Is there a day-of-the-week effect in bitcoin?
Not in our data. Between August 2017 and August 2026 bitcoin averaged +0.33% on Mondays and +0.35% on Wednesdays against −0.34% on Thursdays, but in no market era is the spread between days larger than the noise you get when seven weekdays are compared at once. Rolling weekday averages, as shown on dashboards, had no forecasting value.
The claim
Caporale and Plastun tested four large cryptocurrencies on data through 2017 with t-tests, ANOVA, dummy-variable regressions and a trading simulation. Most coins showed no weekday pattern; bitcoin was the exception, with Monday returns significantly higher than on other days. The same idea lives on in the weekday heatmaps of crypto dashboards, usually computed over the last few weeks or months.
What we tested
BTC and ETH spot, daily returns from midnight to midnight UTC, August 2017 to August 2026, in three eras: before 2020, from 2020 to the launch of US spot bitcoin ETFs in January 2024, and after. In each era we compared every weekday with the rest and judged the most extreme day against the same returns shifted in time so that the calendar no longer lines up — a fair yardstick when seven days are searched at once. We also tested the dashboard version: does a weekday’s average over the past 4 to 52 weeks predict that weekday next time?
Result
No weekday effect survives in any era. The brightest cells — bitcoin on Wednesdays, ether on Thursdays — are about as extreme as the most extreme of seven random averages usually is. Rolling weekday averages had no forecasting value: a rule that traded their sign lost 0.09–0.11% a day before costs.
The arithmetic explains why. With daily volatility near 4%, nine years of data can reliably detect a weekday premium of about 0.4% a day; a three-month rolling window needs about 1% a day — more than 50% a year earned on a single weekday.
Average daily return by weekday (UTC, log returns), August 2017 – August 2026
| Weekday | BTC | ETH |
|---|---|---|
| Monday | +0.33% | +0.20% |
| Tuesday | −0.06% | −0.15% |
| Wednesday | +0.35% | +0.30% |
| Thursday | −0.34% | −0.54% |
| Friday | +0.19% | +0.22% |
| Saturday | +0.11% | +0.30% |
| Sunday | +0.02% | +0.11% |
| All days | +0.09% | +0.06% |
In every era, the gaps between days are within the range produced by chance.
What it means
A weekday chart shows what happened, not what will. Daily crypto returns are so noisy that the gap between the best and worst day of a short window is almost entirely sampling error — and picking the window that looks best makes it worse.
What would change the verdict
One descriptive pattern is worth watching: Thursday was the weakest day for both coins in each of the three eras. It was noticed after the fact, so on its own it proves nothing; a rule fixed today and judged only on data from here on could settle it. A mechanism tied to a specific weekday — a recurring flow with a known date — would also justify reopening the question.
This note re-tests a claim made by the cited authors on our own crypto market data; the summary of their method is ours. Tested · published · last reviewed . Research, not investment advice.