September has historically been one of Bitcoin’s weakest months. But for the fourth year in a row, it ended in positive territory.
Bitcoin gained 6.33% in September, according to CoinGlass data, following gains of 3.91% in 2023, 7.29% in 2024 and 5.16% in 2025. That is four straight positive Septembers after six consecutive losses between 2017 and 2022. CoinGlass data shows September has averaged a loss of about 2.4% since 2013.
The run did not go unnoticed, DeFi app, pointed to it in a post on X. “For years, ‘Red September’ was a rule in crypto,” it wrote, adding that the market now “trades on fundamentals, revenue, and real usage, not folklore.”
DeFi’s claim raises two questions: which of Bitcoin’s traditional rules are failing, and what should traders watch instead?

Bitcoin is breaking more than the September pattern
“Uptober,” the belief that October is reliably strong, also failed last year. October turned negative for the first time since 2018. The fourth quarter then fell 23.07%, despite an average fourth-quarter return of more than 77% in historical data.
Another rule has also fallen.
Before 2026, every green August in CoinGlass data since 2013 had been followed by a negative September. This year, August’s 24.95% gain was followed by another 6.33% rise in September. July, August and September also all ended higher for the first time in that data, giving Bitcoin a 42.71% gain for the third quarter.
The moves do not mean seasonal trends have disappeared. They do show why calendar-based patterns are harder to use as standalone signals.
Is the four-year cycle dead?
The bigger test is the four-year halving cycle. It rests on the idea that Bitcoin moves through roughly four-year periods around halving events, when the reward for mining new Bitcoin is cut in half.
The current cycle has behaved differently. Bitcoin reached its October 2025 peak roughly 18 months after the April 2024 halving, in line with earlier cycles, but the decline that followed was smaller than in previous cycles.
Some analysts say the old cycle has broken down.
Wintermute said in its 2026 crypto outlook that the “four-year cycle is dead,” arguing that structural factors such as exchange-traded funds (ETFs), digital asset treasuries and the movement of capital now matter more than timing.
Bitwise Chief Investment Officer Matt Hougan has made a similar argument, saying the forces behind the cycle, including the halving, interest-rate cycles and leverage, have weakened.
Others disagree. Galaxy Research argued in June that the cycle remains visible in the data, although the swings are shrinking. Each successive cycle has produced a smaller peak-to-trough decline.

The disagreement is less about whether Bitcoin has changed and more about how much the old cycle still explains.
What are analysts watching instead?
Many are watching where money is moving and how financial conditions are shifting. US spot Bitcoin ETFs took in about $3.08 billion over nine straight days through Sept 29. The streak ended the next day with an outflow of about $148.69 million.
Bernstein analysts have pointed to a changing ownership structure. In June, they said combined flows from ETFs and corporate treasury companies had brought about $12 billion of net new capital into Bitcoin in 2026 so far, against roughly $60 billion during 2025. In August, they linked Bitcoin’s rebound towards $80,000 to improving liquidity, renewed ETF demand and larger US Treasury purchases of longer-dated debt.
That relationship is being tested again. The 10-year Treasury yield reached 5.223% on Sept 24, a level not seen since June 2007, and it hit 5.23% on Sept 25, the highest since 2007. The 10-year did hit 5.34% on Oct 1, its highest since 2002.
Read also: ‘Uptober’ test: Can ETF demand keep Bitcoin’s rebound alive after the CLARITY Act setback?