Frame the comparison before reaching for a conclusion. Choose a window, name a possible driver, and write down what you actually observed.
The supplied post asks what happened to the correlation, but does not provide a time series or cause. This page keeps that uncertainty explicit.
This worksheet frames a question about a Bitcoin and money-supply comparison. Its source explicitly provides no time series, live market feed, measured correlation or causal explanation. Choosing a time window or a proposed driver changes the wording of a local brief; the source does not fetch prices, calculate a correlation or establish that driver as a cause. The diagram separates a chosen comparison frame from evidence that has not been supplied. This page makes no investment conclusion. The window selector has four source choices, and the driver selector also has four. Four times four gives sixteen possible wording combinations. Both bars count choices on a common one hundred pixel scale. Changing either selector updates the stored window and driver, reads the current observation note, and saves all three fields locally. The displayed driver is converted to lower case in the brief. These controls change labels and a saved record, not a market model or a measured economic relationship. Typing an observation saves the note alongside the current window and driver. Export downloads JSON containing the fixed source question, current selections, observation, scope and evidence boundary. It does not interpret the observation as proof or send it to a model. Reset removes local storage and performs an actual reload; the default ninety day window, liquidity regime label and empty note then return and are saved by the source initialization. Reload without reset restores the saved fields. The exported brief preserves the missing-evidence boundary in either case.