Dear David, This article in the FT speaks to some of your concerns about trading algorithms. http://www.ft.com/cms/s/0/9c3a1b1a-c33f-11e5-b3b1-7b2481276e45.html#ixzz3yHqWkjYI One of the points it makes is that the algorithms are mainly 'trend following' and do not take value criteria into account: "Their models analyse price data, rather than creatively assessing, for example, how accurately a share that represents fractional ownership of a real business reflects the present and future value of that business." This seems to be supported by my own portfolio, where so far through this bear market the shares selected using my 'momentum-value' criteria have not yet broken their uptrends (take a look at Bellway PLC, Berkeley Group PLC, CVS Group PLC, ITV PLC, Fundsmith Equity Fund, Redcentric PLC and Character Group PLC. Another point made in the article also rings true: "...the advancement and wider adoption of computer-driven trading systems may sow the seeds of their eventual obsolescence. If, in the future, they ever become advanced enough to come close to “solving” trend following in financial markets, then the machines will gradually erode each others’ ability to make money. When everyone is using the same trading strategy, no one is left with an edge and the super computers will cancel out each others’ advantage." What are your thoughts on the article? Best wishes David

David Fuller's view

Many thanks for a very interesting email, certain to be of general interest.  I congratulate you on your ‘momentum-value’ success with the shares mentioned above, to which I have added weekly charts.

Momentum trading, also called trend following is one of the oldest technical strategies known to investors or traders, and can work very well as your examples mentioned above demonstrate.  Indeed, Eoin’s Autonomy fund uses a similar strategy, as you probably know. 

However, there is an old adage which most subscribers will be familiar with:

You have to kiss a lot of frogs before you find your prince (or princess).

For every share which performs with the consistency of Bellway PLC above, there will be plenty of others which may look promising but soon lose their relative strength.  Superior filtering can reduce this problem for long or short positions but the market environment usually has more inconsistencies than the desirable trends which resemble one-way traffic.  The main exceptions tend to early relative strength leaders which have compelling fundamental stories.  The superior consistency of their trends is created by the greatest degree of supply/demand imbalance.

Bull or bear markets will produce the more favourable conditions for persistent trend consistency.  Conversely, choppy market conditions will generate more whipsaws. 

As for the FT article: How Human Traders Will Beat the Machines, some will but the future does belong to machines.  For instance, they can scan thousands of instruments and instantly identify early relative strength or weakness, plus the most persistent consistency characteristics if programmed to do so.  They can also spot losses of consistency across the spectrum of market instruments before individuals, however dedicated and obsessive the latter may be. 

Yes, the machines will destroy each other as clever programmers fine-tune their strategies more quickly than others, although it will not be easy for anyone to hold that position for very long.  However, machine wars also increase intraday volatility, wearing down or frightening away individual traders and investors. When the machines have sufficient AI to programme themselves, humans will be redundant.