If you read my September 25, 2009 post on several blogs I
have speculated that four 1929 dates Chris Carolan identified and which
replicated by his Spiral Calendar Fibonacci F29 interval, may be replicating
from 1987 to 2009 according to the F25 interval. Why 2009?
Here’s a clue; 1929-1987 is 58.0 years, 1987-2009 is 22.1 years, and 22.1
divided by 58.0 is a Fibonacci .381. The four dates are July 11, 2009,
October 16, 2009, November 23, 2009 and December 10, 2009 computed as follows:
http://www.screencast.com/users/Virginia_Jim/folders/Jing/media/e78ae4c6-8eb2-4977-b8d6-001fe77b25c8
Chris Carolan first identified the dates in reference to
lunar synodic months from the first equinox of the year and hence my reference
to months on the above. However, he later discovered the square root of
each number In the Fibonacci sequence extended by the synodic lunar month
(29.5306 days) produced the Spiral Calendar (SC). The Fibonacci 29th
number (“F29”) added to four key 1929 dates projected the same four
“monument” dates in 1987. This time the F25 sequence MAY be
projecting the same four monument dates in 2009. One key difference is
that the F25 dates projected are an additional moon beyond the equinox.
My thought is that the replication was a function Fibonacci numbers counted in
synodic lunar months and not anchored by the equinox. Further, the
“crash” reflection of 1987 is in December; a month not noted in the
works of Stephen Puetz, Chris Carolan or Peter Eliades as being
“crashworthy.” I have an insight (or delusion as it may
be) in that regard in the last paragraph.
The suggested fractal analog appears as follows:
http://www.screencast.com/users/Virginia_Jim/folders/Jing/media/17aa07d7-2698-46fc-8394-cfc9d6bfb6ba
The above chart was prepared by exactly (as much as
possible) correlating the “significant low” date (the red arrow)
among the 3 years and making sure there were an equal number of trading days
among the three years. Hence, there’s as much inter period
comparability as possible. The green arrow which indicates the final
highest closing high would make it appear the final highest high in 2009
appears later than the other years. On a basis of the number of trading
days after the significant low, that is true, it is 4 trading days later.
On a calendar day basis, that is not true, the “final highest closing
high” is only one day later than 1929 or 1987. The four day
variance can be explained by one extra weekend in 2009, one extra trading day,
and the one day late already identified (October 19 versus October 16).
Of the four dates in 2009, two have passed at this point;
July 11, 2009 and October 16, 2009. July 11 represents a significant low
prior to the climb to the final closing high before the onset of a significant
decline. That date is very significant because ON THAT VERY DAY the
markets were looking at a head and shoulders top that was vastly speculated (it
was publicized by WSJ, CNBC, EWI and others) to destine testing the March
bottom. Instead, the July 11 SC analog date suggested it was a
significant low. The actual low came on July 10 (a Friday) and,
therefore, the projection was perfectly accurate.
The October 16, 2009 date may be subject to some
ambiguity. On a closing basis, the ‘final highest closing
high’ occurred on Monday October 19, 2009 at 10,092, one trading day off
but entirely acceptable. The ambiguity is that a higher intraday high
occurred on October 21, 2009 at 10,117. The market closed well below that
level that day. So, is the 1929-1987 SC analog off 1 trading day or 3
trading days? I’ve had SC followers argue 3 days is an acceptable
error for SC and it may be. Certainly, 1 day is more convincing than
3. Reflecting on the ‘classics’ of TA, the masters of many
years ago might say, and Charles Dow and Edwards and Magee certainly did
say, the ONLY price that is important is the closing price. Therefore,
I’m persuaded the second SC analog date, October 16, 2009, is accurate to
one trading day with one caveat; that a new higher closing high might still
occur. If this date remains true and whether it is 1 day or 3 days, the
analog is becoming quite interesting.
What are the probabilities that, if someone walked up to you
on the street and predicted four significant change in trend dates in the
market, including polarity, a year in advance with a 1 (or 3) day
tolerance? Well,365 days factorial X 4 factorial? Considering
these dates could have been projected 22 years ago, 365 days X 22 years
factorial. How many unique combinations of 4 dates might one make that
would satisfy the projection being made in the 1929-1987 SC analog?
We’re talking a permutation of how many factorial 365 X 22….times 4
factorial? I don’t know but a whole lot of combinations of 4
dates. Practically speaking, we’re talking one in infinity, but
that’s just my guess; they don’t teach us that in CPA school.
And that one unique combination divided by that many possible combinations.
Again, the probability of 4 successes is likely infinitely small.
[I’m reminded by Bob Bronson that a sufficiently obsessive quant could be
randomly lucky to find four dates that worked by fitting the right dates to the
right conditions. And to that I’d take offence because I’m
not obsessive (too much), I haven’t tried to fit any dates to any
conditions and I’m hardly a SC quant. Gosh, my golf partners
won’t let me keep score without a calculator. And I’d further
point out I doubt the computer speed and capacity exists to enumerate all the
combinations of four dates over the last 22 years. And all I have
is one computer and Excel.]
Now, consider that the model HAS SUCCESSFULLY predicted the
first two out of the four dates? Does that make the improbable less
improbable? I know it does, but by how much. About that I
don’t have a clue. But, again, it is interesting.
Assuming the analog WILL produce the final two dates
correctly, what will happen? The next date is the secondary high of
November 23 so there must be a low between October 19 and November 23. If
it is comparable to 1929, the drop from the high will be 20%. If it is
comparable to 1987 it will be 10%. That low would occur the second week
in November. 1929 and 1987 declines are computed as follows:
http://www.screencast.com/users/Virginia_Jim/folders/Jing/media/7b53d6e5-b6ca-49e8-ab14-c95f07ded5a3
Which leads to a final consideration. As noted
above, the studies of Puetz, Carolan and Eliades suggest a crash will not occur
in December…historically, they’ve shown it does not. Crossing
disciplines to Elliott Waves and noting that the foremost of the E Wavers have
the markets embarking on intermediate wave 1 of primary 3 of cycle c of this
Supercycle bear, this first wave might become the “opera” and the
larger December event regarded as the “reprise” or the second
act. If that is the case (remember, this is all BS numerology and nothing
more), the fat lady might be behind the curtain, stage left. If so, have
a seat and hold on to the arm rests.
Good luck,
Jim