Why use half- or quarter-Kelly?

Full Kelly is volatile and very sensitive to estimation error in W and R. Many practitioners bet a half or quarter of Kelly to reduce drawdowns.

What happens if I have no edge?

If W − (1 − W)/R is zero or negative, Kelly says bet nothing — the strategy has no positive expectancy at that win rate and payoff ratio.

How do I estimate W and R in practice?

Use your actual trade history: win rate and the ratio of average win to average loss, ideally over enough trades to be statistically meaningful.

Is Kelly the same as position sizing by % risk?

No — the position size calculator fixes risk at a chosen %; Kelly derives an 'optimal' % from your edge, which can be higher or lower than a fixed 1% rule.

Why does full Kelly feel too aggressive?

Full Kelly maximises long-run growth rate but with large swings; small errors in your W/R estimates are amplified, which is why half- or quarter-Kelly is commonly used instead.

Risk & Money Management

Kelly Criterion Calculator

Compute the Kelly criterion optimal position fraction from win probability and payoff ratio, with the safer half-Kelly and quarter-Kelly shown.

The Kelly criterion is a formula giving the fraction of capital to risk per trade that maximises long-run account growth for a known win rate and payoff ratio.

Full Kelly fraction

32.5%

Half Kelly: 16.25%
Quarter Kelly: 8.13%
Payoff ratio: 2.00 : 1

Use with caution. Full Kelly is highly volatile and very sensitive to errors in your win rate and payoff estimates. Most traders use half- or quarter-Kelly to reduce drawdowns.

Worked example

A strategy wins 55% of the time, with an average win of $150 and an average loss of $100.

Payoff ratio R: $150 ÷ $100 = 1.5
Full Kelly fraction: 0.55 − (0.45 ÷ 1.5) = 25%
Half-Kelly: 25% ÷ 2 = 12.5%
Quarter-Kelly: 25% ÷ 4 = 6.25%

Full Kelly suggests risking 25% of the account per trade; many practitioners would use the steadier half-Kelly (12.5%) or quarter-Kelly (6.25%) instead.

How this is calculated

The Kelly criterion gives the bet fraction that maximises long-run growth:

f* = W − (1 − W) ÷ R, where W is win probability and R = avgWin ÷ avgLoss is the payoff ratio.

Because full Kelly is volatile and sensitive to estimation error, we also show half-Kelly and quarter-Kelly, which most practitioners prefer.

When to use this calculator

Use this when you have enough trade history to estimate a win rate and payoff ratio with some confidence — typically dozens of trades of the same strategy, not a handful. Kelly turns those two numbers into the growth-optimal fraction of capital to risk.

Treat the full-Kelly output as a ceiling, not a recommendation. Because the inputs are estimates, most practitioners size at half- or quarter-Kelly, which gives up little long-run growth in exchange for much shallower drawdowns.

It is also a quick edge check: if the Kelly fraction comes out at zero or negative, the strategy has no positive expectancy at those inputs, and the correct position size is none at all.

Common mistakes

Feeding in an optimistic win rate or payoff ratio from a small sample, producing an inflated Kelly fraction.
Betting full Kelly in practice — the growth-optimal fraction is also the most volatile, prone to large drawdowns from estimation error.
Applying one Kelly fraction across unrelated strategies with different edges instead of calculating it per strategy.

Frequently asked questions

What is the Kelly criterion?: Kelly gives the fraction of capital that maximises long-run growth: f* = W − (1 − W) ÷ R, where W is win probability and R is the win/loss payoff ratio.
Why use half- or quarter-Kelly?: Full Kelly is volatile and very sensitive to estimation error in W and R. Many practitioners bet a half or quarter of Kelly to reduce drawdowns.
What happens if I have no edge?: If W − (1 − W)/R is zero or negative, Kelly says bet nothing — the strategy has no positive expectancy at that win rate and payoff ratio.
How do I estimate W and R in practice?: Use your actual trade history: win rate and the ratio of average win to average loss, ideally over enough trades to be statistically meaningful.
Is Kelly the same as position sizing by % risk?: No — the position size calculator fixes risk at a chosen %; Kelly derives an 'optimal' % from your edge, which can be higher or lower than a fixed 1% rule.
Why does full Kelly feel too aggressive?: Full Kelly maximises long-run growth rate but with large swings; small errors in your W/R estimates are amplified, which is why half- or quarter-Kelly is commonly used instead.