Uptober and crypto seasonality: Bitcoin by month and weekday

October has been good, but so have other months

“Uptober” rests on a real record: Bitcoin rose in 10 of 15 Octobers from 2011 to 2025, with a median gain of 11.2%. But February (11 of 16, median +11.7%) and April (11 of 16, +10.0%) look as good, while August and September were the weakest months. October 2025 fell 3.9%.

With only 15 or 16 observations per month and monthly swings of tens of percent, these differences are within what chance produces. A Kruskal–Wallis test of whether the 12 months share one distribution gives p = 0.44. When we shuffled all monthly returns across calendar months 5,000 times, the best month’s median beat October’s actual median 89% of the time. Some month will always look like the winner.

Mean and median tell different stories

November’s average return of +36.1% is the highest, but its median is +8.9%: a few explosive Novembers, such as 2013, carry the average. December’s average is +7.9% while its median is −3.2%, and only 7 of 15 Decembers rose, so the “Santa rally” claim is weak in this data. Averages are pulled by rare extreme months; medians and counts of positive years describe the typical outcome better.

Day of the week: quieter weekends, no reliable direction

Daily returns by weekday, from 2011 to May 2026, do not differ significantly (Kruskal–Wallis p = 0.11; since 2022, p = 0.20). Thursday has had the lowest median since 2022, at −0.32%, but that is a small difference among daily moves that typically span several percent.

The clear weekday effect is volatility. Since 2022 the standard deviation of daily returns was 1.75% on weekends against 2.98% on weekdays, consistent with traditional markets being closed and less macro news arriving. Quieter does not mean safer: weekend liquidity is also thinner.

Weekend altcoin rallies faded

A related claim says altcoins outperform Bitcoin at weekends. Using an equal-weighted basket of 40 liquid altcoins on OKX spot from April 2022 to July 2026, we compared the basket’s return minus BTC’s over Saturday and Sunday with the same 48-hour measure on weekdays. On average the basket trailed BTC by 0.38 percentage points per weekend and 0.47 points per weekday window, a difference indistinguishable from zero (t 0.34).

The pattern did exist for a while: in 2023 and 2024 the basket roughly kept pace with BTC on weekends (+0.08 and +0.10 points) while trailing it on weekdays (−0.67 points both years). It then reversed: in 2025 weekends were worse (−0.68 against −0.32 points), and in 2026 to July −1.28 against +0.13 points.

How the numbers were calculated

Monthly and daily Bitcoin returns use Coin Metrics’ PriceUSD, a UTC day-end reference rate, from 31 December 2010. Monthly returns compare consecutive month-end closes; the last complete month in the snapshot is April 2026. Weekday labels are UTC. The shuffle test permutes all monthly returns across month labels and records the largest of the 12 medians in each draw.

The altcoin comparison uses confirmed OKX spot USDT hourly candles for coins with at least 30,000 hours of history and a median hourly quote volume of at least 50,000 USDT, excluding stablecoins and tokenised gold. Windows are 48 hours between closes of the 00:00 UTC hourly candle: Saturday to Monday for weekends, and windows ending Wednesday, Thursday or Friday for weekdays. Log returns are averaged equally across coins.

Is Uptober real for Bitcoin?

October has been one of Bitcoin’s better months, up in 10 of 15 years, but February and April have similar records and the differences between months are not statistically significant. A strong historical October is not a reliable forecast for the next one.

Do altcoins pump on weekends?

Not consistently. In 2023–2024 a basket of 40 altcoins fared better against BTC on weekends than on weekdays, but the pattern reversed in 2025–2026, and over the full period weekends and weekdays were indistinguishable.

Limitations

One asset for the monthly and weekday tests, 15–16 observations per calendar month, and surviving altcoins on a single venue from 2022. UTC calendar labels are a convention. No fees, spreads or account-level backtest are included; seasonal patterns are not forecasts or trading advice.

Sources and data access

• Coin Metrics: public BTC daily data
• Coin Metrics: PriceUSD definition and timestamp convention
• OKX market data documentation

Cited sources document external facts or data formats; they do not endorse our findings. The public snapshot contains the table and methods. Source-file fingerprints are included for recalculated datasets. Full archived raw files are not distributed here; fingerprints identify the files used, rather than independently proving the results.

Aggregate data snapshot (JSON) · Markdown · Methods and limitations

Live prior 30-day hourly baseline — separate from the historical study

Only the current watch list is shown, up to 12 assets. Missing or stale estimates remain unavailable. Venue and quote currency are kept separate; beta needs matching BTC data. Hourly statistics are not minute-move probabilities.

Observation cutoff (UTC): 24/10/2026, 13:46

Related research