Quote from: dree12 on October 12, 2011, 08:14:40 PM

S3052, in your "bullish divergance" chart are you measuring the change of the divergence of the MACD, a measure of change in trends?! Why is this a useful measurement?

In my opinion the only thing s3052 is saying is that the market could go up because it's not going down as fast as it was before. A slower decline or sideways trend is really all the MACD and RSI show, and s3052 calls this a "bullish divergence", which I disagree with. The market is currently in the second significant sideways trend since the peak in the $30s. During a sideways trend, all 3 parts of the MACD (the two lines and the divergence between them) move towards the center line (0.0). So if MACD was low because of a previous down trend, and then the market starts moving sideways, the MACD will start to move upwards, and the divergence between the two MACD lines will be positive. The reverse is true if the trend was up before it starts going sideways. I think a more relevant thing to consider is that sideways trends are typically followed by the previous trend again, which in bitcoin's case is a down trend.

Two more things that I would consider when interpreting the MACD:

First, the MACD is distorted near the beginning of s3052's "bullish divergence" line due to 6 days of no trade data from mtgox being shutdown after it got hacked. If those 6 periods are taken out of the measurement, the divergence between the MACD lines would go into the negative more sharply, and come back up a little quicker as well, and in my opinion would be more accurate than pretending the market price was constant during that volatile time period. This removes s3052's nice straight line across the bottoms of those MACD divergence lows.

Second, I don't think the big decline and recovery just over 2 months ago (which didn't even recover all the way back up to the previous sideways trend) is very relevant to the general trend and to current and future prices. It's deviation from and return to the trend doesn't have much meaning for the short term or the medium and long term. The context of the market was much different then than it is now. The lack of that trend deviation's meaning today implies that the larger moves in the MACD during that time are not very meaningful today, either, when interpreting the larger trend that is more meaningful to us today. My day trades don't take that period into consideration now, and neither do my long term expectations at this point (however, a few weeks or so ago it was much more relevant to me for both the short and long term...the market context my decisions were based in were much different back then as compared to now). As far as I am concerned when looking at the last 4 months of the market declining, putting a lot of consideration into that big drop/recovery doesn't provide much more meaning than considering the 6 days of no price data at all. s3052's line drawn from the the second MACD divergence low to the third seems to be almost meaningless in today's market.

I think it was just a convenient place to draw a line.

I focused on MACD in this post, but the RSI is obviously based on the same market data and likewise contains similar distorted effects that would likely affect s3052's line drawing. This suggests, at least to me, that the idea of there being a "bullish divergence" isn't very well supported by the data. I think this current sideways trend is just another consolidation before prices start to drop more (the end of the last sideways trend wasn't pretty, either...if that big drop a couple months ago is of any relevant meaning at all, it's that it suggests the possibility that the current sideways trend could end the same way before continue back on the normal slow decline).

Just sharing my opinion.