This post was contributed by a guest author, and does not necessarily reflect the views of Richard or MovetheMarkets.com
I have spent some number of years doing system trading development and I have coded 100’s of systems, all which eventually went bust, or I threw them aside and tried something else. As a result, I have been trading using charts, with my own custom indicators along with the indicators provided with the charting packages. I have always thought a systematic approach would be best, however, in my attempts to develop the ultimate system, I ended up with none. The purpose of this article is to show that you can get good results with commonly available technical indicators.
The truth is, given the current development platforms and technology, there is no magic indicator or system. You have to use a system that implements a core idea well, is not curve fitted, and can handle the market edge cases.
A few weeks ago I found Teresa Lo’s http://www.powerswings.com/ site. She is an experienced trader, who came out of the industry. I looked at her system, and bought her ultimate trading course. She uses a system that is solid, but is not rocket science. It’s a well-implemented trend following system. It works because she knows how to use it. Her swing trading is based on feeding it trending stocks, or she uses it to trade futures at times when they are trending. It is not meant to handle choppy markets, and it is not used for that purpose.
Using ideas from her system and her trading course, I created my own version, and I was amazed by the performance if it is pointed at the right trading instrument. Mine isn’t for sale and hers is, and hers likely better. The point is you are can create a solid system using common building blocks, or by purchasing a realistic system like hers. I don’t know her, I’m just commenting on what I learned at her site and reading her materials.
Here is a snap shot of my system below. It is really simple and consists of the following:
- A WMA which acts as a trend filter.
- An Oscillator (Pick almost any like PPO, Trix, etc. I am using my own RSI, but it really does not matter, trust me)
- Stop Logic (I am using my own modified version of Elder’s Safe Zone, but you can just use a multiple of ATR of price as your stops, like 2.5 or 3)
- Why does it work? Because it catches the trend and it doesn’t get stopped out on a little noise.
- Is it Magic? No. It will only work on something that trends,
- Was it optimized? No. I just picked values that looked correct on the SPY, and applied the exact same values to DIA, QQQQ, and IWM.
Performance
The system was run on the last 5 days of data, 5 minute bars for SPY, QQQQ, IWM and DIA, set with a timer to trigger trades only between 1:50 PM and 4:00 PM eastern, as Teresa Lo suggests with her system. This is because there is too much chop intraday.
The Tradestation reports are below. Using 2000 share lots (Equivalent to 4 mini’s) the system made $17,766 in the last 5 days. This time was picked at random. Will it make this in the next 5 days? Who knows. Do I think a good trend following system will work? Yes, if used to trade trending financial instruments.
What will this system do next week? Probably lose $100,000.
Update: Performance 8-8-2007
4 Trades, 100% profitable. SPY $1,559, IWM $ 1,889, QQQQ $459, DIA $989, Total $4,896.
- TraderD
Could you write something evaluating Lo’s materials and if you found it worthwhile for the cost? I’d assume you are at least 4,896-times happier with the results ;)
The description of your system sounds like something one could pick up in almost any trading book, but obviously there’s more to it.
Thanks, and keep us informed to your system.
Alan
Alan,
I promise to do a review, probably in the next few days. The system is that simple, which was kind of the point. T. Lo I think looks at this upside down, which in this case, is a good thing. Rather than designing a system which will make money, I think she trades what will make money. The standard for the system then becomes one that is good, not one that is fantastic.
- TraderD
Hey! Never thought that you would be a “mystery shopper”.
Really like your comments about the importance of the security and time frame selection. I learned this from the desk and floor traders that I worked with when I was in the business.
Whenever I tried to trade something where there was no action, they would just give me a dirty look and ignore my ticket.
The rest of the post is here.
Cheers!
Teresa
Come to think of it, what I should have done was put a “Range Filter” in the code to stop people from trading when it is dead.
The reason is that if the market is swinging wildly, we know we can make money from the fluctuations, and the position size will be reduced automatically by the algoritm. If we try the other way around (small range and big size), it easily ends in disaster, as we have seen happen to the hedge funds.
@T,
Thanks for the comments. On the range idea, I have never gottten that to work, mainly because of the conjestion before a move that will have a small range, like LBR’s “spring idea”. I’ve tried all kind of filters, and it seems to create another problems, usually delaying entry when you don’t want it to.
On the other hand, you may have better approach than what I tried. I’m the grasshopper, you’re the Sensi ..
- D
Blau’s work has a number of trend filters. They work better on swing trades than intraday. I still like your idea better, which is to figure out what is trending first.
In my stock scans, I have a range of at least $1 a day or sometimes more, averaged, and increasing.
- D
how about an update on your progress
I think TraderD was so disgusted by the site redesign that he hasn’t been back since :-)
lol
Hay TraderD,
Are you still working with Theresa Lo’s PowrTools, and if so how is it going?
Tkx.
I think the only way to trade is with mechanical trading systems. As Welles Wilder suggested, no more than 2 parameters is ideal! Here’s a good into to building a trading system:
http://www.traderslog.com/buildingtradingsystem.htm