Pairs trading sounds beautifully simple: find two stocks that move together, wait for them to diverge, bet on reversion. But anyone who has tried this on ASX-listed mining stocks during a commodity cycle shift knows the strategy has a nasty habit of blowing up exactly when you're most confident it's working. That tension is worth unpacking carefully.
The core mechanic relies on cointegration — a statistical relationship where two price series share a long-run equilibrium even if they drift short-term. Think of it like two dogs on a shared lead. They wander, but they can't stray too far from each other. The lead is the cointegrating relationship. Pairs traders profit from the stretch and snap. The problem? Commodity cycles can quietly cut that lead without anyone noticing until it's too late.
On the ASX, classic pairs candidates include large-cap iron ore producers or diversified miners with overlapping commodity exposures. Historically, companies like these exhibit cointegration because they're driven by the same underlying price — iron ore, copper, gold. When that commodity reprices structurally, not cyclically, the fundamental driver of the cointegrating relationship itself changes. That's a structural break, not a trading opportunity.
The chart above illustrates the core risk elegantly. Left of the break, the spread oscillates around zero — textbook cointegration. Right of the break, it trends away relentlessly. A trader holding a mean-reversion position post-break isn't waiting for reversion; they're just losing. Detecting breaks in real time requires rolling Engle-Granger tests or the Johansen procedure run on fresh windows — not just a one-time setup test done at strategy inception. For deeper grounding, Investopedia's cointegration explainer covers the statistical foundations clearly, while the Wikipedia entry on pairs trading traces the strategy's origins and variants. For the academically inclined, the Wikipedia article on structural breaks outlines the formal testing framework traders adapt for live monitoring.
The practical takeaway is straightforward: never treat a cointegrating pair as a permanent relationship. Re-test quarterly at minimum, track the spread's half-life, and define a maximum holding period so you're not still waiting for reversion that structurally cannot happen.
In pairs trading, the only thing more dangerous than a widening spread is a widening spread you've convinced yourself must snap back.
This content is for educational purposes only and does not constitute financial product advice. Past performance is not indicative of future results. Profit Logic Ltd (ACN 688 669 936) accepts no responsibility for errors or omissions in this content or anywhere on this website. Always seek advice from a licensed financial adviser before making investment decisions.