|
PureBytes Links
Trading Reference Links
|
Paul,
I once tested a similar system that was published by Larry Williams many
years ago. I believe it bought/sold on a breakout of the extreme close
high-close/low-close of the past 48 bars and exited on a $1000 stop or the
extreme high-close/low-close of the past 24 bars. I tested this over 400
years of combined data and found that it consistently made money over every
market tested, and if used in a portfolio approach made money over every
12-month-or-greater time period I looked at. The down side is that the win
ratio tended to be very low, usually between 30% and 40%, and draw downs,
while not huge, were larger than I was comfortable with. But...this does
illustrate just how simple it can be to make money with a no-brainier
mechanical system if one has the discipline to stick with it. No doubt,
this is something similar to what Richard Dennis was doing when he made his
pile of money. Also Dennis cashed in during the inflationary 70s, a time
when a system such as this would have excelled. Let me know if you stumble
upon some significant improvement on this technique, as this type of
trading strategy does appeal to me.
All the Best !
Bill Shumake
-----Original Message-----
From: Paul Cote <cote@xxxxxxxx>
To: RealTraders Discussion Group <realtraders@xxxxxxxxxxxxxx>
Date: Sunday, February 15, 1998 10:21 AM
Subject: channel breakout
>Systems Designers:
>
>In supercharts, I have set up a system gotten from Krutsinger's secrets
>of the masters. The system is by Rich Saidenberg. You creat two
>indicators that form a band. The top indicator is the 3 day moving
>average + two times std deviation of fifteen bars, the whole thing
>offset by ten [10] The bottom band is 3 day moving average minus the
>std deviation thingy. Now, when you close above the top line, you go
>long, when you close below the bottom line you reverse. This sytems
>seems to nearly mirror a channel breakout ala turtles. Curious if
>others have tried such a system. This system is very easily programmed
>into supercharts. Preliminary results are attractive albeit you get a
>lot of intraday drawdown, however it seems to perform very well.
>
>Paul Cote
|