Discretionary portfolio management is full of traps and pitfalls.
Trendrating data and technology can help to address the key problems.
Many strategies are primarily based on fundamental analysis and old school-traditional data.
Poor consideration and respect for actual price trends. This is why the majority of mutual funds underperform the benchmarks, (Source: S&P SPIVA).
The traps of human behavior, (Behavioral finance key flaws).
The disposition effect – selling winners too early, and sitting too long on the losers.
Tendency to forecast and overconfidence.
Biases and attachments.
A large part of risk management tools are based on the research and the development from academics developed 20 to 30 years ago.
Those theories and metrics are helpful, but cannot capture the complexity of today’s market, partly driven by new variables (a faster more reactive big money flow, momentum players, social media influence and more).
Precious time is spent by portfolio managers in reading analysts reports, opinion pieces and research whose value in delivering alpha is highly questionable.
There is a lot of noise that eats time and can generate a confusing picture.
Compliance requirements are increasing, and portfolio managers must provide evidence of a robust diligence process behind their investment strategies.
Totally subjective decisions or the use of a limited arsenal of analytical tools can impact credibility and confidence.
The management of equity portfolios can be performed on a discretionary basis, a systematic basis or a combination of both approaches. Systematic portfolio management is gaining traction as it often provides better returns, lower deviation from an expected outcome, improved discipline and control, full transparency and consistency of the rules governing investment decisions and lower costs. Systematic management can easily fit into most investment strategies (value, growth or contrarian) as the starting point is the universe of stocks that satisfy the defined qualitative or quantitative criteria. These systems are then used to decide the selection and timing of investments. The underlying models are usually tested across years of history to validate the added value they can generate across different market cycles (bull trends, bear trends, ranging phases, high and low volatility) and across a large number of securities. After the testing phase, a real-life validation is required to confirm the quality of the models.