Every trader eventually asks some version of this: "If my edge is positive, why can't I just bet as much as possible?" It sounds almost insultingly simple. But this question sits at the heart of one of the most important — and most misunderstood — ideas in quantitative finance. The answer will genuinely change how you think about position sizing forever.
The core tension is this: expected value (EV) maximisation tells you to take the biggest possible bet on any positive-edge proposition. Kelly Criterion says that's a catastrophic mistake. Both frameworks are mathematically rigorous. They simply optimise for entirely different things — and in real markets, that difference is the gap between compounding wealth and blowing up your account.
Here's the analogy that makes this click. Imagine a coin flip where heads pays 2x your bet, tails you lose your stake. EV says bet everything — expected return is 50% per flip. But after just two flips, heads then tails, you're back to zero. You can't play again. The EV calculation ignores that ruin ends the sequence permanently. This is path dependency in its rawest form.
John Kelly's 1956 paper reframed the entire problem. Instead of asking "what maximises my expected return this bet?", Kelly asked "what maximises the expected logarithm of my wealth over time?" That small shift is profound. The log function penalises losses far more than it rewards equivalent gains — exactly mirroring how real account growth works. The Kelly fraction formula, f* = (bp - q) / b, where b is the net odds, p is win probability, and q is loss probability, produces the bet size that maximises long-run geometric growth. Bet more than Kelly and you grow slower. Bet double Kelly and you go broke with probability one, regardless of edge. Practitioners often use fractional Kelly — commonly half Kelly — to reduce volatility while preserving most of the growth advantage, as documented extensively in Investopedia's Kelly Criterion guide. The deeper mathematics behind optimal growth strategies trace back to information theory, and the Wikipedia entry on Kelly criterion covers the derivation cleanly. For traders wanting the original source, John Larry Kelly Jr.'s biography contextualises why a Bell Labs scientist was solving gambling problems in the first place.
The practical takeaway is simple: calculate your edge, run the Kelly formula, then consider betting half that fraction. Your account survives bad runs. Survival lets you compound.
EV tells you whether to play. Kelly tells you how much. You need both — but only one of them keeps you in the game long enough to matter.
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.