Forum Archive Index - February 2003

["tennyson@caverock.net.nz" <tennyson@caverock.net.nz>]

Hi Travis,
 
>> 
>>Please explain how you can 'fixed stop' or 'trailing
>>stop' that is 
>>not based on the price of a stock.
> 
>It is a function of the trade size and other things as
>well as the price.  What I meant to say, and I admit I
>screwed it up, was that stop loss orders are designed
>to automatically terminate a trade as soon as a loss
>exceeds a preset amount.  It doesn't depend on any TA
>concept such as support/resistance or trendline breaks
>or whatever. 
> 
>

OK, at last I understand the question here.  I'll start with the 
point of agreement.  

There is no argument that any trader needs a risk management 
strategy.  The argument is over how to do it.  Either method A or 
method B.

A/ Travis is suggesting a risk management strategy which triggers as 
soon as a certain percentage of your wealth is lost on a trade.  In 
other words if your share drops by (say) 5% in value (the exact 
figure you use would have to be determined) in a single day you 
should sell it *no matter what the share price history of that share 
is*.

B/ OTOH if Phaedrus held that same share that dropped by 5% in 
value, he would sell if that share dropped below the long term trend 
line (assuming it was a long term trade we were talking about).  
Contrary to Travis's example, Phaedrus's behaviour will be influenced 
by the long term trading history of the share.

What Travis is asking here is some statistical evidence that 'B' is 
better than 'A'.

To settle this I propose the following experiment.   

If you consider yourself a successful trader (and no-one here is 
arguing that trading can be successful, the only argument is over the 
technique) have a look back at your history of trading.  This means 
*all* of your trades, not just the trades you made money on.  

Got your trading history in front of you?  Now look at the 
recent high point of the share (just before you sold it) and the 
point at which you *actually* sold it.  No trading system is 
perfect so your actual results will always be less than the 
best possible hypothetical results.  Record this figure as a 
percentage figure.  By looking back over your trading history. you 
are now building up a database of your actual results in such a way 
that they can all be compared with each other.  

Record exactly how far below the local share price peak (representing 
the theoretical 'perfect trade') your exit point is.  You will get an 
answer like 'sold the share at 96% of the price of the local share 
price peak'.

Because you are using a trend following system, every exit point 
on every trade will not give you the same answer for each trade.   
You will get a series of answers like 96%, 94%, 97%, 85%, 99% and so 
on.

Now you have the data to come up with your own 'rule of thumb'. 
Something like: "I quit all my trades at an average figure of 4% 
below the most recent local share price peak".

Now, from here on in, this is where the real experiment starts.  You 
need two seperate pieces of paper to record your results.

Parallel to your existing sharemarket trading going forwards, I am 
asking you to try out a second parallel hypothetical system based 
on your own 'rule of thumb'.

At the end of this experiment you will have two sheets of paper.  One 
will have your profits recorded from your trading (this is something 
that all professional traders do already, so I am not asking for 
any 'extra effort').  The second sheet of paper will have the record 
of all trades assuming that they had been exited using your simple 
rule of thumb strategy only.  In this alternative hypothetical case, 
this means 'sell' all shares as soon as the price drops below 96% of 
the previous days listed value  ( I am using this 96% figure as an 
example.  The actual figure you will use depends on your own past 
results and your own personal 'rule of thumb' , but you get the idea 
I hope! )

The point of this exercise is not to prove whether you are a 
successful trader or not (that is given).  The point is to determine 
if your successful trading system can outperform a crude 'sell your 
shares if you lose a certain amount of your capital in one day' 
system.   Do that, and you will have some serious evidence that your 
trend line trading system works.   Do anything less and Travis will 
continue to dismiss your claim that your trend line system is 'giving 
you an edge' as ravings of your own selective sampling.

SNOOPY


---------------------------------
Message sent by Snoopy 
e-mail  tennyson@caverock.net.nz
on Pegasus Mail version 2.55
----------------------------------
"Sometimes to see the wood from the trees, 
you have to cut down all the trees."



----------------------------------------------------------------------------
To remove yourself from this list, please use the form at
http://www.sharechat.co.nz/chat/forum/