Study · re-run every night · updated
Do weekdays, weekends and Fed weeks matter?
116 tests on Bitcoin, Ether, the 100 biggest altcoins and the altcoins against Bitcoin (Aug 17, 2017 to Oct 7, 2026): each day of the week, the weekend, the turn of the month, the quarter's last week, the week into the monthly options expiry, the six days before a Fed decision and 10 market regimes. 0 passed the first cut (with 116 tests at once, a luck test of about 3.5 was needed) and 0 also earned on the newer 40% of the history. The closest: the altcoins against Bitcoin did better than usual on Saturdays, by 0.33% on average (luck test 1.3 on the older part, 2.4 on the newer), not past the cut. Before fees. Past patterns, not financial advice.
Tests
116
21 states × 4 targets
First cut
0
passed on the older part
Held
0
also on the newer 40%
Luck test needed
3.5
with this many tests at once
§ 01 · The calendar
The calendar
as of · our daily closes, Aug 17, 2017 to Oct 7, 2026
Each bar is how much better (or worse) than an ordinary day a market did on that day of the week, over the whole history. A faint bar has a luck test under 2: it could easily be chance.
| When | Bitcoin | Altcoins | Alts vs Bitcoin |
|---|---|---|---|
| Mondays · 1 day | +0.27% t +1.5 | −0.10% t −0.4 | −0.39% t −2.6 |
| Tuesdays · 1 day | −0.16% t −1.0 | −0.23% t −1.2 | −0.09% t −0.7 |
| Wednesdays · 1 day | +0.26% t +1.5 | +0.14% t +0.6 | −0.14% t −0.9 |
| Thursdays · 1 day | −0.38% t −1.9 | −0.45% t −1.8 | −0.04% t −0.3 |
| Fridays · 1 day | +0.11% t +0.6 | +0.47% t +2.0 | +0.40% t +2.6 |
| Saturdays · 1 day | −0.01% t −0.1 | +0.27% t +1.7 | +0.33% t +2.5 |
| Sundays · 1 day | −0.09% t −0.7 | −0.11% t −0.6 | −0.07% t −0.7 |
| Weekends · 2 days | −0.10% t −0.7 | +0.16% t +0.7 | +0.26% t +1.6 |
| Turn of the month · 4 days | +0.61% t +1.0 | +0.40% t +0.5 | +0.30% t +0.5 |
| Options-expiry week · 6 days | −0.71% t −1.1 | −0.87% t −1.0 | −0.26% t −0.4 |
| Week before the Fed · 6 days | −0.77% t −0.6 | −2.86% t −2.0 | −2.09% t −2.3 |
§ 02 · What passed0 held
What passed, what came closest
as of · the strategy search's gate
116 tests, the calendar and 10 market regimes; with that many at once, a luck test of about 3.5 was needed to pass the first cut. 0 passed it and 0 also earned on the newer part.
Not one of the 116 tests passed both cuts.
The closest misses, not past the cut
| Test | Older t | Newer t | Against usual | Cases |
|---|---|---|---|---|
| The altcoins against Bitcoin did better than usual on Saturdays | 1.3 | 2.4 | +0.33% | 434 |
| The altcoins against Bitcoin did better than usual on Fridays | 2.3 | 1.4 | +0.40% | 434 |
| The altcoins against Bitcoin did worse than usual on Mondays | 2.2 | 1.3 | −0.39% | 435 |
| Bitcoin did worse than usual on Thursdays | 1.1 | 2.3 | −0.38% | 476 |
| Ether did worse than usual on Thursdays | 1.7 | 1.7 | −0.54% | 476 |
| The altcoins against Bitcoin did worse than usual in the six days before a Fed decision | 2.0 | 1.3 | −2.09% | 63 |
§ 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.
A test is one state (a day of the week, the weekend, the turn of the month, the quarter's last week, the week into the monthly options expiry, the six days before a Fed decision, or a market regime), one market and one hold. Its cases are the closes where the state is on, a run counted once per hold so no two overlap. Each case is measured against the same market's average move over the same years, so a rising market does not make every state look good.
It passes 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. The altcoins against Bitcoin is a long-short spread with four sides a trade; the others have two.
Our price history holds today's coins: coins that died before we began collecting are missing, which flatters the older years.
§ 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 6, 2026; its numbers last moved on Oct 6, 2026. Each night its headline numbers moved is listed here, newest first.
- 116 tests · 0 past the first cut · 0 held
§ 05 · More
More research
as of