Trade Win $ Loss $ Win Score Loss Score

1 5,000 1
2
3 12,000
5,000
1
1
Kelly C
K% = W-
4 9,000 1
5 17,000 1
6 6,000 1
7 5,000 1
8 5,000 1
9 7,000 1
10 4,000 1
11 3,000 1
12 6,000 1 Based on your past trading performance,
13 10,000 1
14 10,000 1 The Inputs to the Kelly Criterion are as fol
15 12,000 1
16 4,000 1 W = The winning probability facto
17 5,000 1 (1-W) = The Losing Probability Factor
18 6,000 1 R = The Win / Loss Ratio
19 6,000 1
20 4,000 1
21 5,000 1 W
22 5,000 1 (1-W)
23 12,000 1 R
24 9,000 1
25 17,000 1 Kelly %
26 6,000 1
27 5,000 1
28 5,000 1 The Kelly Criterion has returned a figure o
29 7,000 1 of their capital on their next trade, in orde
30 4,000 1
31 3,000 1
32 6,000 1
33 10,000 1
34 10,000 1
35 12,000 1
36 4,000 1
37 5,000 1
38 6,000 1
39 6,000 1
40 4,000 1
41 4,000 1
42 3,000 1
43 6,000 1
44 10,000 1
45 10,000 1
46 12,000 1
47 4,000 1
48 5,000 1
49 6,000 1
50 6,000 1
51 3,000 1
52 6,000 1
53 10,000 1
54 10,000 1
55 12,000 1
56 4,000 1
57 5,000 1
58 6,000 1
59 6,000 1
60 10,000 1

$ $ W 1-W W (1-W)
272,000 148,000 30 30 0.50 0.50
 Kelly Criterion Win %
Loss %

K% = W-[(1-W)/R]
Win $
Loss $
R
Kelly %

your past trading performance, the Kelly Criterion tells you the position sizes you should be taking on your next trade.

to the Kelly Criterion are as follows

The winning probability factor / the probability a trade will be a winning trade
The Losing Probability Factor / the probability that a trade will be losing
The Win / Loss Ratio

= (30 winning trades/60 Total Trades) = 0.50

= (1-0.50) = 0.50
= 272,000 / 148,000 = 1.84

= 0.50 - [(1-0.50)/1.84]
= 22.79%

Criterion has returned a figure of 22.79% meaning based on this traders past performance they should be staking 22.79%
pital on their next trade, in order to maxmise their performance.
 50.00%
50.00%
$272,000
$148,000
1.84
22.79%

taking on your next trade.

hey should be staking 22.79%

 Kelly Criterion
K% = W-[(1-W)/R]

>On the surface, this traders performance may seem mediocre. The tr
>The total $ profit on winning trades is $272,000. The total $ loss on lo
>Therefore, on winning trades, this trader has made 1.84 times what t
>Even though this trader only makes money on half their trades, this t
>It can be argued that this is a good performance, given that the trade

If R remains constant K increases as a traders W increases i.e. The reco

If R remains constant K decreases as a traders W decreases i.e. The rec

If W remains constant K Increases as a traders W/L Ratio (R) improves

If W remains constant K decreases as a traders W/L Ratio (R) worsens
 Win % 50.00%
Loss % 50.00%
Win $ $272,000
Loss $ $148,000
R 1.84
Kelly % 22.79%

mediocre. The trader has made money on 30 trades and lost money on 30 trades
e total $ loss on losing trades is $148,000
1.84 times what they lose on losing trades.
heir trades, this trader has made a profit of ($272,000 - $148,000) = $124,000 over the last 60
ven that the trader only achieves a 50% winning percentage.

eases i.e. The recommendation is to stake larger amounts if a trader is right more of the time
reases i.e. The recommendation is to stake smaller amounts if a trader is right less of the time

atio (R) improves i.e. The recommendation is to stake larger amounts as a trader makes more
Ratio (R) worsens i.e. The recommendation is to stake smaller amounts as a trader makes less
00 over the last 60 trades.

more of the time

ht less of the time

ader makes more money from each trade.

trader makes less money from each trade.
 >We can now reverse the causality of the
Win % 30.00% >Instead of relying on 60 of our past trade
Loss % 70.00% vast majority of our trades wrong and stil
Win $ $150,000 >We can apply a 1-3 rule in which we stat
Loss $ $50,000 >For example a 15% stop loss requires a 4
R 3.33 >By applying a 1-3 rule we can now afford
Kelly % 9.00% >If we apply this 1-3 trading rule to every
must always be 3.33
>In this scenario, the Kelly Criterion tells u
positions at any time.

Win % 25.00% >What if we got greedy and squeezed the

Loss % 75.00% order to break even?
Win $ $166,500 >Here, we are applying a 1-4 rule or [(1/4
Loss $ $50,000 >If we apply this 1-4 trading rule to every
R 4.00 must always be 4.
Kelly % 6.25% >In this scenario, the Kelly criterion tells u
of 16 positions at any time.
 e causality of the formula.
0 of our past trades, we can work in reverse by establishing a trading dynamic which allows us
es wrong and still make money.
in which we state our minimum target returns on every trade will be 3X greater than our stop
p loss requires a 45% target return.
we can now afford to get only 30% of trades right to break even or [(1/3.33) X 100] = 30%. W =
ding rule to every trade, then R must be 3.33 because if we fix R at 3.33 then our $ ratio of Win

ly Criterion tells us that each position size should be 9% as a % of our capital or a maximum o

and squeezed the level of R even higher? To a point where we only had to get 25% of our trad

a 1-4 rule or [(1/4) X 100] = 25%

ding rule to every trade, then R must be 4 because if we fix R at 4 then our $ ratio of Winners t

ly criterion tells us that each position size should be 6.25% as a % of our capital or a maximum
c which allows us to get the

ater than our stop losses.

X 100] = 30%. W = 30%.

our $ ratio of Winners to Losers

or a maximum of 11

t 25% of our trades correct in

ratio of Winners to Losers

ital or a maximum