Ask a quant how much to bet on their best idea and watch the room go quiet. Position sizing is the unglamorous cousin of strategy development — everyone wants to find the next signal, nobody wants to do the maths on how much to stake. Yet getting this wrong can blow up a perfectly profitable strategy faster than any losing streak. The Kelly Criterion sits at the centre of this debate, and it's messier than the textbooks suggest.
The direct answer is this: most institutional quant teams deliberately use half-Kelly or even quarter-Kelly sizing, and it is entirely rational. Full Kelly maximises long-run geometric growth in theory, but in practice it produces drawdowns so savage that no investor — human or institutional — will sit through them. A strategy sized at full Kelly can lose 50% of its equity in a perfectly normal bad run. Half-Kelly cuts that expected drawdown roughly in half while sacrificing only about 25% of the theoretical growth rate. That trade-off is almost always worth making.
Think of it like driving on a highway. Full Kelly says: drive at the absolute maximum speed your car can physically sustain on a perfect road. Half-Kelly says: back off 20%, because the road isn't perfect, your speedo might be slightly wrong, and a blown tyre at top speed is catastrophic. The institutional quant adds another layer — they're driving someone else's car with passengers who will demand you pull over the moment things get uncomfortable. Volatility tolerance isn't just mathematical; it's behavioural and contractual.
There's a deeper problem that makes the full-Kelly debate almost academic: you never actually know your edge with certainty. The Kelly formula requires precise inputs — your true win rate, your true payoff ratio. In live markets, both are estimates, often optimistic ones fitted to historical data. Research published in the Journal of Portfolio Management has long documented that traders systematically overestimate edge, meaning the "correct" Kelly fraction is always smaller than you calculate. Fractional Kelly acts as a built-in humility tax. For anyone wanting the mathematical foundation, the Kelly Criterion explained on Investopedia is a clean starting point, while the deeper probability theory sits in Wikipedia's Kelly criterion entry for those who enjoy staring at logarithmic utility functions. The broader framework of optimal f position sizing on Investopedia offers a practical companion concept worth understanding alongside Kelly.
Start today by calculating your full-Kelly fraction on any active strategy, then immediately halve it. Run it that way for 90 days and compare your sleep quality to the prior quarter.
The best position size isn't the one that maximises theoretical returns — it's the one you can actually hold when markets go sideways.
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.