I would very much appreciate seeing your correlation. Either with pictures or dates.
Mt. Gox Bitcoin May to today (late July 2011) vs.
Nasdaq Composite 1994 to 2008
Whoa, that's not what I meant. First of all, I compare nothing of NASDAQ after 2003, only the last big rally and the time where the market is not in "bubble destruction mode". Also, the comparison needs multiple corrections. I wouldn't expect Bitcoin to be as close as the NASDAQ and DJIA bubble were between one another, simply because the timings in the digital age are heavily distorted, Bitcoin is much smaller, and even with that said, everything happened at blazing speed in the first half of this year.
Timing issues I expect:
- Bitcoin is illiquid. Times in which the price appears very stuck, especially until a real bust started in August, should not be considered much in comparison with the much more liquid NASDAQ.
- Similarly, the MtGox hack is something one cannot just throw in there. It probably delayed the process, just cutting it in as natural is most likely not a valid approach.
- These saw-tooth like down-up spikes are distorted because of Bitcoin's extreme timing. What is important is that a similar pattern repeats a few times, and includes a characteristic drop, turnaround, and vain rally multiple times and on similar time-scales.
- I personally would not match the "bursting" phase of initial sell-offs with the same scale factor as that of the following "bust" phase, where price is repeatedly lowered in spiky price curves. They have different psychological components that might have changed separately: during the "bursting", the inflow of people and their reaction time to a changed scenario is important, which is much faster in an internet-driven community. But in the bust phase, panic thresholds and the time people take to change their minds is important, which has not changed as much. Thus, in Bitcoin, the bust phase appears longer compared to the initial bursting.
The point of this wasn't to become an exact price prophet, but rather to make qualitative forecasts on price. After the superexponential breakout, prices were very likely to drop, and after that had happened, the following rallies ended in that typical bust-pattern, each attempt to return driven down by the great bears until all denial was destroyed.
It's cheating to match the graphs in a non-trivial way in retrospect, so I won't. Also, it's quite extreme to intend predicting an
exact price curve. This analogy was good enough to make predictions on expectation value, that is, produce money on average. In my case, it saved me from massive losses, and gained me a little in day-trading. I know it working once doesn't prove a thing, but once you "get" what the analogy is about, it is quite striking IMO. I wanted to test it again on silver -- I was quite frustrated my order on put options wasn't executed, just before the 40-30 drop.
Anyways, I'll shut up on this now, as it's a thing of the past now. Bitcoin "bubble mode" might have ended already. At least I can't use the analogy to predict when, or if, the trend stops now. All I want to say is: I think this was more than a coincidental similarity.