Every quant who has ever backtested a strategy knows the gut-punch moment: the gross information ratio looks spectacular, then you factor in transaction costs and the whole thing collapses like a soufflé in a thunderstorm. This question — how do I know if my alpha actually survives contact with real markets — is genuinely one of the hardest problems in applied quantitative finance.
The direct answer is this: your net information ratio equals your gross IR penalised by the drag from turnover multiplied by your all-in cost per unit of trading. Grinold and Kahn formalised this elegantly. If your strategy turns over 200% annually and one-way costs on ASX mid-caps run 40–60 basis points including market impact, you need a gross IR of roughly 0.8 just to post a net IR of 0.5. Most strategies fail this test quietly.
Think of it like running a food truck. Your gross margin on each burger looks healthy. But once you account for the cost of driving to every market, parking fees, and the time spent setting up, your net margin tells a completely different story. Turnover is your fuel bill. A strategy that trades constantly is burning cash on every lap, and ASX execution — with its relatively wide spreads in the small-cap universe — is an expensive circuit.
The practical construction involves three inputs: your forecast IC and breadth (giving gross IR via the Fundamental Law), your strategy's expected annual turnover expressed as a fraction of portfolio value, and a realistic all-in cost estimate per trade. For ASX large-caps, 20–30 bps one-way is defensible. For small-caps, 50–80 bps is more honest once slippage and spread are included. Running sensitivity tables across these parameters, rather than picking single-point estimates, is how professionals stress-test a strategy before committing capital. For deeper grounding in the underlying theory, the information ratio methodology on Investopedia is worth revisiting, and the formal treatment of IR construction on Wikipedia clarifies the variance decomposition. The mechanics of how market friction compounds are also well-explained in the transaction costs overview on Investopedia.
Today's takeaway: open your best backtest, calculate your annualised turnover, multiply by your honest round-trip cost estimate, and subtract that drag from your gross alpha. Whatever survives is what you actually own.
Gross IR is vanity. Net IR is sanity.
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.