This article was written by Roy Tipping and published in The IRS Report in November 2011.

Stopping your losses

In 1978 I built my first computer and taught myself computer programming. I was fascinated by the various theories of investment and thought that perhaps I could bring my developing computing skills to bear on these theories. I went down lots of blind alleys, which were nonetheless highly instructive, not least when my father lost most of his life savings as a result of poor investment advice from a major British bank. At that point I realised that the professionals were as much in the dark as I had been.

In 1989 I acquired my first shares as a result of demutualisation, I was now faced not just with a set of theoretical computational and mathematical conundrums, but the realisation that somehow I had to find a way of succeeding where the professionals had failed. As a result of my father's experience, I am instinctively an investor who, above all other considerations, wishes to protect capital. For me income from capital has never been a major consideration but the increase and protection of capital has been paramount.

Over the years I found that there were 12 parameters that seemed to maximise profits and minimise loss. Some of these are pretty esoteric - though essential - but two well-known and related parameters have stood the test of repeated computer runs of many billions of calculations, and those are the stop-loss and its close cousin the gain-lock.

Whatever the reason for an investor acquiring a traded security, they immediately have the worry about when to get rid of it. Securities don't just rise or fall linearly - they have an innate volatility surrounding the underlying trend. Sometimes the volatility hides the fact that the underlying trend has changed and the investor realises this too late. This is where a rigid stop-loss/gain-lock policy pays for itself. The problem is that there is no universal value for these parameters and they will be different for investors with different horizons.

The value investor who is investing for income will probably stick with their chosen companies through thick and thin: in effect they are using a very large stop-loss/gain-lock, perhaps approaching 100%. The day trader will work with incredibly narrow intra-day stop-losses, perhaps well under 1%, so that their frequent trades don't drain their capital rapidly.

As a momentum investor, I believe that there is nothing new under the sun, and that the Extraordinary Popular Delusions and the Madness of Crowds first identified by Charles McKay can be exploited for personal gain. The stop-loss/gain-lock values that seem to offer the very best returns are around 40% for the stop-loss and 30% for the gain-lock using daily closing prices (not intra-day). What this means in practice is that if I buy a security and it immediately starts to fall in value, I will cut my losses when it falls to a value of 60% of the price that I paid (so I lose 40% of the cost) at the close of trading. If I have chosen well, and the security rises in value, any drop below 70% of the post-purchase high will be viewed as a change in trend and I will cash in my profits the following trading day. I won't be right all the time, but my calculations show that I should do well in the long run.

Gilts are rather different and for a momentum investor a stop-loss of 9% and a gain-lock of 4% appear to be best.

In the TMR portfolio, this means that Intermediate Capital Group (201p for a loss of 33%) and Quihang Equipment (18p for a loss of 25%) should have been sold at their respective stop-losses.

You can see all of Dr Tipping's articles at www.TheIRSReport.com

Dr Tipping writes in The IRS Report every quarter

Call 0800 756 5437 or click here for more information