|
PureBytes Links
Trading Reference Links
|
Hi everybody, just lurkin quietly in the background and thought I would add
a couple of words on this topic since I too bought both TD_books. First let
me say that I am neither a promoter nor a demoter of TD's work. The memory
of having worked night a day for at least a month sometime ago with one of
TD's indicators makes me want to tell this story.
I got interested in TD's work when I learned about the counting rules in the
"sequential". The counting of bars reminded me of the the professional card
counter on the blackjack table and I suspected that there was something of
value there. If you have read Tom's book you will find that he goes into
great detail about the counting process (the "9" the "13" and the various
qualifiers) but unfortunately he fails miserably in quantifying his results
IMHO. The problem I think is that he suggests multiple strategies of
entries/exits leaving it up to the individual trader to select them. This to
me is a mistake as it makes it extremely difficult to QUANTIFY the
performance of the indicator and that's where the weakness in Tom's material
is. In his books he never presents a single study showing how any of his
indicators when used with specific entry/exit (his favorites) would have
performed in any one market. Instead he chooses to present isolated cases
with charts from specific time periods where the indicator "worked as it
should". In his first book I don't remember a single chart showing any of
the indicators failing and this is the first thing I want to know when
looking at something new (how it fails when it does fail). Although I am
impressed of the depth of TD's thinking in developing this type of indicator
(unique in the trading literature) I find his work incomplete and at times
confusing. Was this done on purpose? Only TD knows.
After reading his first book 2-3 years ago I coded my owned version of the
sequential in MATLAB. My testing proved inconclusive as to whether the
sequential had any predictive power. At times it worked well with price
collapsing or rising substantially almost immediately after recording "13's"
while other times price just paused or continued without reversal. Sometimes
it reversed well after the count was complete and sometimes it didn't. I
might add that I used mostly intraday data. In summary my best efforts to
quantify the most famous of TD's indicators resulted in the classic
"sometimes it works sometimes it doesn't" conclusion. At this time I no
longer research or use any of TD's indicators in my trading but I still keep
his books on my shelf. Maybe if I get some free time I might look at some of
his other ideas.
PS:
I find his last week's call of an overbought market in the SPOOS without a
sell entry interesting.
|