Methodology
Why the Bitcoin Barometer exists.
Price tells you what Bitcoin costs today. It does not tell you where you are standing in the cycle. The Barometer was built to answer the second question from Bitcoin's own on-chain record, and to be honest about the half of the range where it works less well.
Every decision a member makes sits inside a bigger one: is this a market where you are early, or a market where you are late. That question is almost impossible to answer from price. A chart at forty thousand looks identical whether it arrived from ten thousand or from ninety.
Bitcoin has an unusual property that helps here. It keeps a public record of its own holders. What they paid, how long they have held, when they spend, what miners are earning. Read against Bitcoin's full history rather than against a headline, that record carries information about cycle position that price on its own does not.
The Barometer exists to turn that record into one number a member can actually act on the basis of, and to refuse the temptation to turn it into a signal. It answers where you are. Deciding what to do about that is a separate job, and one the tool deliberately leaves to you and to the Monday written analysis.
Every metric is ranked against its own behaviour from 2009 to the end of 2021. That cutoff was fixed and has never been moved forward. Everything from 2022 onward is scored by a model that has never seen it. A model refitted to recent history will always look impressive on recent history. Freezing the window means the years you can check are the years it could not have learned from.
Rather than asking whether a metric has crossed some remembered threshold, the Barometer asks where today sits within that metric's own historical distribution. Fixed levels decay. A rule of thumb calibrated on the 2013 cycle was already misleading by 2017 and meaningless by 2021. Ranking a measure against its own past is the version that survives contact with a changing market.
An earlier version held the score at a fixed point once conditions got extreme enough at the low end. The effect was that every deep bottom printed the same reading, which flattened real differences between them. That floor is gone. The deepest readings now vary. A deep reading and a considerably deeper one no longer look identical on screen.
When the specific conditions associated with capitulation are present, the tool raises that as its own visible flag rather than quietly pushing the headline score around. You can see whether it has fired, partly fired, or not fired at all. A score that moves for reasons you cannot see is a score you cannot audit.
It is much better at bottoms than at tops
This is the most important thing to understand about the tool, and it is structural rather than a bug waiting to be patched. Every Bitcoin cycle so far has printed less extreme on-chain readings at its high than the cycle before it. As the asset matures and the holder base broadens, the measures that once ran to genuine extremes at a top no longer travel as far. The consequence is that the Barometer catches bottoms emphatically and grows progressively less sensitive at the top end with each cycle that passes.
What follows from that is a reading discipline. Weight the low end of the range heavily. Treat the high end as directional rather than precise, and never as an instruction to sell. The tool has never been validated as a top signal and is not published as one.
It reads early at both ends
Cycle bottoms and cycle tops both tend to register on-chain months before price turns. That is a property of the data, not a calibration problem, and no amount of tuning removes it without destroying what makes the tool useful. It means the Barometer is a poor instrument for deciding what to do this week and a reasonable one for understanding the season you are in.
Some of the internals stay internal
How the fifteen measures are combined into the headline number is not published. That is a deliberate line, and the reason it is stated here rather than glossed over is that a partial disclosure presented as full transparency would be the dishonest option. What is published is what the tool reads, what it puts on screen, and where it fails.
Early or late is the decision underneath all the others.
The Barometer is where members start that read each week. The Monday analysis is where it gets used.
Discussion