The Kelly criterion is one of those ideas that sounds like a cheat code the first time you hear it: a formula that tells you the exactly optimal amount to bet to grow your money as fast as mathematically possible. It's real, it's used by professional gamblers and some legendary investors — and if you apply it naively to trading, it will hand you position sizes that can cut your account in half on a normal losing streak. Understanding both halves of that sentence is the point of this guide.
What Kelly actually calculates
Kelly answers one question: given an edge, what fraction of my capital should I put at risk to maximize long-run growth? For a trade with a win rate and a payoff ratio (how much you make on a win versus lose on a loss), the formula is:
Kelly % = W − (1 − W) ÷ R
where W is your win probability and R is your payoff ratio (average win ÷ average loss). Say you win 55% of the time and your winners are twice your losers (R = 2). Kelly = 0.55 − 0.45 ÷ 2 = 0.325, or 32.5% of your capital. The Kelly calculator does this instantly and also shows the half and quarter figures — which you'll want in a second.
Why full Kelly is too much for real trading
Betting 32.5% of your account on a single trade should make your stomach drop — and it should. Full Kelly is optimal only under assumptions traders never actually meet: that you know your win rate and payoff exactly, that they never change, and that you can stomach enormous swings. In reality your edge is an estimate, and if you overestimate it, full Kelly doesn't just reduce your growth — it can wreck you. Even when your numbers are right, full Kelly produces gut-wrenching drawdowns; a 50% account drop is a routine event, not a disaster.
The practical answer: bet a fraction of Kelly
This is why serious practitioners use half-Kelly or quarter-Kelly. Here's the beautiful part of the math: betting half of Kelly gives up only about a quarter of the growth rate but roughly halves your volatility and drawdowns. You sacrifice a little upside for a huge improvement in survivability — and survivability is the whole game, because you can't compound an account you've blown up.
In our example, full Kelly said 32.5%. Half-Kelly is ~16%, quarter-Kelly ~8% — still aggressive by most standards, which tells you something: if Kelly is handing you numbers far bigger than the 1–2% per trade you'd size with a normal risk-based approach, that's a signal your estimated edge is probably too optimistic. Treat Kelly as a ceiling and a sanity check, not a target.
The most useful thing Kelly tells you
Honestly, the single most valuable output of the Kelly formula isn't the exact percentage — it's the sign. If Kelly comes out negative, you have no edge, and the optimal bet is zero. No position sizing trick rescues a strategy with negative expectancy; the only winning move is not to play. So before you argue about half versus quarter Kelly, use it to confirm you have a real edge at all. Pair it with expectancy to see the full picture of whether a strategy makes money.
Run your own numbers in the Kelly criterion calculator, look at the half and quarter figures rather than full Kelly, and treat anything it tells you above your normal risk budget with healthy suspicion.