Sponsored Content

Diversification Underpinned by Science

If you’re looking for a diversifying mainstay in your portfolio, here’s the trend for you.

Sponsored by 


If you were trying to explain the investment strategy of trend following to someone in five seconds or less, you’d say something akin to “It’s as easy as buying what goes up and selling what goes down.”

It is hardly that simple, of course, but if like many investors your goals include maximizing forward-looking returns at scale, it’s difficult to ignore the logic of the strategy – particularly when it’s underpinned by a systematic algorithmic approach that allows it to be managed within a tight risk envelope.

Few have studied the art of trend following on a par with Capital Fund Management (CFM), where every strategy is built upon a quantitative and scientific foundation. II recently spoke with Philip Seager, Head of Strategy, Quantitative Investment Solutions at CFM, in a conversation that began by addressing the fact that trend following – much like the trends themselves – seems to run hot and cold with investors depending on when you ask them about it.

Is trend following back in fashion just now?

Yes and no! There are two camps emerging – one stating that trend following is dead while the other is stating that it is alive and well but delivering a level of risk adjusted returns consistent with its long-term average. Certain discretionary investors have even espoused the merits of it. During speaking engagements, I sometimes quote George Soros, who essentially says that trend following works until the trend becomes exhausted, at which point he goes in the opposite direction to the trend. He has been a successful discretionary investor, and even if he’s been doing it in a discretionary fashion, part of his strategy involves trend following. I would argue that other discretionary investors follow trends subconsciously, to an extent that you can model their return streams and you will see a component of trend following. I would further argue, in fact, that it’s rather difficult to not follow trends, even if you aren’t aware of it. It’s quite ingrained in the human psyche.

Right now, there’s potentially a fashion for saying that the “trend is back,” similar to what happened in 2014. Prior to 2014, people were dismissing the trend. It was a case of “The trend is dead!” and then at the end of 2014 “Long live the trend!” The general feeling toward it is purely based on performance – which shows trend following is so ingrained that people even follow the trend of trend following performance!

CFM quote

CFM quote

Do you find that investors are as well versed as you would like in the complexities of trend following?

They are, but they’re not well versed in managing their expectations, and that’s more the issue. Trend following is a great strategy, but its main advantage is that it tends to be uncorrelated with standard traditional benchmark strategies, such as going long equities or long bonds. It’s not so much that it’s better than the traditional benchmarks, but that it has real added value in combination with them. When investors’ expectations aren’t managed, they tend to set them too high and enter a trend following strategy thinking it’s going to be an extraordinary performer in their portfolio.

What should their expectations be?

For example, the S&P 500 went through an extended flat period of 12 or 13 years from the 2000 dot-com bubble onwards. Given that the expected levels of risk adjusted returns between trend following and an S&P 500 investment are similar, such prolonged periods of flat performance are likely and should be expected for an investment in trend following. Indeed, very long back tests of trend following and equities reveal such flat periods of performance have occurred in both even prior to 2000.

Your approach, however, isn’t the same as those discretionary investors you mentioned?

You’re right, it isn’t. As a systematic quant algorithmic asset manager we can analyze much more data, trade more instruments, and manage a portfolio to a very narrow envelope of risk. A discretionary investor cannot. A systematic and algorithmic manager therefore can exploit the trend more fully.

Getting back to those expectations a bit – what are some of the misconceptions that investors assign to trend following?

One big misconception is that it’s a bulwark against equity downturns. There is an aspect of that – if you’re trending on an equity market and it goes down strongly or crashes on a time scale comparable to the trend, then your short equity position makes money – the longer the crash continues the more you make! This “convexity” on one contract is mechanical. It does not, however, guarantee positive performance outside of prolonged crises! Any performance from trend following in normal times should, we believe, arise from exploiting a behavioral bias in markets.




But if you only trend on equities, then your risk adjusted returns are not great. You try to improve your risk adjusted returns by trending on a big portfolio. You go beyond equities and start trending on commodities, on FX, and on bonds and everything you can. Now, the more diversified your portfolio gets, the less protection on equities you have. I think there is some protection on equities in a broader trend portfolio, but not as much as people think.

How do you address this misconception with investors?

We feel investors should use trend following in their portfolios as a diversifier. Your expectation of returns should be that it’s about same as equity markets and bond markets, but the investment tends to be uncorrelated.

So, there is some equity protection, perhaps just not as much as some investors believe.

As I said, you do get some protection, and you can modify your trend following approach to increase protection on the equity downside. At CFM, we have developed a second trending program that tries to maximize the amount of protection and hedge that we get from trend following, but there is a tradeoff: more protection, but about 20% lower risk adjusted returns. We’re transparent about this trade off and believe some clients want this benefit.

Is there an ideal investor type or mindset for trend following?

One who is trying to maximize forward-looking returns with scale, and who is looking to diversify away from equities and bonds. It’s the scale that should resonate. The main parameter in a trend following strategy is the time scale over which you define that things are going up or going down. The level of risk-adjusted returns does not vary strongly with that time scale. I would say trend following is not well suited to performance chasers.

Why is that?

In trend following, you tend to get infrequent big returns on big moves of the underlying markets, and frequent small, average, close-to-zero returns. Combined, the infrequent big returns and frequent small ones have a positive skew – the smaller returns collectively are of a lesser magnitude than the infrequent large returns.

If you are a performance chaser – and let’s face it, most people are, because that’s what generates the trend in the first place – you tend to invest after the big infrequent moves. You are then generally disappointed by observing the trend not really doing much for an extended period of time, which is a performance chaser’s nightmare. Often, they’re not patient enough to stay in it until they witness the next acceleration.

What do they miss out on due to that lack of patience?

Everyone is familiar with the 2008 financial crisis. As it became worse and worse, and all those bets on premiums became correlated, everybody lost money – with the exception of investors in trend following, because the crisis went on long enough that you captured it with trend. Trend followers were short risky assets and long quality assets – the crisis persisted beyond most trend timescales and the positioning continued to reap the rewards.

Learn more about pioneering scientific investment strategies.


Any description or information involving investment process or allocations is provided for illustration purposes only. There can be no assurance that these statements are or will prove to be accurate or complete in any way. All figures are unaudited. This article does not constitute an offer or solicitation to subscribe for any security or interest.