There are a lot of technical commentaries out there right now looking at the current pullback in comparison to all the other pullbacks since the low in March 2009. I've seen lots of means, medians, ranges, averages, % above/below 50 day MA, etc. bandied about, all inferring that the recent past is a good approximation for what is likely to happen this time.
Quite apart from conducting dubious statistical analysis with limited explanatory power on a meaningless sample size, what kills me is that they do it everytime and they invariably miss the forest for the trees.
The technicians are always very good at describing what has happened, but like all other prognosticators, they are not so good at telling us what will happen.
Addendum: It's all about the graphs. It also really bugs me when someone uses a volatile data series depicted in graphs to make their case. They invariably extrapolate some future scenario based upon (1) an historical pattern, (2) the current trend, or (3) both 1 and 2, but with so much noise in the data, it is a crapshoot.
No comments:
Post a Comment