on average in a day, HBAR ended higher the next day more often than on any day
220 cases · earned out of sample: right 57% of 1,678 calls since 2019, against 49% for that side on any day · first proven · its own pageShare on Xcopy link
-
PROVEN
IOTA
When Bitcoin fell 3% or more in a day, IOTA ended higher the next day more often than on any day
331 cases · no clear edge out of sample · first proven · its own pageShare on Xcopy link
-
PROVEN
XMR
When Bitcoin rose 3% or more in a day, XMR ended lower the next day more often than on any day
261 cases · no clear edge out of sample · first proven · not proven any more · what moves XMR
-
PROVEN
MATIC
When Bitcoin fell 3% or more in a day, MATIC ended higher the next day more often than on any day
240 cases · no clear edge out of sample · first proven · its own pageShare on Xcopy link
-
PROVEN
XLM
When Ether fell 4% or more in a day, XLM ended higher the next day more often than on any day
339 cases · earned out of sample: right 54% of 5,799 calls since 2019, against 49% for that side on any day · first proven · its own pageShare on Xcopy link
-
PROVEN
ETC
When its sector fell 5% on average in a day, ETC ended higher the next day more often than on any day
274 cases · earned out of sample: right 57% of 1,678 calls since 2019, against 49% for that side on any day · first proven · its own pageShare on Xcopy link
§ 03 · Stopped holding27 stopped
Stopped holding
as of · proven when the week began, not past the bar now
27 findings that had been proven no longer passed the bar at the week's end. A finding that stops holding stays in the record; it no longer counts toward a coin's odds.
-
STOPPED
AAVE
AAVE, the day after Ether fell 4% or more in a day proven when the week began, not past the bar now · its own pageShare on Xcopy link
-
STOPPED
ALGO
ALGO, the day after Ether fell 4% or more in a day proven when the week began, not past the bar now · what moves ALGO
-
STOPPED
BEAM
BEAM, the day after it closed back above its 200-day average proven when the week began, not past the bar now · what moves BEAM
-
STOPPED
BEAM
BEAM, the day after a big S&P 500 down day came proven when the week began, not past the bar now · what moves BEAM
-
STOPPED
CTSI
CTSI, a week after it rose three times its usual daily move proven when the week began, not past the bar now · what moves CTSI
-
STOPPED
DOCK
DOCK, the day after it traded five times its usual volume proven when the week began, not past the bar now · what moves DOCK
-
STOPPED
GAL
GAL, the day after a US CPI release came proven when the week began, not past the bar now · what moves GAL
-
STOPPED
GAL
GAL, the day after Ether fell 4% or more in a day proven when the week began, not past the bar now · what moves GAL
-
STOPPED
LSK
LSK, the day after it rose three times its usual daily move proven when the week began, not past the bar now · what moves LSK
-
STOPPED
MDT
MDT, the day after it rose 5% or more while Bitcoin fell 2% or more proven when the week began, not past the bar now · what moves MDT
-
STOPPED
NEAR
NEAR, a week after it rose three times its usual daily move proven when the week began, not past the bar now · its own pageShare on Xcopy link
-
STOPPED
NEO
NEO, the day after it rose three times its usual daily move proven when the week began, not past the bar now · what moves NEO
§ 04 · Fired this week5 fired
Proven drivers that fired
as of · the event happened this week, so its window opened
5 times this week, the event behind a proven driver happened. What followed is counted in that driver's own record.
-
PROVEN
OGN
When it traded five times its usual volume, OGN ended lower a week later more often than on any day
40 cases · no clear edge out of sample · happened · its window ends Oct 12, 2026 · what moves OGN
-
PROVEN
CKB
When it rose three times its usual daily move, CKB ended lower the next day more often than on any day
21 cases · no clear edge out of sample · happened · its window ends Oct 12, 2026 · its own pageShare on Xcopy link
-
PROVEN
NMR
When it rose three times its usual daily move, NMR ended lower a week later more often than on any day
28 cases · no clear edge out of sample · happened · its window ends Oct 13, 2026 · its own pageShare on Xcopy link
- PROVEN UOS When it rose three times its usual daily move, UOS ended lower a week later more often than on any day 16 cases · no clear edge out of sample · happened · its window ends Oct 13, 2026 · its own pageShare on Xcopy link
-
PROVEN
RAD
When it rose three times its usual daily move, RAD ended lower a week later more often than on any day
19 cases · no clear edge out of sample · happened · its window ends Oct 12, 2026 · its own pageShare on Xcopy link
§ 05 · The trading rules
The trading rules
as of · the nightly strategy search and its book
On its latest night of the week (Oct 11, 2026), the strategy search tested 3,338 versions of its trading rules; 9 passed both cuts.
9 rules in the book still pass; 9 entered it this week. Every rule that ever passed stays in the book with its paper record.
What the search keeps, and the closest misses it threw away →
§ 06 · The market
The market that week
as of · our daily closes, Sunday to Sunday
Bitcoin fell 3.2% from the Sunday before to Oct 11, 2026, Ether fell 7.0%. 72 of the top 217 coins with both closes ended the week higher. The week brought 1,586 dated events.
Our live next-day calls whose window closed that week: 389 of 674 right (58%); calling every coin lower, the way more than half went, was right 56% of the time over all 6354.
Our live 7-day calls whose window closed that week: 456 of 661 right (69%); calling every coin lower, the way more than half went, was right 69% of the time over all 1174.
The full weekly recap: every event and what prices did after it →
§ 07 · How to read them
How to read these notes
as of · the bars each lab holds a finding to
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 or against ordinary days.
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 thousands tested at once need far more, which is what the cuts below are for.
A coin driver or a series finding is proven when it passes a 10% false-discovery cut over every test of its night, rests on 12 cases or more and holds in both halves of its history. Likely ones pass a looser bar, at about the rate pure chance would give: leads to watch, not facts. A trading rule must pass a 5% false-discovery cut over every test at once on the older 60% of the history, then a luck test of 1.645 or more on the same side on the newer 40% it was not chosen on, every case counted.
This week is closed: its numbers were stored for the last time on Oct 12, 2026 and will not change.
§ 08 · More
More research
as of
Every week: week 47 · week 46 · week 45 · week 44 · week 43 · week 42 · week 41
Get the day's plays in your Telegram before they go out on X, tested like this one → 7 days of Pro free · no payment