Devising a successful fund selection strategy is not easy and many advisers fall into the trap of using recent past performance data in their decision-making process. Unfortunately, explicit performance screens or implicit ones such as reliance on quartile ranking, crowns or star ratings provide virtually no help in the search for funds that outperform.

The problem is that intuitively you would expect there to be some relationship between recent performance and future performance. Manchester City may not win the Premier League next season, but the fact they won this year makes it probable that they will do well next season.

This does not work for funds, there are too many variables, some of which are very sensitive. As a result, using past performance will leave advisers just banging their heads against a brick wall.

We can demonstrate this using a simple scatter chart. Taking data from the IA UK All Companies sector (from which we exclude any funds benchmarked against the FTSE 250), funds are ranked according to their three year performance over 2014 and 2016.

These can be plotted against how the same funds ranked in 2017. If past performance has any predictive value we should expect the good funds in 2014-6, to go on to do well in 2017 and be grouped together. Read more from…

thumbnail courtesy of