I ran the numbers three times. Kelly said risk 18% of my account on this setup. The edge was there — 60% win rate, solid R:R, clean backtest. I cut it to 9% because 18% felt insane. The trade won. Then I watched myself slowly creep the size back up over the next month, chasing that 'optimal' number. By week six, I was risking 15% per trade and my account was swinging like a pendulum. One bad week and I'd lost five weeks of gains.
The Kelly Criterion is mathematically perfect for maximising long-term growth — but it assumes you're a robot. The formula is simple: f* = (bp - q) / b, where b is your win/loss ratio, p is your win probability, and q is 1-p. Plug in a 60% win rate and 2:1 payoff, and Kelly tells you to risk 20% of your capital. The problem? Real traders have real emotions, real drawdowns, and real mortgages. A 20% position size means a string of three losses wipes out half your account. Most people can't stomach that — I certainly couldn't.
The turn came when I started tracking my actual behaviour versus my planned behaviour. I'd set a Kelly-based size, then unconsciously reduce it when I felt uncertain or increase it when I felt confident. I wasn't following the system — I was sabotaging it with gut feel. The realisation was brutal: if you can't execute the size consistently, the optimal size is irrelevant.
The solution isn't abandoning Kelly — it's using fractional Kelly. Half Kelly or quarter Kelly gives you most of the growth benefit with drawdowns you can actually survive psychologically. I now use quarter Kelly as my baseline, which typically puts me at 3-5% risk per trade depending on the setup. It's not mathematically optimal, but it's behaviourally sustainable. The Kelly Criterion works brilliantly in theory, but real trading requires accounting for risk management that keeps you in the game long enough for the edge to play out. Your position size should be the largest amount you can risk while still executing your system with zero hesitation — and for most humans, that's nowhere near what full Kelly suggests. Survive first, optimise second.
This content is educational only and does not constitute financial advice. Past performance is not indicative of future results. Always seek licensed financial advice before trading.