Kelly criterion

How much of the bankroll to put behind an edge, what that costs you in growth if you size it differently, and how deep a hole it can dig on the way.

%

Above the 50.0% these odds imply, so there is an edge.

×

A win returns your stake plus 100 per 100 staked.

%

Share of the stake you give up. 100% is an ordinary bet.

Only used to turn the fraction into an amount.

50%

Most people who use this bet a fraction of it.

Stake this much
5.00%
500 of a 10,000 bankroll
Full Kelly
10.00%
1,000 at full size
Edge per unit staked
+10.00%
fair odds here are 50.0%, you say 55.0%
Growth per bet
+0.38%
75% of the best possible rate
Bets to double the bankroll
185
at this stake, compounding
Chance of ever losing 50%
12.5%
down to 5,000 at some point
Stake, growth rate and drawdown risk at several fractions of Kelly
Fraction of KellyStakeGrowth per betOf the best rateEver down 50%
Full10.00% · 1,000+0.50%100%50.0%
Three quarter7.50% · 750+0.47%94%31.5%
Half5.00% · 500+0.38%75%12.5%
Quarter2.50% · 250+0.22%44%0.8%
Tenth1.00% · 100+0.10%19%0.0%
One run of bets, at three different stakes
Yours — 50% of Kelly
10,000
staking 5.0%
Full Kelly
10,000
staking 10.0%
3× Kelly — over-betting
10,000
staking 30.0%
1002005001,0002,0005,00010,00020,00050,000 wiped out — 1% of where you started 050100150200250 bets placed

Every stake plays the identical sequence of wins and losses, so anything that separates them on screen is the sizing and not the luck. The run ends when the last one is busted, or after 250 bets.

Outcome of each stake over the same sequence of bets
StakeOf bankrollHighest it reachedAt the end
Yours — 50% of Kelly5.0%31,64826,864
Full Kelly10.0%55,21838,668
3× Kelly — over-betting30.0%13,000wiped out at bet 105

What a bad run looks like

Betting 50% of Kelly, there is a 12.5% chance the bankroll is down 50% at some point along the way, even while it grows in the long run. Full Kelly has a coin flip's chance of halving at some point, which is why almost nobody bets it straight.

down 50%

Kelly maximises the expected log of the bankroll rather than the bankroll itself: g(f) = p·ln(1 + f·b) + q·ln(1 − f·a), which peaks at f* = p/a − q/b. With a full loss on the downside that is the familiar (p·b − q)/b. Maximising expected money instead would tell you to stake everything on any favourable bet, which goes to zero almost surely.

The drawdown numbers come from the diffusion approximation, where the chance of ever falling to a level α while betting a multiple k of Kelly is α^((2−k)/k). It holds when each bet is small relative to the bankroll, and gets optimistic when it is not.

The simulation stakes a fraction of whatever is left each time, so the bankroll is multiplied rather than reduced by a fixed amount and never actually reaches zero. It just keeps shrinking until a bet is not worth placing, which is why the run ends at 1% of the starting bankroll rather than at nothing. Every stake in it sees the identical sequence of wins and losses, so nothing separating the lines is luck.

The real weak point is not the maths, it is the first input. Kelly assumes you know the win probability, and a confident estimate that is a few points too high will happily hand you a stake well past the ruin line. Betting a fraction of Kelly is mostly insurance against your own estimate.

Trying to size the confidence in a number rather than a bet? The significance tool gives you the posterior on a conversion rate.