What this chart shows
Each cell is the percentage change from the first recorded price of that calendar month to the last. The current month is necessarily incomplete — it shows the return so far, not a finished month — and the earliest month (July 2010) is partial for the same reason.
The stat cards above summarize the full grid: how often a month closes positive, the average monthly return, and the best and worst single months on record.
How to read it
Scan a column (say, every September, or every December) to eyeball whether a specific month has historically leaned one way. With roughly a dozen-plus observations per calendar month, this is thin evidence by any statistical standard — but it's the same evidence people are citing when they make seasonality claims, laid out so you can check the claim yourself rather than take it on faith.
Limitations — why this isn't a seasonality trading signal
Bitcoin has traded for around 15 calendar years. That means any single month (say, every October) has at most 15 observations — nowhere near enough to establish a statistically reliable seasonal pattern, no matter how consistent it might look in this table.
Each of those observations came from a different market structure and macro backdrop (pre-exchange era, ICO boom, COVID liquidity, ETF era). Treating them as repeated trials of the same experiment, which any seasonality claim implicitly does, is a stretch.
This page reports what happened, not why. A run of strong Novembers could reflect genuine seasonality, pure coincidence, or a few outsized events landing in the same month by chance — the grid alone can't tell those apart.