- As Q2 of 2024 unfolds, the geopolitical environment is as dynamic as ever. But if you’re a systematic trader, chances are the whims of current markets aren’t part of your calculus.
- Aspect Capital cofounder Anthony Todd says that his firm’s investment decisions are strictly based on its models and research, not on prevailing economic conditions or market sentiment.
- How is Aspect Capital taking trend following into the mid-2020s and beyond? What’s Anthony’s take on the mood of 2024 so far?
“Bear in mind, we’re not economists. Everything we do is entirely consistent and static,” says Anthony Todd, cofounder of Aspect Capital, a systematic investment management firm based in the U.K. with offices in London and Stamford, Connecticut.
Founded in 1997 by Anthony and his partner Martin Lueck (a repeat guest on our show), Aspect Capital manages over $9 billion in a range of systematic investment solutions, including trend following, macro/currency and multi-strategy/customized solutions. Its approach is both “scientific and creative,” with an emphasis on research.
“If you ask anybody in the company, we’ll all have a view about what’s happening in terms of the global economy,” Anthony says. “But it has no bearing on our models, on our research, on the positions we’re taking at the moment.”
Why isn’t the current economic milieu part of Aspect’s methodology?
Anthony joined Alan Dunne and me for a Top Traders Unplugged episode to answer that question, and discuss the role of systematic trading and trend following in today’s economy, including the rise of AI in the space. Read on for highlights of their wide-ranging conversation.
The most dangerous words in finance
Anthony says that during the “tough period of performance between 2009 and 2018,” a number of investors, commentators and Aspect’s competitors took the view that market behavior itself had changed — that since the markets became faster to respond to market news, the right thing to do was to take in-kind action with lightning-fast responses in the form of trading.
“We looked at that in extensive detail,” he adds. “And we could not reach that same conclusion. The most dangerous words in finance are this time is different. We cannot actually see any evidence of that market behavior change. So we retained our aggregate level of responsiveness but did a lot of work trying to identify ways of repeatedly improving our responsiveness at those turning points.”
However, Anthony says, we’ve seen “a complete seismic shift in the market environment.”
“If you look at the period post-global financial crisis, we saw an extended period of coordinated central bank intervention, quantitative easing on a global basis and a zero-interest-rate kind of policy — significant levels of government intervention,” he explains. “And it was that significant government [and] central bank intervention in the markets that distorted market price behavior and limited the ability for markets to trend effectively across multiple fronts.”
Consider the alternative
Alternative data and machine learning have been significant areas of investment for Aspect Capital over the last seven or eight years — “particularly in our systematic global macro program, but also elsewhere,” Anthony explains.
In the context of investment management, “alternative data” is non-traditional data that can provide investors with insights beyond the conventional financial statements and economic indicators. Asset managers and hedge funds often use alternative data to gain an edge over the competition, make better investment decisions and produce astute predictions about the financial markets.
At Aspect Capital, alternative data include non-pure, price-based data like “nowcasting … more instantaneous measures of certain key economic indicators,” as well as flow data that “provide valuable additional sources of return,” says Anthony.
‘Intuitive and explicable’
Aspect Capital runs a “sleeve of 24 machine learning models” in its “mousetrap” program.
But those models are “constrained” by the firm’s central tenets: everything should be hypothesis-driven; intuitive and explicable to clients; and consistent in style (“avoiding style drift”).
“We don’t undertake research on the basis of just trusting the data,” he notes. “Any [economic] effect we’re trying to identify in the first place has to be backed up by a hypothesis about market behavior.”
That’s a very different strategy from the typical “data-driven approach” many firms use — which is essentially the idea that “if you have enough data and are strong enough … because statistics, then you can just trust the results.”
“We don’t work on that basis,” he argues.
“We see huge advantages in the application of AI, particularly in relation to increasing productivity in research and … in terms of coding. But we’re moving cautiously. … We need to actually make sure that the work we do using AI is hypothesis-driven. We’re not just asking AI for a solution.”
Although Aspect’s team pride themselves on providing a high degree of transparency to its investors, “we’re highly protective of our IP,” Anthony says.
Sophistication and diversification
In terms of the investment philosophy, Aspect Capital has been consistent with their approach for 25 years.
Looking back on the early days of Aspect Capital back in the late ‘90s, “our aim was to build a medium term trend-following program, capturing trends over a period of two to three months over a broad set of diversifying markets,” Anthony explains. “And at the highest level, that’s exactly what we’re trying to do today.”
What has changed is the “level of sophistication” of those models — today’s are much more complex and diverse than they were 25 years ago.
That’s due to a number of factors, the most obvious being the number of markets that Aspect trades. When the firm launched in December 1998, its portfolio consisted of 64 markets. Today, the company trades in 180 markets, many with multiple contracts — so its portfolio trades more than 240 contracts overall.
Machine learning
In its early years, Aspect Capital primarily used daily cut-price data. But the level of data granularity it uses today is “significantly finer.”
Anthony and his team’s filters for identifying uptrends or downtrends in each market are considerably more advanced than they were a quarter century years ago.
“If you translate that trend strength into a position in the markets — we talk about the application of a position function — that’s the way we translate a trend signal into a position in the market,” he says. “A significant amount of work and research has gone into refining those position functions. Over the course of the last five years, a significant focus of research has been put into trying to improve our responsiveness at turning points in markets.”
“We’re very excited,” Anthony says of the rise of AI in trading. “But we move carefully … and make sure our approach is consistent with our central tenets.”
Amid the rapid advancements in AI, we should all remember to remain this steadfast in our commitment to our guiding principles.
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.