Imagine a simple
coin-tossing game where you win whatever you stake if heads comes up, lose
what you stake if tails comes up, and you are charged 1% of your stake
each turn to play. Can you win money at this game? If you are familiar
with the concept of expectancy, then you will probably answer ‘No’ since
over many turns the amount won will be equal to the amount lost (assuming
the coin is a fair one) and after factoring in the 1% cost of playing you
will lose money overall.
In fact, there is a way to win this game, and that is to understand that
the longer you play, the more you will lose, so the optimum strategy is to
bet everything you have on just one toss of the coin; just like Ashley
Revell did when he sold everything he owned, took the $135,300 to Las
Vegas, and bet it all on ‘Red’ on one spin of the roulette wheel. Mr.
Revell was fortunate and he won, but I am not recommending that you bet
everything you have on one trade!
Obviously risking everything on one trade is not a useful strategy since
we want a game we can play for long periods of time to generate a
consistent income. So how can we change the game so that we can win? There
are three aspects to the game which can be adjusted to increase our
chances of winning consistently:
• We can tip the chance of a winner in our favor from 50/50
• We can increase the size of the payout from 1:1
• We can reduce the cost of playing the game
Tipping the chances of a
winner is not possible in a fair coin toss game, but it is possible in
trading. There are two ways to approach this: identify conditions that are
more favorable to your winners and include them in your system definition,
or identify circumstances where a loser is more likely, and skip those
trades. For example, if you notice that most of your winners are entered
on days where the overall market has moved in the same direction as your
trade, then only enter trades when the overall market is moving in the
correct direction. This means that your trade is in the same direction of
the overall market, rather than against it.
Another example might be that trades that are entered just before major
news announcements, like earnings calls, often get stopped out as losers
due to increased volatility, so you should skip those trades.
There may be many patterns of winners and losers that you can identify for
your own systems and careful study of past trades is definitely
worthwhile. Note that we do not want to increase our win percentage too
significantly (i.e. to greater than 60%) since this would indicate that we
have ‘curve-fitted’ our system to historical results that are unlikely to
continue into the future.
It is also important to note that for some types of trading (i.e.
long-term trend following strategies) it may not be possible to have a win
percentage that is greater than 50% (and it may be much lower) and that is
where the second aspect of improving your system comes into play: the
average size of winners versus losers.
Increasing the size of the payout so that the winners win more on average
than the losers lose depends on the way you handle your stops. Having
large winners in relation to losers can make up for a low win percentage,
and mean that you will still make money playing the game. One method is to
have a trailing stop that moves up as a trade becomes a winner. If you
have fixed stops for losing trades that limit losses, but trailing stops
for winning ones that allow winners to grow, then you are increasing your
chances of your average winner being larger than your average loser.
Generally it is better to be strict on losers by having tighter stops that
keep losses to a minimum and generous with winners by having stops that
allow profits to grow. In any case you want to make losers small and
winners large, so never add to a losing trade – that would be doing the
opposite of what you want to achieve.
Lastly, reducing the costs of trading is probably the simplest change you
can make, and can mean the difference between winning and losing overall –
especially for systems that have lower expectancy. There are many online
brokers now that charge 1c per share for equity trades (and comparably low
fees for other instrument types) and there is no reason why you should be
paying more than this if you are trading electronically.
Every trader should do whatever they can to maximize the expectancy of
their trading system or method by considering each of the 3 aspects just
described. If we do some, or all, of these things then the amount we win
now becomes a
factor of how much we stake, and how often we play because we have created
a true ‘edge’ where we know that the system we are trading should make
money (if traded accurately). Calculating the expectancy of your trading
system or method
tells you whether you are playing a game you can win, and is a very
important piece of information that every trader should know before they
risk real money.
If the game is rigged against you because your trading methods lose money
regardless of how accurately you implement them, how can you ever be a
successful trader?
.gif)
