Is 100 percent mechanical trading possible?

[StratOpt]

For me, the mechanical, means that I can come in the office this morning, I switch on the strategies that are trading this day, and then I go on to my other business of developing strategies and testing strategies and writing code for people. The only thing that I need to do is make sure that when I hear an order being sent that it actually did what it was supposed to be doing and that no errors have occurred. At the end of the day, you can turn them off and move on to another day. The strategies that I am employing are in no way as complicated as some neural network artificial intelligence program, nor do they need to be.

 

[pedro01]

If you are making 44% a year - why do you need to spend time writing code for people ?

Did you only start with 25c ???

 

[fantastic]

Pedro01

I think he started by being paid to write codes for an accomplished trader.
But I am not sure. StratOpt can you confirm this.

 

[thierry911]

I understadn that StraOpt.. keep writing codes... Thats what he likes and started with. THat also give him the opportunity to talk to other people and to get new ideas to develop new strategies.
You need to be creative in this job.. One way is to keep talking to people to share ideas.

 

[Richard Madoff]

I'm almost certain you can make more than 48% with an automated system if you are a bit more clever about the way you manage your risk

 

[meanreversion]

Nearly a two year gap in posts there ... for your first comment, you could perhaps have picked a more recent thread .. ?

 

[dreisign]

I use a 100% mechanical system and it is working out very well in backtesting. A consistent 30% per year.

 

[meanreversion]

I use a 100% mechanical system and it is working out very well in backtesting. A consistent 30% per year.

You need to tell us a little more than that.

1. What is the maximum drawdown you encounter in backtesting
2. What is the trade frequency
3. Have you incorporated slippage and commissions
4. What is the % win rate

 

[Citizen2007]

Yes it is.

 

[dreisign]

Win % rate is 66%. Max drawdown is only 10-12%. The longest string of negative territory is 4 months. I trade about 40 times a year, about 3 times per month. The only downfall to the system is that to achieve the 30% per year I need to risk 5% per trade. However, I am comfortable with that based on my max drawdown stat.

 

[dreisign]

Commissions and slippage included

 

[Big Spender]

Hi
I’m looking for a system to trade indices.
Dow, S+P etc.
Can anyone help?

 

[Richard Madoff]

Win % rate is 66%. Max drawdown is only 10-12%. The longest string of negative territory is 4 months. I trade about 40 times a year, about 3 times per month. The only downfall to the system is that to achieve the 30% per year I need to risk 5% per trade. However, I am comfortable with that based on my max drawdown stat.

Personally 5% is well above what I would risk (2% max). But if you have multiple systems running across multiple timeframes then the total amount I would risk at any one time would rise to 8%

 

[numbertea]

I do believe that it can be done if enough and the right things are considered. The system I built has selections for determining what criteria make for a good 8 day long or less downtrend, it employs 140 variables for criteria selection. That is a small part of my system. I created my own program because the simple logic used in the commercial ones I've seen can't provide me with the complexity I need to accomplish the systems I think up. Papertrading has gone very well so far, I use any stock in NASDAQ 500.

 

[Richard Madoff]

I do believe that it can be done if enough and the right things are considered. The system I built has selections for determining what criteria make for a good 8 day long or less downtrend, it employs 140 variables for criteria selection. That is a small part of my system. I created my own program because the simple logic used in the commercial ones I've seen can't provide me with the complexity I need to accomplish the systems I think up. Papertrading has gone very well so far, I use any stock in NASDAQ 500.

I am impressed - I don't think your average trade2win member is anywhere near this stage. I am going to start by building a simple system and build up in complexity gradually. I wanted to know if you could recommend any backtesting software. I am quite handy with programming, so I am not bothered about difficulty, I just want good software for backtesting/system development. From looking around I think that Tradestation is the best but there might be something I've missed.

Thanks!

 

[numbertea]

I have been told that AmiBroker is a good backtesting framework where the programming you do to create a system is in C. I tried out tradestation and liked that and it's Easy Language is fairly intuitive and complete for all kinds of conditional logic. Since both these software packages require serious programming anyhow I figured that I might as well just build my own framework and not pay someone for theirs. The charts I create are exactly the way I want to see them, personal preference is completely satisfied when you do it yourself. I plan on trading for years and so I think I made a good decision to spend the 20 hours of programming it took for me to make the initial program. Now I add as complex a pattern or requirement as I want...no limitations. I built my system with php and serve the pages to myself with apache on same computer... fast easy graphical development and super easy array handling.

 

[Black Swan]

I am going to start by building a simple system and build up in complexity gradually.

Why?

 

[Chartsy]

remember not to backtest random stuff, your system has to have logic driving it, tendency to fluctuate around mean, trend etc

 

[forker]

As much as I would relish a box in the corner of a room making me money without any effort. I do enjoy the process of manually extracting money. It's a case of getting my hands dirty and feeling a sense of achievement at the end of each month

 

[the hare]

As much as I would relish a box in the corner of a room making me money without any effort. I do enjoy the process of manually extracting money. It's a case of getting my hands dirty and feeling a sense of achievement at the end of each month

I've done both, and I can assure you looking after a box in the corner of the room takes a great deal more work than watching charts and occassionally clicking a mouse !