Keel's Kelly criterion calculator turns your win rate and your average win and loss into the share of your account to risk on each trade. It shows full Kelly and the half and quarter Kelly most traders use, because full Kelly assumes you know your edge exactly.
From your strategy backtest. Example: 42 means 42% of trades close in profit.
Enter it as a positive number.
Full Kelly maximizes growth but is psychologically punishing. Half-Kelly is the common practical default.
How likely is a deep drawdown at that size? Check the risk of ruin at 4.89% a trade.
The Kelly criterion answers: given a known edge, what fraction of bankroll bet maximizes long-run logarithmic growth?
f* = (b × p − q) / b
p = probability of winning
q = 1 − p
b = avg_win / avg_loss (positive)The result f* is your bet size as a fraction of bankroll for each trade. Compounding at f* gives the highest expected long-run wealth, but the path there is extreme. Full Kelly drawdowns can hit 50%+ even when the underlying edge is positive; the variance of returns scales with f.
Why half or quarter Kelly: Kelly assumes you know p and b exactly. In real trading, you estimate them from limited samples. Overestimate the edge and full Kelly bets more than the true Kelly size; at twice the true size the account stops growing, and beyond it shrinks. Half Kelly is much more forgiving. Half Kelly keeps about 75% of full Kelly’s growth with about a quarter of the variance.
Use this calculator with backtested inputs: the win rate, average win and average loss from a backtest are the right numbers to feed in, not gut-feel estimates.
A backtest shows a 55% win rate, an average win of $150 and an average loss of $100. Then p = 0.55, q = 0.45 and b = 1.5, so full Kelly is (1.5 × 0.55 − 0.45) / 1.5 = 0.25: risk 25% of the account on each trade. Half Kelly is 12.5% and quarter Kelly 6.25%. Even the quarter Kelly figure is far above the 1% a trade many traders use, which is why Kelly works best as a ceiling.
To turn the Kelly fraction into a position, combine it with your stop in the position size calculator, and check how deep the losses can run with the max drawdown calculator.
Keel is a strategy builder for Hyperliquid. Describe an idea, backtest it with fees, slippage and funding, and run it live on Hyperliquid’s crypto perps. Start free, and connect a wallet when you’re ready to trade.
The share of your account to risk on each trade that grows it fastest over many trades, given your win rate and your average win and loss. The formula is f* = (b × p − q) / b, where p is your win rate, q = 1 − p, and b is your average win divided by your average loss.
Take your win rate and your average win and loss from a long run of trades or a backtest, and enter them here. The result is the Kelly fraction: the share of your account to risk per trade. Most traders then bet a fraction of that, usually half or a quarter, and use a position size calculator to turn it into a position.
Most traders use between a quarter and a half of full Kelly. Full Kelly assumes your win rate and payoff are exactly right, and they never are. If you overestimate your edge, full Kelly bets too much and the drawdowns get very deep. Half Kelly keeps most of the growth with much smaller swings.
Betting a fixed fraction of the full Kelly size, such as half or a quarter. It gives up some long-run growth in exchange for smaller drawdowns and protection against an edge that is smaller than you think. This calculator shows full, three-quarter, half and quarter Kelly.
Your numbers have no edge: with that win rate and payoff, the strategy loses money on average. Kelly would tell you to take the other side, which in practice means do not take the trade. Check your win rate against the win rate you need to break even before you size anything.
Risking a fixed percent, say 1% a trade, ignores how good the trade is. Kelly sizes up when your edge is bigger and down when it is smaller. Many traders use Kelly as a ceiling and a fixed percent in practice, never risking more than half Kelly.
Use both. Size each trade by a fixed share of the account at the stop, and take the smaller of that and half Kelly. The fixed percent keeps any one loss small; the Kelly ceiling stops you from risking much when the edge is thin or uncertain.
From as many trades as you can get. A backtest returns your win rate, average win and average loss over the whole history you test. Run the same strategy over different periods to see how stable those numbers are before you size a live position on them.