Random System, Loss Control?
by Mark Harris
In his Settlement column in the January 1991 issue of STOCKS & COMMODITIES ("Trading simply:
Minimizing losses"), Technical Editor John Sweeney says, "I haven't tested a random set of rules yet, but
I wouldn't be surprised if they worked given good loss control." I thought it might be interesting to check
out a very random system with a certain degree of loss control. I'm hedging on this a bit because the
system I've set up literally only gives a degree of control over losses.
The basic idea of the system is as follows: On some randomly chosen day, we buy a contract on the
Standard & Poor's 500 (S&P 500) index. The price used will be the closing price. We will have in mind a
"stop" value, which will be some number of points (say x) below the closing price. Now if, at the
following day's close, the price is below our stop value, we will consider this a losing trade executed at
the close. On the other hand, if the price is up, we will hold and raise our stop so that it remains x points
below the closing price. We will continue this process until, eventually, we are stopped out.
It is quite clear that we aren't going to control our losses to x points each trade. What we'd really like to
do is place an order specifying "if the close is going to be below my stop, then (prior to the close) sell at
my stop." Now that's the kind of order I would like to be able to enter, but so far no one's been interested
in accepting it. I will return to this point a little further on when I discuss "slippage".
After a transaction is completed, we wait another random number of days and do it again.