An adaptive system with feed-back loops creates momentum (positive and negative) or, described another way, trends for much of the time. While momentum or trend following is based on some complex ideas and theories its practical implication is reasonably simple. Financial markets can be thought of as acting in the same way to tides in the ocean. The primary trend can be thought of as the tide which moves in and out in a slow predictable manner. Within each tide (primary trend) there are many waves (secondary trends) which in turn move in and out but all the time cumulatively in the same direction as the tide. Our process aims to identify these trends near their inception and follow them until they come to an end using the daily swings as a means of entry. The following chart illustrates our process.

The Pruden Model.

Source: From "Life Cycle Model of Crowd Behaviour", by Henry Pruden, Technical Analysis of Stocks 7 Commodities 17, no. 1 (January 1999).

Momentum can be represented by price action but this is by no means the only way of considering it. Earning revision, volume, sentiment, characteristic and news flow all have momentum characteristics which can be exploited by way of a systematic strategy following changes in these variables.

Significant evidence exists to demonstrate that trends which can be exploited also occur in measures of relative strength. The use of relative strength trend following has proven to be an effective methodology for stock selection as it keeps us in tune with the primary themes and helps us avoid the problem areas. Recent studies (for example, Returns to Buying Winners and Selling Losers: Implication for Stock Market Efficiency Journal of Finance 1993) have demonstrated the value of following a relative strength strategy.

A focus on relative strength was first documented in the mid 1960’s when Robert Levy produced his study documenting that past winning stocks tend to continue to produce superior results. This analysis has been repeated many times since in a variety of differing formats all of which have confirmed the basic proposition: assets classes, sectors and individual stocks demonstrate persistent performance which can be exploited even after transaction costs have been taken into account (see Orajczyh and Sadhia).

Financial markets exhibit what has been termed fractal behaviour. This characteristic enables us to identify low risk buy points in assets and individual securities which are demonstrating positive momentum.

Expressed another way, we look to gain our initial exposure to trending stocks as soon as possible after the trend begins. We do not look to establish our full investment at this stage as the risk of it being a false signal is highest at this time.

Once a successful investment has been made, by which we mean that a positive return has been achieved we add to our position by identifying the many short-term pull backs we see in securities with positive momentum.

We believe that investing is a matter of probabilities. By identifying new trends, or corrections in establish trends, lower risk investments can be made than if no account is taken of the prevailing trend. As the chart from Merrill Lynch shows, the probability of a stock with positive momentum and earnings staying in the top quartile of market returns is over 72%.

Probability of a Stock Moving Quadrants

Source: Merrill Lynch Global Quantitative Strategy. Since December 1988

Similar results have been obtained by others who have looked at this issue. Recent studies have suggested that momentum strategies similar to ours are successful 70% of the time, a figure we consider to be an accurate reflection of the process. The chart below from AQR Capital Management demonstrates the effectiveness of a momentum strategy.

Performance of Stocks with good and Bad Momentum
Average Annual Returns for Portfolios Grouped by Momentum (January 1927 to December 2008)

Source: AQR Capital Management. Data is based on monthly returns from overlapping portfolios. Momentum is calculated as the past 12-month returns excluding the most recent month. *Return in excess of the beta-adjusted CRSP Value Weighted Index.

Too many investments go wrong because the person taking the decision becomes emotionally attached to the decision (Behavioural Finance Confirmatory Discounting). They cannot accept that they are wrong and so stick with positions when the evidence becomes over-whelming that would be best served not to. This is a well recognised trait identified by behavioural finance (anchoring and mental accounting). By believing in investment as an exercise in probability no such emotional attachment exists and so the correct action can always be taken.

Actions concerning future activity carries with it uncertainty. In an uncertain environment it is important to ensure that both the probability of success is on your side and that when successful, the returns are significantly larger than when unsuccessful.

Investing with the probability of success on your side is the logical way to proceed and is at the core of our process. However, there will be times when the markets reverse following an investment. No-one likes to take a loss but it is better to take a small loss than a large one. As the table below shows, the smaller the loss the quicker it can be recovered. Recognising these facts, whenever we make an investment we identify at what point we should consider selling it if it subsequently turns out that sentiment towards it reverses. By taking small but quickly recoverable losses, should they occur, enables us to keep focused on the better and stronger performing stocks and assets.

Controlling Investment Risk
Winning by not losing using stop loss

Logic links momentum investing, a focus on probability of success and stop losses. A complete strategy needs all three elements working together to be successful.