Uptober 2026: Does Bitcoin's Best Month Still Work?

The short answer

"Uptober" is the crypto market's nickname for October, the month in which bitcoin has historically posted its strongest average returns. The label is earned, but thinner than it sounds: using monthly candles from Binance's public market data for BTCUSDT (accessed 22 September 2026), bitcoin closed October higher in seven of the last nine years, with an average gain of +18.1% and a median of +11.0%. Strip out the outlier 2017 bull mania and the average drops to +14.4%.

The two things most Uptober posts leave out: the streak broke last year, and the pattern is close to meaningless outside bitcoin itself. Bitcoin trades near $86,206 as of 22 September 2026 — roughly 32% below its October 2025 high of about $126,200 — so this October starts from a very different place than the last one.

Uptober in numbers: every October since 2017

Our own calculation, monthly open-to-close on Binance spot pairs. Percentages are simple monthly returns, not compounded.

October	BTC	ETH	SOL
2017	+47.6%	-0.1%	—
2018	-3.8%	-14.7%	—
2019	+10.3%	+0.7%	—
2020	+28.0%	+7.4%	-46.5%
2021	+39.9%	+42.9%	+43.2%
2022	+5.5%	+18.4%	-2.1%
2023	+28.5%	+8.6%	+79.7%
2024	+11.0%	-3.2%	+10.6%
2025	-3.9%	-7.2%	-10.3%
Positive years	7 of 9	5 of 9	3 of 6
Average / median	+18.1% / +11.0%	+5.9% / +0.7%	+12.4% / +4.2%

Source: Binance public market data (monthly klines for BTCUSDT, ETHUSDT and SOLUSDT), retrieved 22 September 2026.

Uptober is a bitcoin story, not a crypto story

This is the part the seasonal chatter flattens. Ether's median October return is +0.7% — statistically indistinguishable from nothing — and it fell in four of nine Octobers. Solana has only six Octobers of history and split them three-three, with the average dragged up entirely by a single +79.7% month in 2023 and a +43.2% month in 2021.

If you are trading altcoins on an "Uptober" thesis, you are borrowing bitcoin's statistic and applying it to assets that do not share it. Whether alts participate depends on rotation, not the calendar — the mechanics of that rotation are covered in our altcoin season index breakdown, and the fundamentals case for ETH specifically in our 2026 ethereum outlook.

What 2025 changed

October 2025 closed -3.9%, the first losing October since 2018, and November 2025 followed with -17.6%. That matters for two reasons. First, one bad year in a nine-year sample moves the numbers a lot — this is a small dataset, not a law. Second, it happened in the month bitcoin printed its all-time high near $126,200, which is the classic shape of a seasonal pattern breaking: the trade works until enough people are positioned for it.

Anyone quoting a clean Uptober win rate today is quoting a figure that is one year out of date.

Our read: how much is the seasonal edge really worth?

Run the arithmetic honestly. Nine October observations, average +18.1%, standard deviation about 18.7%. That is a t-statistic near 2.9 — nominally "significant" — but the month was not chosen at random. It was chosen because it was the best of twelve. Picking the strongest month out of twelve and then testing it makes a result like this far easier to hit by chance than the raw number suggests. With nine data points, the honest description is a weak positive tilt, not a tradeable edge on its own.

Two further checks we ran on the same data:

• September's mirror pattern has already failed. Bitcoin averaged -3.0% across the last nine Septembers, but the last three were all positive — and September 2026 is up roughly 9.7% month-to-date through 22 September. The "weak September, strong October" pairing has lost the first half of its own story.
• October strength does not carry into November. November closed higher in only four of the last nine years, with a -36.6% month in 2018 and -17.6% in 2025. Uptober, where it exists, is a one-month effect.

What would make us take the seasonal tilt more seriously this year: confirmation from flows and positioning rather than the calendar — sustained spot ETF creations, rising open interest with funding staying flat rather than spiking, and bitcoin reclaiming its 2025 range. What would invalidate it: another leverage flush of the kind that produced the November 2025 drawdown.

How to use seasonality without getting hurt by it

Seasonality is a context filter, not a signal. A useful way to hold it: if your own analysis already favours a long, a historically positive month is a mild reason to size normally rather than defensively. It is never a reason to enter a trade you would not otherwise take, and it is never a reason to widen a stop.

Concretely, that means the same risk process as any other month — defined invalidation level, position size set from the stop distance, and no leverage added because a statistic from 2017 says the month is friendly. If you want the longer-horizon framing for bitcoin rather than the monthly one, our bitcoin outlook for the end of 2026 sets out the scenarios we are working from, and traders who want setups called in real time can follow the crypto signals desk.

Trading crypto involves substantial risk of loss and the figures above are historical, not a forecast. Nothing here is investment advice. If you are setting up to trade the quarter, do it from a properly funded, risk-sized account — you can open a trading account when your plan is written, not before.