Track Record Investing
As the splashy spectacle that was the Rio Olympics slowly recedes into memory, the event can be looked at through an analytic lens. The BBC published a fascinating article: Why do swimmers break more records than runners? And who knew this was even the case?! Those who are schooled in analysis (that includes you, discerning reader!) may ask: what are the conditions that propel this phenomenon of outperformance? It turns out, that there are four factors, three of which are:
Drawing parallels with the investing world, we reflect on how investing track records are achieved, how they are evaluated, and how they can result in winning a mandate.
Compared with the results of swim meets, analyzing investment track records is much more complex because the environment is not controlled, there has been limited innovation, and optimization often turns out to be the result of changes in the macro environment.
In swimming, records are measured with a singular measure: time to the finish line. In the world of investing, one challenge is that success does not have a uniform measure. Different investors measure performance according to different goals.
When ranked via differing metrics, a lot of managers can claim to be top quartile. Therein lies one of the challenges of identifying a clear winner. Additionally, success is measured over a discrete time period.
Cumulative and consistent success over an extended period trumps a single and stunning victory, which may have been flattered by a particularly beneficial investment exit, or the success of a given month or year. The manager’s skill is not always about getting it right, but rather how they adapt when they have gotten it wrong. Investors who understand this characteristic of their managers may develop a level of trust based on investment philosophy and process that effectively diminishes the focus on short-term quantitative track record.
Much like qualifying rounds leading up to the finals in swimming, most investors use quantitative screens to dictate a binary outcome for a particular investment rather than establishing boundary conditions. Boundary conditions on the other hand allow for evaluating managers on multiple merits, which is important since purely quantitative measures do not guarantee success.
I am critical of the quantitative approach that focuses largely on track record for specific periods, as it induces short-term thinking, fuels herd mentality and eliminates managers who have been on the wrong side of luck.
Thinking about the role of luck brings to mind an experience I had a few years ago when I was tasked with sorting through literally hundreds of resumes to prepare a list of candidates to interview. A colleague jokingly suggested that I randomly stack the CVs on a table and turn on a fan. The candidates whose CVs remained on the table should be invited for an interview. Those whose resumes flew off the table were unlucky, and it was not desirable to work with unlucky people. While this was harmless humor, it reflects the initial conditions bias in investment decisions.
Despite the universal performance disclaimer, past performance generally is used to project future return potential. History provides the only concrete manifestation of the results for any combination of manager skill, macro environment, and industry trends.
The central issue in the current environment is that growth going forward will be volatile given that:
Thus, the ability to break new investment performance records has been constrained by the investment environment and (d)evolving alpha opportunity set. The expectation to generate 10-12% returns in a sustainable fashion is not realistic. Unlike a swimming pool, the investment environment is not controlled.
As track record has seemingly co-opted the whole podium, taking the gold, silver and bronze in the investing world, there have been significant quantitative innovations in multi-factor predictive data models which evaluate track records using algorithms to generate insights and portfolio recommendations. But for these algorithms to work, the track records need to be valid.
What are the remaining areas of track record that investors need to validate so that the output of any such model remains relevant? What checklist can they employ, to remain effective vs. relevant for a given fund, strategy or environment?
Given the pace of investing, and the increasing amount of information flow, how does one deal with increasing complexities of responsibility and ensure that all of these points are validated for each manager that is evaluated? Signup to experience a digital way to ensure you ask these questions using auditable checklists.
Does the past have the ability to predict the future? Or is more required to pick managers who are smart, lucky and adaptive in the shifting investment landscape? There is more to successful long-term investing than a specific period outperformance, and for these reasons, track record investing will continue to be under scrutiny. This overly narrow approach of time bound screening needs re-calibration and optimization to solve for future performance potential rather than just looking in the rear view mirror. Particularly when provided a technology-enabled solution, investors and allocators will optimize by focusing more on the quality of the returns.
This shift will support innovation in the areas of both qualitative and quantitative analysis, paving the way for a new generation of investment champion.