This summary is written by Rich based on a conversation in our CTA series between Simon Judes, co CIO at Winton, and the podcast hosts, Niels and Alan.
Simon shared a brief background of Winton, a systematic hedge fund founded by David Harding 25 years ago.
The company manages $10 billion in AUM, employs around 200 people, and focuses on three types of products: a quant multi-strategy hedge fund, a diversified CTA, and pure trend products.
Simon joined Winton in 2008 and reflects on how his initial experiences in the industry shaped his approach to markets.
Simon explained their investment philosophy, which focuses on applying empirical scientific research methods to discover and improve successful investment strategies.
By approaching markets with an open mind, they seek to make accurate predictions about market behaviour. Their scientific approach involves testing multiple hypotheses and avoiding data mining traps.
Simon emphasized that while an economic rationale can add confidence to a strategy, it isn't the only criteria for success. Winton's sources of opportunity are diverse, ranging from behavioural issues to economic relationships and market microstructure, making it difficult to generalize who is on the other side of their winning trades.
Simon clarified that Winton's perspective on trend following has not fundamentally changed over time, even though David Harding mentioned in 2018 that the long-term Sharpe outlook for trend following was around 0.5. This level might not be adequate to run a hedge fund solely on trend following. Harding's observation was based on a consistent decline in the performance of trend-following strategies since the 1970s.
Despite the declining performance, trend following continues to be Winton's primary strategy.
However, the company has made adjustments by diversifying its product offerings. Winton now provides three different products, with some focusing solely on trend following, while others have a lower allocation to this strategy. This diversification allows Winton to continue capitalizing on trend following while also taking advantage of other strategies to enhance overall performance and manage risk.
By having a variety of products, Winton can cater to different investor preferences and risk appetites, ensuring that their offerings remain relevant and competitive in the market.
The Challenge of Timing Trend Following
Simon recognized that some investors treat trend following as a trade and attempt to time it, leading to preconceived expectations about its future performance. He highlights the importance of being empathetic to investor concerns and addressing their underlying fears.
While timing trend following is not feasible, Judes suggested that the risk of underperformance after allocating funds to the strategy can be managed. To do this, he proposes that investors can gradually average into a trend-following strategy over an extended period, such as several months. This approach helps address investor concerns by reducing the impact of sudden underperformance and providing a smoother investment experience.
Judes emphasized the importance of accurately describing the trend-following strategy and its potential reactions in different market scenarios. This helps investors gain a better understanding of how the strategy functions and can adapt to various market conditions. Clear communication is crucial to ensure that investors have realistic expectations of the strategy's performance and do not rely on misleading labels or oversimplified characterizations.
Simon encouraged investment managers to focus on clear communication about the trend-following strategy and its potential performance in different market environments. By understanding investor concerns and proposing a gradual averaging approach, investment managers can help investors build confidence in the strategy and reduce the impact of preconceived notions about its performance. This, in turn, promotes a smoother investment experience and fosters long-term trust in the strategy.
Speed of Trend Following, Machine Learning and Parameter Selection
Simon delved into the importance of speed in trend following strategies and how it impacts their performance. He also shed light on the role of machine learning in the trading process and the challenges associated with applying these techniques to slower strategies.
Judes explained that faster trend following strategies have experienced a decline in performance over time. This decline is attributed to the fact that many investors were trading faster strategies in the past, which led to increased competition and reduced profit opportunities. On the other hand, slower strategies have managed to maintain more consistent returns, albeit with a lower Sharpe ratio.
The choice of speed in trend following is largely dependent on the investor's objectives and the specific market environment. If an investor is purely focused on maximizing the Sharpe ratio, slower trend following strategies might be more suitable. However, in times of market turbulence, such as during a crisis, being more reactive to market changes could be beneficial for the investor, even if it comes at the cost of a slightly reduced Sharpe ratio.
Judes also discussed the limitations of using machine learning in trend following strategies. Machine learning algorithms typically require a large amount of training data to function effectively. In the context of slower trend following strategies, there is often insufficient data to train these algorithms effectively. As a result, machine learning techniques may not be as useful for these slower strategies as they are for faster, intraday trading strategies where more data is available.
Simon highlighted the importance of considering both the investor's objectives and the specific market environment when determining the appropriate speed for a trend following strategy. He also emphasized that systematic managers should focus on their core expertise and not attempt to time market events or predict the actions of institutions like the Federal Reserve.
Niels asked Simon about parameter selections in trading strategies. Simon explained that his firm uses mostly universal parameters but accounts for differences in various markets, such as cost structures and market impact. He also discussed the role of automation and human intervention in parameter selection. While systematic firms like his tend to automate many processes, there is a need for human responsibility at some level.
Simon highlights the dilemma of whether to choose parameters manually or rely on algorithms. The right answer depends on the specific situation, and sometimes, it is a good idea to use an optimizer. However, it is essential to avoid creating a black box where decision-making is entirely delegated to algorithms, making it difficult to understand the reasons behind choices.
Simon's firm generally assigns responsibility to humans early in the process, such as selecting trend following speed and asset allocation. These decisions are made with input from thorough analysis and scientific rigor, but ultimately, humans sign off on them.
Fees and Trend Replication
Niels broughtup the topic of CTA replication strategies and asked Simon for his thoughts on the risks associated with these strategies. Simon disagreed with the view that these replication strategies mitigate single manager risk, stating that they are essentially single managers themselves. He suggested that a fund of funds approach is more effective in mitigating single manager risk.
Niels mentioned a CTA replication ETF with a fee just shy of 1%, while Simon revealed that his firm offers a major markets trend strategy for 80 basis points with daily liquidity. Simon questioned the advantage of investing in a replication strategy based on regression analysis when investors can directly invest in the original strategy at a competitive fee.
Niels inquired about Simon's perspective on trading more markets and if it leads to improved performance. Simon concured that increased diversification and capturing more trends are advantageous. He disagreed with the notion that there is no performance difference between trading a limited number of major markets and engaging with a larger variety of alternative markets.
Simon emphasized that over the past decade, alternative markets have consistently outperformed major markets, except for the previous year. He stressed the importance of incorporating truly diversifying and alternative markets into the trading strategy, rather than merely expanding the number of markets without considering their potential for diversification.
Niels clarified his reference to classical markets, which encompass a wide range of commodities. Simon agreed that when adding more markets to the trading strategy, they should be genuinely alternative and additive to enhance diversification. He mentioned that his firm's focus is on incorporating genuinely diversifying markets, such as Chinese futures, regional gas and power markets, freight markets, and marine fuel, all of which have contributed significantly to the overall performance in recent years.
This discussion highlighted the importance of diversification in trading strategies and the potential benefits of incorporating alternative markets into the mix. By carefully selecting genuinely diversifying markets, traders can enhance their performance and better capture market trends. This approach requires thoughtful consideration of market correlations and the potential value of including non-traditional or alternative markets in a diversified trading strategy.
Drawbacks of Trading Less Liquid Markets
Niels asked Simon how he balances the benefits of performance from alternative markets with the potential drawbacks of not being exchange-traded and taking on counterparty risk.
Simon explained that while the majority of trading activity and liquidity in alternative markets may not take place on an exchange, it is still possible to clear positions through an exchange. Clearing positions on an exchange means that a central clearinghouse acts as an intermediary between the buyer and seller, managing the risk associated with the trade.
This process significantly reduces counterparty risk, which is the risk that one party involved in the trade may not fulfill its obligations. By clearing positions on an exchange, traders in alternative markets can manage their risks in a way that is similar to trading more traditional, exchange-traded assets.
He also emphasized that the barrier to entry in these markets contributes to their extra alpha since it's harder for speculative trend followers to easily expand their strategies into them. This requires organizations to structure their trading and middle office teams to handle the additional markets efficiently, which involves significant effort in software and automation. Despite the challenges, Simon believes the potential benefits make it worthwhile to invest in alternative markets.
Role of Trend Following in a Larger Portfolio and Risk Management
Alan discussed the role of trend following strategies in the context of a broader portfolio and asks Simon about customization and risk management. Simon explained that they customize programs based on clients' requirements, and often it's more about the amount of exposure to alternative markets or specific markets like Chinese markets.
When it comes to risk management, Simon noted that the CTA industry often equates risk with volatility, which can be misleading. Instead, he believes in considering various other factors and running scenario tests and stress tests. By looking at how current positions would have performed in different historical market environments, they can better understand potential risks.
At present, Simon is concerned about a classic risk-off move where bonds rally and equities decline simultaneously. To manage such risks, they may adjust the leverage of the portfolio based on signal strength, market correlation, and other risk measures.
Niels asked Simon about the potential risks that could affect all major asset classes simultaneously and how to balance caution with realistic scenarios. Simon explained that they look at historical scenarios and also consider stylized scenarios, such as changes in correlations that could affect their positions.
Niels also asked about capping the trend follower's ability to go long equities and the number of trend following strategies an allocator should have in their portfolio. Simon believes it depends on the investor's objectives and preferences. Capping the ability to go long equities might reduce long-term performance but could be suitable for some investors. Regarding the number of strategies, there is a trade-off between diversifying to reduce single manager risk and paying more in fees. The allocator must decide where they want to be on that spectrum.
Return Expectations of Trend Following
Alan asked Simon about return expectations and whether the historical Sharpe ratio of 0.5 for trend following is the default expectation going forward.
Simon agreed and explained that long-term simulations serve to show the existence of trending behaviour in markets, which challenges the efficient markets theory. However, the expected Sharpe ratio for a traded strategy also depends on the number of people trading the strategy, which impacts returns. The 0.5 Sharpe ratio expectation considers the industry's growth, performance changes over time, and the best trading speed currently. Lower trading costs and fees are also factors that contribute to return expectations.
Misconceptions of Trend Following
Simon disagreed with the notion that trend following is a “dumb strategy” involving herding behaviour. He explained that trend following can often be a contrarian strategy that goes against popular narratives.
Simon provided examples of oil prices in 2014 and German fixed income yields, where trend following strategies bet against the prevailing market beliefs, resulting in successful trades.
He emphasized that trend following doesn't rely on theories about what is supposed to happen but focuses on actual price movements, enabling it to act as a contrarian strategy in certain situations.
Outlook for 2023
Simon is excited about the ongoing original research being conducted at Winton to find new alpha sources and improve existing ones.
This research is taking place across various strategies like trend following, systematic macro, long-short equities, and credit. In addition, he emphasizes the importance of risk management through hypervigilance and monitoring various possible scenarios.
These areas of focus are what Simon believes will be important for 2023 and beyond.
This is based on an episode of Top Traders Unplugged, a bi-weekly podcast with the most interesting and experienced investors, economists, traders and thought leaders in the world. Sign up to our Newsletter or Subscribe on your preferred podcast platform so that you don’t miss out on future episodes.