Higgenbotham wrote:
> Except for the disparity in time (number of days for the pre-crash
> pattern to play out), I see a lot of similarity in the above 2
> patterns
> When I day trade, this is what I'm doing on a micro scale. I'm
> looking at market behavior and comparing it intuitively to
> situations I've seen before. That's hard to do because the time
> sequences don't normally correspond one to one, but there are
> vague similarities.
> A lot of people try to categorize these similarities into
> definable, identifiable categories of patterns. I don't find that
> to be helpful.
> What I see above broadly is 2 patterns that are similar enough
> (particularly the ending rebounds) to be actionable if one were to
> consider pattern as the only criteria.
When I was counting days, what I did was look at the interval from the
beginning of the earnings period to the October 1929 crash, and then
compare that to a similar period for this year's Q1 earnings. This is
based on the idea that the 1929 crash was triggered by Q3 earnings
reports that were below expectations.
I think that a meaningful comparison might be possible if it compares
two intervals of a few weeks. But I just don't see how comparing
15-year intervals can make sense. There might be a war in one
interval but not the other, or there might be sequestration or some
other fiscal policy in one interval but not the other, or something
else that would make the two intervals incomparable.
Things like Kondratiev cycles might be comparable, but even then the
timespans are approximate, rather than an exact number of days.
As you say, when you day-trade, you're working on a micro scale. That
makes sense to me.
