Also, an exponential curve on linear chart and a straight line on a log chart are very similar. Log chart shows a strong deviation from the usual exponential growth of Bitcoins:
![](https://ip.bitcointalk.org/?u=https%3A%2F%2Fi.imgur.com%2FA5aZP99.png&t=665&c=AD3IsLjZ-nMKZg)
Although I am unsure Bitcoins can continue such exponential growth right now, the current trend of the last few week was way over the top and the predictions for Bitcoins, short of ridiculous and overoptimistic, in my opinion.
That's a nice graph you got there. Kind of proves your point, doesn't it? Only, it has as much to do with your selection of the time interval as it does with anything else. Oh, and the fact that your trendline is going through the lowest points of the graph. Which is not a problem per se, but it implies that you're trying to predict the likely lower bound/minimum price of btc.
How about this: Let's take a log chart of the entire duration of btc trading (okay, almost the entire duration, let's start ~12 weeks in, when we have some data, and volume picked up).
What would a straight line in this log chart look like if it would run from the price at time (0+12 weeks) to the price now. In other words, what would an arbitrary (but fixed) exponential function look like that is accurate at time 0 and now?
Looks kind of different, doesn't it?
Disclaimer: I'm absolutely, totally not claiming that this is the actual (or even most likely) growth function that governs btc price. I'm just putting it up to point how arbitrary your selection of a time interval and intersections points was.
Now, you could perhaps argue that the time interval you selected was actually somewhat motivated, maybe because of the news picking up the bitcoin topic, and the loads of dumb newbie money flooding the market as a result. Again, nothing wrong with that argument per se, but I could construct a similar argument as well:
My exponential function shows that both the 2011 peak, and the 2013 peak were outliers (as they should). And so was the (now corrected) drop to 50 following the recent peak. Interestingly enough, according to my -- admittedly somewhat retarded :) -- function, during almost all of 2011 btc were overvalued, while they were undervalued nearly all of 2012 and even during the beginning of 2013.
I'll say it again: I'm not seriously arguing that the function I plotted is the actual growth function of btc price. I simply wanted to point out that your projection was somewhat arbitrary.
On the other hand, your decision to sell at 250 was almost certainly right, who would seriously argue otherwise? The open question is if we're now moving (slowly? rapidly?) upwards from the current position, or if we're in for a long, painful correction.