VALYNOXSTRATEGY RESEARCH TECHNOLOGY
AlphaPrimeAlphaPrime
Free trial · Download
Free tools / Risk and performance

Kelly criterion calculator

The Kelly formula gives the share of your account to risk per trade that grows it fastest over time, if your win rate and win/loss ratio are right. Enter yours to see full, half and quarter Kelly.

Runs in your browser. Nothing you type is uploaded.

Result

Full Kelly (risk per trade)
25%
Half Kelly
12.5%
Quarter Kelly
6.25%
Expected growth per trade, full Kelly
4.569%
Expected growth per trade, half Kelly
3.443%
Win rate needed to break even
40%

Turn on JavaScript to recalculate. The results below are for the example values.

How it works

More risk is not always more growthExpected growth per trade by risk per trade (win rate 55%, ratio 1.5)Half KellyKelly2× Kelly: no growthRisk per trade0Past the peak, more risk only adds drawdown.

The formula

Kelly % = W − (1 − W) ÷ R, where W is the win rate and R is the average win divided by the average loss.

Example: a 55% win rate with wins 1.5 times the size of losses gives 0.55 − 0.45 ÷ 1.5 = 25%.

What the percentage means for futures

It is the share of the account you lose when a trade reaches its stop, not the margin. Combine it with the position size calculator: risk % in, contracts out.

Why traders use half Kelly or less

Your win rate and ratio are estimates, usually from a backtest, and they are rarely as good live. Betting more than Kelly is far worse than betting less: at twice Kelly the expected growth falls to about zero, while half Kelly keeps about three quarters of the growth with much smaller drawdowns.

Questions

Is the Kelly criterion good for trading?

It is a useful ceiling: it tells you how much risk your numbers can bear. Because the numbers are uncertain, most traders who use it risk a half or a quarter of the Kelly amount.

Can I use Kelly with a win rate under 50%?

Yes, if the wins are large enough. A 40% win rate with wins twice the size of losses gives 0.40 − 0.60 ÷ 2 = 10%.

Why is my Kelly percentage so high?

Usually because the win rate or ratio is too optimistic. Backtest numbers include luck and leave out some costs. Use numbers from out-of-sample tests or live trading, and still bet less.

What does a negative Kelly mean?

The strategy has no edge with these numbers: on average it loses money, and the formula says to risk nothing.

Test the idea on years of data, not one trade

AlphaPrime generates and backtests futures strategies on your GPU, with contract specs, trading costs and sessions built in, then exports the strategy code for your trading platform.

Free 7-day trialSee how it works

For education. Not investment advice. Check the exchange before you trade.