Use this one instead, accounts for decreasing volatility. Formulas for buy & sell lines:
>Buy: Price = EXP(LN(Date)*5.5046 - 36.8846)
>Sell: Price = EXP(LN(Date)*4.4549 - 26.2872)
Where date is days since April 26, 2009 (being day 0) - somewhat arbitrary for log scaling
Indicator (graph B) just scales the price so that 0 & 1 correspond to the buy and sell lines. Few more details here: warosu.org
Other urls found in this thread:
Also, lines were calculated on May 15, 2022, well before the dump to 17.6k
checked thanks for posting user
nice chart op
but why does it stop curving in the future
same with the rainbow chart
its all curling at an ever increasing angle of curve until the future projection then it stops curving
if you ignore the data after 2020 what top would it predict? 60k or higher?
also is there a url or smt?
dude can you smartfags just tell me if i should buy or not? no need to make thse graphs and make up new words. just say if i should buy or sell
dca
I’m waiting for a further drop. I’m buying Litecoin under $44, Bitcoin under $17000
It doesn't stop curving, that's just an illusion. This is what it looks like on a log-log scale. You get fixed % returns (log scale) over each % increase in time. It takes more and more time for the same % returns, but it doesn't ever go flat.
>Rainbow chart
>Seems to forget how a rainbow in reality looks like
i also made a chart
tell me what you think
Bump
It's wrong because it models price as an exponential function of time. OP is correct in using a power-law relation.
seems about right
As of now, the buy line is at 18.9k
I'm not going to start cherry-picking when I already get an r-squared of .95 from the entire price history. The graph as it is only has 4 free parameters to worry about.
good thread OP
does eth follow this?
What do you project as the next top for price and date
I wonder the same thing. How would the chart look if you created it using the data up to January 2017? Or up to January 2020? When you have all the data available, it's easy to create a chart to fit it.
>muh r2
so it wouldnt have predicted the 60k top, gotcha.
It looks exactly the same. I know because I posted this exact analysis throughout 2021 and instead of being praised was called a "fucking bobo". Now that the chart is printing 'buy' all the Any Forums smoothbrains are finding profound wisdom in it.
I guess technically 5 free parameters, since setting time 0 (for the purposes of log scaling) is arbitrary. I went with the time that gave the highest r-squared, downweighing earlier years when BTC was flat at $0.1 for a long time
op calls that cherry picking, kek
Even in June, all anyone did was grift, REEing about it hitting 8k
warosu.org
Honestly your model seems too optimistic. ~250k in 2025 seems too...high, i dunno