Bitcoin's price on a log scale against its long-term trend, banded from cheap (cool blue) to expensive (hot red). A long-horizon guide, not a prediction.
The Rainbow Chart plots Bitcoin's price against a long-term trend line fitted to its whole history, then paints bands around it. Cool blues mean Bitcoin is cheap versus its trend; hot reds mean it's expensive. It's a slow, big-picture sentiment map - not a price prediction.
Price is below its long-term trend. In past cycles these bands (Fire Sale, Buy, Accumulate) have lined up with bottoms - historically the strongest accumulation zones.
Price is around its long-term trend (HODL / Still Cheap / Bubble?). Neither a screaming buy nor a warning.
Price is far above its trend (FOMO, Sell, Max Bubble). Every past cycle top has printed up here - historically a time for caution.
A long-horizon guide that moves over months. The bands are a fitted model, not a law - and because Bitcoin matures, recent peaks have reached lower bands than older ones.
It is a long-term valuation guide. A logarithmic regression line is fitted to Bitcoin's entire price history to estimate its "fair value" trend. The chart then draws coloured bands above and below that line. When price sits in the cool blue bands it is historically cheap relative to trend; when it climbs into the hot red bands it is historically expensive. It was popularised as a light-hearted way to picture where we are in the cycle.
The fair-value line follows a logarithmic regression of the form log10(price) = a · ln(days) + b, where days is the number of days since the Bitcoin genesis block. Each band is that line shifted up or down by a fixed amount on the log scale, which is why the bands fan out as price grows. On this page the band centreline is least-squares-fit to Bitcoin's real price history, and the price line itself is real daily BTC price from CoinGlass (roughly 2010 to today), with the live spot price anchoring the final point.
Because it is a logarithmic model, each cycle's gains are expected to be smaller in percentage terms than the last - the "diminishing returns" idea. That is why Bitcoin reaching a red band in 2013 looked very different from reaching one in 2021, and why recent cycle tops have peaked in progressively lower bands.
No. It is a backward-looking fit of past data and has no predictive power on its own. The model is sensitive to its start date and coefficients, the bands are arbitrary, and past patterns may not repeat. Treat it as one piece of context about long-term value, never as a buy or sell signal.
Use it as a long-horizon sanity check, not a timing tool. It tells you roughly how stretched Bitcoin is versus its own history, which can help frame risk. This page is general information only and not financial advice - see our Disclaimer.