As traders we must feel comfortable living life on the right edge of the chart, where what happens next is unknown.
As I mentioned many times in earlier posts, attaining a proper mind-set is key to long-term success. That said, a key mental characteristic of professionals is confidence; confidence that their systems will be profitable, confidence to pull the trigger despite not knowing how the next trade will turn out and confidence to trade their system with the utmost discipline even when faced with many consecutive losses.
A bit of a paradox if you ask me. How on earth can someone have confidence in a future outcome when, by definition, the future is unknown? That’s the mark of a true professional…to be able to march confidently in the face of uncertainty.
Suffice it to say that most people do not start out with that degree of confidence in the markets and most are not willing to do the necessary legwork to acquire it either. But that doesn’t have to be you. Take the advice in this article to heart and you’ll be trading with the mind-set of a pro in no time flat!!
Since no one can know what will happen in the future the only benchmark you have to work with is what’s happened in the past.
I am assuming that you already have a system, either one you purchased or one you’ve devised yourself, complete with entry, exit, stop-loss and risk parameters as we discussed earlier. So you’re armed with a system and have high hopes but what do you really know about the dynamics of your system? If you haven’t backtested it then you know nothing about it except its rules.
Too many traders get their hands on a system and simply assume it will work. They blindly place their trust in the system without truly knowing where they stand with it. This is why you MUST back test. By testing it you will give yourself an idea if the system can work in the future. It is the only fighting chance you have with a system before you put money on it. If it doesn’t work in testing, don’t expect it to work going forward either.
Whatever you do, PLEASE don't just take my word on any system suggestions i make in this blog! You have to make your own assessments, even on the ones you believe to be top-notch. I have saved myself fortunes by rigorously testing any system I get my hands on. And it is through this tedious, time consuming process you gain your confidence to trade a system, especially during rough times.
So, how exactly do you go about backtesting and what information do I need to get out of it?
First and foremost you have to get your hands on loads of historical data. Some advanced brokerage platforms provide it with their charting and other times you need a separate piece of software. Once you get it, this is the info you need to analyze from your backtested data:
1. Total number of trades you tested (I recommend a minimum of 100 trades over several market conditions)
2. Number of winners
3. Number of losers
4. Accuracy. (What percentage of trades turn out to be winners)
5. Value of winner. (What each winner was worth in dollars, on average)
6. Value of losers. (What each loser was worth in dollars, on average)
7. Maximum number of consecutive losses
8. Maximum number of consecutive wins
9. Maximum drawdown (what percentage did your account value lose from a previous equity peak)
10. The mathematical expectancy of your system – ABSOLUTELY ESSENTIAL CONCEPT
Now that I’ve told you why you have to backtest and what you have to get out of it, what about HOW? That is a good question to which there are a couple of answers. You can manually backtest or if your system is completely mechanical you can program it. I'll discuss each option and where to find the right resources for each in the next post.
The premise here is that even though past results are no guarantee of future performance, it’s fair to at least assume that if a system performed well over a long period of time in the past that the odds are favorable it will continue to please going forward. This is the way you acquire the necessary confidence to trade a system!
This is a crucial topic and I have lots more to say about interpreting your results, expectancy (what's that?!?), win/loss distributions (huh?!?!), software you can use to facilitate the process and actually make it fun. Until then, be well!
Thursday, August 28, 2008
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment