Here is utilize multiplied moving averages using robertoc's open source script. Playing with multiplied averages using days that are close multiples of pi, I may have recognized a trend. The 157 day MA (pi*50) value multiplied by 4 corresponds closely to the 2014 double peak when the price crossed the Moving average. Multiplying the length value by 2 to get the 314 day MA (pi*100) value multiplied by 4 which, when the price cross it in 2018, signified the 2018 top. Multiplying the length value by 2 again makes the 728 day MA (pi*200) value multiplied by 4 which so far corresponds to the 2021 local top. Using only the 350 day MA x 5 does not account for diminishing volatility over time, and so I don't think it can be trusted. Diminishing returns can be seen looking at the 350 day MA x 5 in 2014, when we went way above it and 2021 when we went way below it, so the change in length of the moving averages allows for the market cycle top signal to account for diminishing returns. If you do not believe in diminishing returns over time, this model can be used as a risk metric, but I think there is a more interesting trend to notice.
Like the famous pi cycle top, this model also has pi hidden in the indicators; however, even the pi cycle top does not account for decreased volatility, as with each blow off top, the moving averages utilized get closer to not crossing each other. For those two reasons, I gave it the informal name of Alternative Pi Cycle Top Based On Diminishing Returns.
This model will eventually stop working because every "market cycle" the length of the moving average multiplies by two. Unless we get greatly lengthened market cycles, the growth of the moving average length will outpace the growth of the length of all price history for bitcoin.
This is very dubious. This model should probably not be relied upon. Just speculating
Like the famous pi cycle top, this model also has pi hidden in the indicators; however, even the pi cycle top does not account for decreased volatility, as with each blow off top, the moving averages utilized get closer to not crossing each other. For those two reasons, I gave it the informal name of Alternative Pi Cycle Top Based On Diminishing Returns.
This model will eventually stop working because every "market cycle" the length of the moving average multiplies by two. Unless we get greatly lengthened market cycles, the growth of the moving average length will outpace the growth of the length of all price history for bitcoin.
This is very dubious. This model should probably not be relied upon. Just speculating
