Why our 'no-thought' stock screen needs a rethink

Less than a year after reviving our quarterly No-Thought portfolio, I am starting to lose patience with it. The six-part screen, which has been in service in some form since 2004, again lagged its benchmark in the first three months of the year. On its own, short-term underperformance is not a reason to get fed up. But it is the nature of the underperformance, which I’m increasingly convinced is due to its odd construction, that irks me. Whether the portfolio needs to be tweaked, reinvented or abandoned, I haven’t yet decided.

What then, is going wrong? In a word, focus. Although the No-Thought comprises six separate 20-stock screens – longs on the FTSE 350 constituents with the best one-year momentum, dividend yields, lowest five-year volatility and largest market values, and shorts on the worst one-year performers and most volatile – its long-time steward tended to see the whole portfolio as a shorthand for momentum investing.

Even on that score, I am not sure it does the job. The ongoing success of our quarterly Momentum Classics screen (which draws its selections from the FTSE 100) is, I believe, down to its focus on three-month share price appreciation. By extending the momentum window to one year, the No-Thought increases the risk of picking stocks whose rallies peaked months ago (and vice versa for the negative momentum shorts).

This lack of responsiveness hit the No-Thought last quarter, and again seems to have been a factor in the opening months of 2023. Both the positive and negative momentum screens underperformed the FTSE 350, whether dividends are included or not.

There are other issues. The value screen, which selects the highest trailing dividend yielders, has consistently failed to beat the benchmark, while the equally blunt megacaps filter has recently reverted to its underperforming type.

But the high and low-beta screens deserve a special mention. While I can see some logic in buying low-volatility stocks, this isn’t captured by the five-year focus on a share’s beta. Shorting high-beta stocks on the other hand – while the most successful sub-screen over the past two years – strikes me as a very high-risk factor to pursue. Indeed, over three years, it has underperformed the FTSE 350 by 93.6 percentage points, highlighting the huge risks involved in repeatedly betting against high-risk stocks.

The resulting screen is one that has proved less volatile than the FTSE 350 over the past five years. But its somewhat contradictory range of factors has largely served to cancel itself out. Given the risks – of allocating a third of its holdings to shorts, and leaving everything else to single-metric filters – the No-Thought portfolio’s very modest rewards have not nearly been worth it.

As a result, I have little faith in this quarter’s gargantuan batch of stocks to generate alpha. I’ll give it some thought, but I'm tempted to conclude it may be this screen’s last outing.