Every systematic trader hits a flat period eventually. Your strategy stops performing, your equity curve goes sideways for months, and the temptation to abandon ship and look for something better becomes hard to resist. Nick Radge has seen this pattern play out with clients for decades, and his response is always the same: the problem is not the strategy. The problem is not having enough strategies.
Nick has been trading since 1985. He started out as a floor trader on the Sydney Futures Exchange and has spent the last 30-plus years developing and managing systematic equity strategies through The Chartist. When he talks about diversification, he is not talking about spreading money across a few different stocks. He is talking about building a portfolio of strategies, each designed to perform in different market conditions, so that the equity curve stays smooth even when individual strategies go through their inevitable rough patches.
This episode breaks down his layered approach to diversification, with real equity curve data showing how combining strategies with low correlation can smooth out performance even when both are long-only and trading the same asset classes.
Watch the full episode below, then read on for the complete breakdown.
The beginner cycle: why most traders sabotage themselves
Nick has a name for a pattern he sees constantly, especially with newer traders. He calls it the beginner cycle.
It works like this. A trader finds a strategy, starts trading it, and after two or three months of flat or negative returns, concludes it is not working. They switch to something else. That new strategy also goes through a flat period. They switch again. Each time they flip, they miss the recovery of the strategy they just abandoned.
He showed a real example from one of his own strategies, a short-term mean reversion approach trading the Russell 1000 with an average hold of about three days. From January 2015 to point one on the chart, the strategy returned around 30% net. From point one to point two, it gave almost everything back. A trader starting at the peak, or worse, joining near the top and riding it down to point two, would be feeling pretty beaten up at that stage and thinking about quitting.
Then 2020 arrived. The strategy returned 31% or 32% that year. The person who quit at point two missed all of it. The person who stuck with it through the flat period was rewarded. Nick’s line on when to start a strategy is worth remembering: “The best time to start is 20 years ago. The second best time is now.”
What diversification actually means in practice
Most traders understand diversification in theory. Most still trade with too much concentration in practice. Nick’s approach is built around one core idea: instead of trading a portfolio of stocks, trade a portfolio of strategies.
Each strategy will go through periods of poor performance. That is unavoidable. But if you have two strategies that perform well in different conditions, the combined equity curve should be smoother than either strategy in isolation. The math is not complicated, but actually building and maintaining multiple uncorrelated strategies takes real discipline.
Nick focuses primarily on Australian and US equity markets, long-side only. Short selling in Australia was severely restricted following the global financial crisis (GFC), and borrowing stock remains difficult. But he is clear that this long-only constraint does not mean the strategies are correlated with each other. As he demonstrates, two long-only strategies can have correlation as low as 0.22 when they respond to market conditions differently.
Mean reversion plus trend following: a concrete example
The clearest illustration from the episode involves two specific strategies placed side by side across the same time period.
The first is the US high frequency strategy: a short-term mean reversion approach on Russell 1000 stocks with an average hold of around three days. The second is an ASX trend following strategy with an average hold of about nine months.
Over the period shown, both strategies produced similar total returns. But the journeys were entirely different. The mean reversion strategy performed well during the flat-to-volatile periods that frustrated trend followers. The trend following strategy held up during periods where mean reversion got hammered. Together, they produced a much smoother equity curve than either would have in isolation.
During the GFC specifically, the trend following strategy switched to cash from early 2008 through mid-2009. Clients pushed back hard at the time. Why were they paying for a strategy sitting in cash? Nick’s answer came in late 2008 when the market had fallen 50% and his clients had lost only about 12%. Recovery from a 12% drawdown is manageable. Recovery from 50% is a different psychological challenge entirely, and many of the people who suffered that loss were unable to act when the recovery came.
Meanwhile, the mean reversion strategy was thriving during that exact period. High volatility is what short-term mean reversion likes. The wild daily swings of 2008 were fuel for the system. So while the trend portfolio sat in cash, the mean reversion side was generating returns. The combination worked because the two strategies liked different environments.
How to diversify across markets and styles
Nick trades Australia and the US, and he applies this same logic across multiple dimensions:
- Geography: Australian equities and US equities behave differently, have different liquidity profiles, and are denominated in different currencies. This geographic split adds a layer of structural diversification that correlation alone does not fully capture.
- Strategy style: Trend following and mean reversion are almost opposite in their market preferences. Trend following needs sustained directional moves. Mean reversion needs volatility and pullbacks. Combining them means you are rarely entirely in the wrong environment.
- Time frame: A three-day average hold and a nine-month average hold respond to market news and regime changes at completely different speeds. This difference in time frame is itself a form of diversification.
He noted that the US market is extremely cheap to trade for short-term strategies, especially compared to Australia. Commission costs matter a lot when you are holding positions for three days and trading frequently. This practical consideration shaped his market selection.
The case against timing your strategies
One of the traps Nick sees traders fall into is attempting to time when to switch strategies on and off. The question he gets constantly is: “What’s the best time to start?” The answer, as he puts it, is always the same. The best time was 20 years ago.
Trying to wait for a strategy to be “on an upswing” before committing to it is a version of the same mistake as abandoning a strategy during a flat period. You do not know when the upswing will start. After 35 years in markets, Nick says he has never met anyone with a reliable ability to predict what the market will do in the next week, two weeks, or two months.
The correct response to a strategy going through a difficult period is not to time your exit and re-entry. It is to have a second strategy that performs well precisely when the first one is struggling. Let the portfolio do the work of smoothing the returns, rather than trying to manually manage each strategy’s active periods.
Allocations: fixed and unmanaged between strategies
One practical question that comes up is whether to shift capital between strategies based on recent performance. If the trend strategy is sitting in cash, should that capital be reallocated to the mean reversion strategy?
Nick’s answer is no. Each strategy gets a fixed allocation, and that allocation stays through thick and thin. When the trend strategy was sitting in cash during the GFC, the cash position was part of the strategy design. That was not idle capital; that was the strategy doing exactly what it was supposed to do. Shifting that capital to the mean reversion side based on short-term performance would have disrupted the intended risk profile and introduced a market-timing decision that the system was never designed to make.
Sitting in cash is a position. For beginning traders especially, this is counterintuitive. The urge to always be doing something is strong. Sometimes the most productive thing a systematic trader can do is wait.
What correlation should you target between strategies?
The correlation between Nick’s US high frequency strategy and his ASX growth portfolio is 0.22. Both are long-only equity strategies. That low correlation across two ostensibly similar strategies demonstrates how much variation in correlation is possible through differences in time frame, market, and style.
He does not dynamically adjust allocations based on rolling correlation. The correlation figure is used during strategy design and portfolio construction, not as an active management signal. Once the portfolio is built and the allocations are set, he stays with those allocations and lets both strategies run.
His target for students in his trading system mentor course is straightforward: build at least two systems that are not correlated. Get that working before adding more. A simple, well-designed two-strategy portfolio will outperform most traders who have been jumping between single strategies for years.
Related episodes
- Market regime techniques for systematic traders with Cesar Alvarez
- Nick Radge on the best trading systems and what makes successful traders
- 6 ways to detect a failing trading strategy with Kevin Davey
- How to profit and protect in any market with Tom Basso
Want to build a more resilient trading portfolio? Subscribe to the Better System Trader podcast for weekly interviews with the world’s top systematic traders and quantitative researchers.
