Quote from: Revalin on January 02, 2012, 07:24:45 AM

Quote from: sunnankar on January 02, 2012, 05:37:58 AM

Post hoc ergo propter hoc.

I do understand the fallacy. Correlation != causation. I've said it in great frustration many times. But now I feel like I've fallen into a bizarre alternate reality where that message got repeated too often, and everyone has now concluded: "... and therefore, causality does not exist."

Quote

Quote

Different individuals value the same things in a different way, and valuations change with the same individuals with changing conditions. . . .The impracticability of measurement is not due to the lack of technical methods for the establishment of measure. It is due to the absence of constant relations.

We are talking about economics with human action and not physics. But I will concede that despite the complexities and unknowables we can know something and from that something derive tendencies.

I agree with everything you just said there. I absolutely agree that the relationships are not constant. But working within your concession, and with a year of good data that coarsely supports the theory, I think it's fair to postulate ergo propter hoc, and that I'm not running afoul of a fallacy in doing so.

Your graph clearly does not show the relationship that you are claiming. Difficulty does not equal price times a factor, even with a delay. It doesn't even equal price times a polynomial.

You need a more complex model. There are more than two quantities, and the connections between them aren't linear. For one thing, the amount of mining that happens is correlated to the value that miners see in bitcoin, which is only weakly related to the price.

The simple model has no predictive power. The price has just gone from $2 to $5 for the second time ever. Look closely at your chart and ask it what the difficulty is going to be, and when. Is it going to triple in a few weeks? Or is it going to increase by a hundred thousand or so? Seven months ago, tripling and increasing by about 100,000 were the same thing. This time, they are not. And look at November and December, the price and difficulty seem to be decreasing and then increasing, but this time together. What happened to the lag?

An increase in price can cause an increase in difficulty, but it is far from clear that it will always. Like ol' Ludwig said, your graph shows a unique historical fact, not an eternal principle.