For Trading System Beginners, beware of systems that...

W

WDGann

I actually had the chance of meeting Larry Conners from Tradingmarkets.com this week. We were talking about a few things briefly and something struck in my head and based on that I'd like to mention a few insights, not about Larry but about system trading, in general.

First, I don't believe in buying a system from 3rd parties. I also have a good idea about money management, system trading and it's statistical edge toward it (well, an expert in Gann Analysis too, look at my name... ha ha). Well, we were talking about money management and he mentioned about RINA systems and how it was great regardless of Tradingmarkets.com's interest and how thorough it was. Though, I like to stick to using WealthLab for diversified and Tradestation for single derivative product for flexibility, when testing money management rules to backtest. (Larry is a great guy, though... Thanks for the time, if you are part of ET)

While driving home after the meeting, I was thinking about 3rd party charting software vendors and eventually led to trading system vendors. Tradingmarkets is a newsletter service especially have a strong marketing and service for daytraders. I used to daytrade before and have a good idea about NASDAQ LVL 2 stuff.

Now, I put the daytrading and 3rd party trading system vendor together... it doesn't fit. Then something just reminded me of something when I made a transition from a long-term system trader to a day-system-trader. I was reminded of the irony of a system-trader especially for trend-following systems that are out there and for intra-day trading. Reason? Execution.

Daytrading and most 3rd party trend following system doesn't work. First, when daytrading all the orders you put in are Limit orders(OK, there are SOES but they're practically bad fills). Another thing is when daytrading Support/Resistance is extremely strong in any case. On a Daily Bar you don't see it as much but when you are Daytrading, the resistance lvl is extremely evident and one of the keys to success in daytrading is how you deal with resistance/support lvls. Because of this, when the resistance is broken the market just tanks like it never wants to go back.

Here's the thing, most trend following systems, waits till the trend occurs. Then they go in and try to ride it. So in another words:

!. The market is going hovering around the resistance lvl
2. It breaks through the resistance and tanks... then makes a trending signal.
3. You get the buy order from the system. Meanwhile market is tanking
4. You put your order in...
5. You get massive slippage.

This does affect the expectation of the system drastically. I actually tested the some trend-following systems out with intraday data, it works great on the results. Then I put a few criterias into the trend-following systems with Uptick rule and responsive delay on order entry.

Funny thing is most of these system can't live up to the initial performance. Luckily, the system I developed on my own for my personal daytrading doesn't have to deal with this too much. So I didn't have a clear idea of it as I do now. It was practically part of me through instinct, but if you are starting to daytrade and thinking of buying a system, ask for how realistic the order entry criteria is.

While I was writing this I remembered how few trader buddies of mine were mentioning about it and how I've previously read about it in different materials.

I guess actual trading is not the only inspirations you get for trading. Thanks to Larry.
 
Recently, I had some daytraidn ides that I wanted to try so I subscribed to TS6. It was a smart move. Not because I made money, but the education was way more valuable. Not only did I realize that I don't care to daytrade, but I learned some caveats.

Oe of those was what you referred to WD. I tested several ideas and some produced net profits, some not. Some of the net profits were small, some were large. All the profitable ones were generated on 300-500 trades over the previous 100 days. When I added in commish, obviously the profits decreased.

But the real unknown was the slippage. The difference between a big profit and a big loss over time was a nickle. One nickle's difference on each side of the trade killed the profit.

So when the system said by at the close of this bar or buy at the open of the next bar, great. But in real time, it was impossible to get the same price "the system" got. I was lucky and tried it real time when there happened to be larger than usual profits. So yeah I made a little money, but in the long run, I would have lost.

I would have been a good little system follower and entered whenever I was supposed to, but almost never at the same price.

Most systems all use the disclaimer, "slippage and commissions not included." They are unknown variables to the writer. To the buyer, the commish can be easily calculated. But not the slippage. No you can't just use limit orders because ou will then miss moves.
 
Originally posted by inandlong


Oe of those was what you referred to WD. I tested several ideas and some produced net profits, some not. Some of the net profits were small, some were large. All the profitable ones were generated on 300-500 trades over the previous 100 days. When I added in commish, obviously the profits decreased.
More...

My trading is 90% Gann based. The reason so many people do not believe in Gann is because they can not make it work for them. To be perfectly honest, its not their fault, rather, the way the 1 X 1, 1X2, 1 X 4 lines, etc. are calcualted in their charting programs are flat out wrong. It is much more complicated mathmatically than just picking a point and drawing a trendline that rises one unit in price by one unit in time.
 
I completely agree with Sarasota.

Actually I put my % of using Gann is 50%. I do get a good idea about the market through Gann Analysis but I trade my system. The analysis I make usually gives me an idea of the market before the signal comes out. If the system confirms and gives a signal, I go in. If they contradict, I don't go in.

Because my analysis is always faster than the system's signal, so if I am right with the analysis, I get a signal. If my analysis is wrong then the signal doesn't come out.

Also, it's a matter of edge. Gann Analysis gives me the very large edge but it's inconsistant. System Trading gives me a somewhat large edge and it's consistant.

Gann Analysis is like a Monte Carlo criteria for the system. One thing I can add to this is that, the way I trade is conceptual. Well, Gann is Gann... angles, time counts, cyclical analysis, and etc.. My trading system is a different matter, it's not a Gann related system. But the concept between the two are the same. When using Gann I have certain concept in perspective, the system I trade is based on the same concept with different techniques.

As an example, I try to catch a trend and strength using Gann, then I use a trend-following system.

For my daily homework, I go through the complete process though.

PS. I don't use variations of Gann Analysis that are out there. I use purely what Gann mentions in his books and pretty darn good at it.
 
You must log in or register to reply here.
Back
Top