|
PureBytes Links
Trading Reference Links
|
Chris Jackson wrote:
>
> I would appreciate feedback from anyone who has experience with this
> approach.
>
>
>
> "COMMODEX is the oldest, most diversified, most consistently profitable
> daily futures trading system in the world".
>
>
I used Commodex for about 6 months and even resubscribed to it off and
on after that, although I'm not using it currently. Commodex is a fairly
active trend following system that has been published for a long time.
The system gets in pretty quickly and rides trends for a long time if
the trend continues. Track records are available at their website,
www.commodex.com. These are said to be the published signals which is an
advantage over backtesting where you have to worry about hindsight. They
claim to have never had a losing year.
There are some things you should know about the track record and
system, however. These are related to the fact that it takes positions
based on the closing price. The track record assumes you got in at
today's close whereas you are actually getting in at tomorrow's open.
The differences might cancel out, but one suspects in a trend, tomorrows
open tends to be higher and so you get in a bit worse on average
than what their computer will say. (If you get in on Globex you might
get the an entry price very similar to the close.) Also, more
significantly, their computer treats stops in a way you can't actual
trade! Supposedly these are close only stops. Yet if they are hit the
computer calculates the loss based on the stop not the close. But if it
bounces off the stop the computer stays in. You're going to have to
choose whether to use stops or SCO. The computer doesn't. What I did and
would do now if I was using it, would be to have my own trailing stops.
If you select a stop that is somewhat similar to Commodex's but one that
you're comfortable with either as a stop or stop-close-only, then you
will approximate their performance. But how do you actually test this?
The only way would be to get their track record, pull up the contract,
"get in" where they did, run your stop and calculate profit and loss for
each trade manually. I'm not aware that they sell the system rules but
there used to be a book that described it.
The system takes into account volume and open interest so it's
actually analyzing action from one day prior to the day published. Since
it's a reasonably long term system, this may not be too bad a
disadvantage but you will need to be comfortable with that. This feature
is already accounted for in the track record, though.
Their track record shows, like many systems, a profit by the
difference of large numbers. Maybe 2/3 or 3/4 of the commodities they
track show a profit in a given year, the others losses. This is typical
of any system, but this was my other problem when I used it. The
portfolio I chose didn't happen to have the big trends that year. I
can't say I was successful but I believe it was because I was
overtrading my account more than
the system. While I did not exhaust my account completely, I felt it
necessary to take a break.. and of course that's when some winning
trades came along. As you may already know, make sure you understand the
drawdown of any system or newsletter you consider. This is as important
as the overall profits. In other words: during the year, how
low will the dip in your account be, before you come out profitable at
the end of the year. You could get a rough idea from Commodex's track
record just by looking at the closed out trades. The features noted
above make this more approximate, however. Approximate, but a real track
record.
If I was guessing, I'd say the system trades each commodity about
10 - 15 times per year. It's sometimes very good at calling turns in the
market, particularly if there's a short consolidation between the down
and up trend. It's slower at calling a reversal if it's v-shaped. They
did quite well in 1997 I believe.
In summary, it seems to function like a pretty good entry system.
But you
would probably have to run your own stops.
Conrad Bowers
|