Profitable EFS strategy.

[demarcog]

The strategy I'm attaching is extremely simple, is not a holy grail of any kind but does make a good profit on intraday (15 minute) data for volatile instruments such as the SKF.

Trading 100 shares with slippage of $10 and commission of $1 made about $48K during the last 120 trading days on 15 minute 9:30-16:00 time template.

Unfortunately for me I'm down for the last 120 days hopefully you can muster the discipline and impulse control that I unfortunately have not. Don't repeat my blunders by trying to combine discretionary and systematic trading...."stick to the system" 100 percent of the time.


This EFS is backtest compatable and will run in real time and trade automatically. If anyone can suggest more profitable versions feel free to let me know I could also use the help.

I realize the last thing almost everyone would ever do is share or publish a profitable trading strategy. Why not help someone avoid the misery I've experienced and still experiencing? In addition Alex, Jason, Steve and many others have been very helpful to me over the years and I'd like to "pay forward" their kindness and assistance.

Hopefully someone will find this helpful if you do please let me know.

Glen

g[email protected]

[Doji3333]

Try this.. I've been helping clients build trading systems for quite some time.. I don't believe this is the holy grail either, but I believe I have improved your basic logic by using two new indicators and changing your logic a bit. The other thing I did was to change it from a "rotational system" to an "entry/exit system"..

The purpose of these changes is to try to grab any profits that are available before a trade turns dramatically into a loss.

I could probably improve this further for you, but right now I think I've found my holy grail and am busy developing IT.

BTW, it is great that you posted this strategy here and provided the source. I hope we can continue to build this for esignal clients.

One other thing, I don't use the internal Global Broker Functions, so you'll have to address the commented sections of the code for the EXIT signals.

Let me know if this generates better profits or not on your tests??

B

[Doji3333]

results

I was playing with this and comparing the code between the original version and my new version. Here are some general determinations from running a simple ES 15min 6:30~13:00 test..

_ my new code takes a few more trades (but that is to be expected seeing as though I changed it from a rotation mode to an entry/exit mode).
_ my new code appeared to have a more consistent monthly return than the original. Although the original code appeared to do well on my ES test, the equity curve for my version actually resulted in less drawdown (which is a good thing).
_ my new code appeared to have a greater "per trade" result as well as other results.

What this tells me is I was able to improve the system by probably 15~20%. Now, my tests include a $12.50 slippage and $5 RT commissions.

Here are the images.

The first three images are from the original system. The last three are from my revised code.

Let's try to keep this going as I find this interesting.

[Doji3333]

Let's try this differently. I posted the images to my site so I can included them into one message...

Old Code Results:







New Code Results:

[demarcog]

Brad,

Very much appreciate you taking interest and posting the improved code and yes let's keep this going.

As you observed there are problems with this system:

- the first being it absolutey "gives back" too much of the incurred profit I wanted to try adding profit targets and trailing stops.

Glen