JTrader
Guest
- Messages
- 5,741
- Likes
- 507
Take a look at Forexfactory or forex-tsd.com to see how quickly auto systems come and go.
I have spent over 12K on the services of a very good programmer and learnt a few things in the process.
1. Learn to trade successfully before any thoughts of implementing an ATS otherwise there will be continuous tweaking and layering.
2. Backtesting is only useful in the implementation of the system ie checking the ATS can handle all eventualities. As far as basing future success on previous performance, it's a myth for a myriad of reasons.
3. If you can really define your strategy in terms of mechanical rules then there may be a chance of short term profit. A big 'if' though.
4. Don't blame the programmer for your own shortcomings. Programmers like things to be defined. Treating them like traders is a mistake.
5. Understand the complexity of the task. There's a lot more to it than just buying and selling. For example what if the order is only partially filled, or the stop gets missed, or there is a power cut, or the data feed gets corrupted, or the ATS gets stuck in a loop and keeps adding to positions? All these things I have experienced and worse.
6. You are competing against companies which employ the brightest and the best and have millions of dollars at their disposal. Do you honestly think you can knock up a better and more successful auto-strategy on tradestation and spend the day cruising around your yacht while the money rolls in?
It depends on your strategy