Episode 9: Testing our System

We now find ourselves at a point where we need to make a decision - which set of rules do we want to trade.

Transcript: 

'Successful Trading' - Episode 9: Testing our System

Presented by Jeff Bryant, HomeTrader.

Hello - and welcome to Episode 9 in our series on becoming a Successful Trader.

So far in the series we’ve looked at the elements of a trading plan and why we need one.  We have gone through examples illustrating various elements and the process as a whole.

We now find ourselves at a point where we need to make a decision - which set of rules do we want to trade.
   
Now, before I go any further, I am want to reiterate the Financial Services mantra:

Past performance is not a guarantee or a reliable indication of future results.

What is important is that beyond the legal aspects of this statement is its reality. 

Understanding that it is the fact is a key step forward along the pathway to successful trading.
    
Now, one aspect of a proper trading plan is the rules are unambiguous and therefore can be replicated.

You can look back in time and look to (say) June, 2003 at (for example) XYZ and see whether you might have got taken a position or not.  The answer will be cut and dried; yes or no. 

Further, if you had taken the trade, you can then see, again from your rules, how many you would have bought and when and where you would’ve got out.

In so doing – you can effectively build up what I refer to as the career of your plan. 

You can look back and imagine that you had got your trading float together and it’s the evening of January 1, 2000 and you start trading.

You can build up the list of trades your system could have taken and then analyse the results. 
   
Imagine you have a set of rules that would have got you thru:

* the tech crash
* the 12 month period sideways that followed
* September 11
* the bear market run to the Iraq war
* the sub-prime correction
* and what will be referred to for decades as simply, 2008.

If your trading plan gets you thru all of those markets, you’d have to consider you’re in pretty good shape - certainly a lot better off than your investor mates who were by early 2009 down, on average, a MASSIVE 55%. That’s on average!  Quite a number were down a lot more than that! And they are still down more than 20% over 2 years after it started!
    
The concept is pretty simple.  You imagine you have a trading float, say $20000 & it’s January 1, 2000 and you start scanning the market. Night after night you build up a list of eligible trades.

You then literally look back over that list to see how your plan is likely to have performed.
`    
Sounds easy enough… but to do this manually, trade by trade, would take ages! But, if your rules that can be programmed into a computer, you can use one of this decade’s great advances in trading – the Trading Simulator. A simulator can do what I’ve just been talking about in just seconds!

Here’s how they work. Firstly we program our rules so that the simulator can read them. I want to stress here – you don’t need to know to be able to do this. 
You can access this level of intellectual property through us at HomeTrader without having become a computer programmer.

Of course, for those who do want to go to this level – we do have programming packs, and User Groups are generally available to help you get started.

In this example I’ve set it to start from 2000 and I’ve set up a trading float of $20,000. Clicking Backtest gets it started. It then works its way from there looking for stocks that meet all entry conditions. It knows how much money it has because it can keep track of its cash position. Once it finds an entry, it knows how many shares it will buy, then looks for the exit signal.

Following this process, the simulator builds up a list of such trades – effectively creating what I refer to as the trading career of the system.

We can check this list – have a look at FUN for example. How do we know we got entry on June 17, 2003? Because our entry rules are unambiguous.

We can also confirm entry in 2699 shares at $1.44, and exit on Dec 4, 2004  at $2.17.

And of course with all of this information, we also have the profit or loss of the trades.

From this list we can then analyse all manner of information. We can plot the profits and from this answer a couple of key questions. We can see the worst drawdown over that entire period. In other words – would we have survived?  A quick look tells us that the worst we saw here was a drop of just over 16% - not too bad compared to the damage sustained by the average investor over 2008.

Then the profit itself.  Did we make enough to make it worthwhile? The average of around 25% per annum in our example is certainly better than the banks would have given us.

Other stats we might like to review include:
* Number of trades
* Win rate
* Average win
* Average loss
* Ratio of the two
* Ave trade size
* Ave days in trades

I could go on, but you get the idea.

Now we come back to a point I raised earlier. You are looking to make a decision as to how we will approach the market.

You have some choices:

You can take advantage of all that backtesting can offer us to help us decide whether our rules work or not,

OR -

You can take someone’s word for it – and it would want to be someone you trust!

OR –

You can just plain guess and hope for the best!
   
I certainly want some idea how my rules might have gone BEFORE I would trade them and that’s why I prefer to trade a method that I can backtest.

In other words, I want to know it works - before I start throwing money at it all. I know that my future trading will be different – that’s a given. But in making a decision in the present and having the confidence to follow it through in the future - there’s no substitute for Back-testing as far as I’m concerned. Next time, we’ll look at some more advanced strategies to add to our trading arsenal

In the meantime, if you’d like further information, you can visit our website at hometrader.com.au, and book in to one of our free 2-hour information seminars.