Youmu's H&S EA Ver0.1.3 Need Filter

Jonathan You · #1 (**permalink**) 10-17-2006, 09:39 AM

I'm developping a EA based on Starc Bands and Hurst-StDev Bands, here is the back-testing results of 2006/10/13. I need a filter of Starc Bands and Hurst-StDev Bands, Anyone has good idea or indicator, Please PM or Send me, I think it will be it'll be profitable for us.

H&S EA works on any time frame and mainly on M30. This Version will be expire on 2006.11.1.

juanchoc · #2 (**permalink**) 10-19-2006, 07:21 AM

Hello!
Your statement looks good at first sight, but if you look at the trades your ea took, you will see that theres not much of money management, and that the risk is really high.

The best thing to do is to block the ea from opening more than 1 position on a specified time range. Remember that opening many trades at almost the same time, is almost the same as opening 1 big trade. Main diference is that you have lots of trades and so lots of spread and commision costs, and im not even talking about all the risk that you are runing.
For example, at the begining of the statement, if you add up all the lots of the open positions you have at the moment of trade #22, you'll see that you have 7.3 LOTs. LOL

but thats a lot for a 10k account, at that point you are risking 73% of your initial equity!, so its only a matter of when you will go broke. Imagine yourself selling at trade #22, and some minutes after that, you have a new eventon the other direction, you've just lost almost all your account, and are left with only 27% of your original account size, so you would need to make more than 370% of your equity to get to the breakeven point...

So, recomendations:
1- Test the EA on a biger time range 4 months -2 years, and try to see on what market sittuations it develops better.
2- Limit the number of open orders, you can limit it to 1 order at a time, or to 1 order in a period of time (depending on the timeframe you operate).
3- Try to use some money management rules, ex. dont risk more than 2% of your balance per trade (if 10.000 x 2% = 200 or 0.2 lots).
4- As for the filter, check if its really necesary to add a filter after implementing the first 3 recomendations i gave you, because remember filters not always filter bad trades, but sometimes also filter good trades (besides, if you follow recomendation #2, you shouldnt be so worried about filters).

Cya,
JCC

Jonathan You · #3 (**permalink**) 10-19-2006, 07:42 PM

Quote:

Originally Posted by juanchoc

Hello!
Your statement looks good at first sight, but if you look at the trades your ea took, you will see that theres not much of money management, and that the risk is really high.

So, recomendations:
1- Test the EA on a biger time range 4 months -2 years, and try to see on what market sittuations it develops better.
2- Limit the number of open orders, you can limit it to 1 order at a time, or to 1 order in a period of time (depending on the timeframe you operate).
3- Try to use some money management rules, ex. dont risk more than 2% of your balance per trade (if 10.000 x 2% = 200 or 0.2 lots).
4- As for the filter, check if its really necesary to add a filter after implementing the first 3 recomendations i gave you, because remember filters not always filter bad trades, but sometimes also filter good trades (besides, if you follow recomendation #2, you shouldnt be so worried about filters).

Cya,
JCC

Thanks a lot juanchoc, I'm coding the money manage of the EA upon your recomendations.

msaqibc · #4 (**permalink**) 06-10-2007, 02:10 PM

i was looking for an automated starc band system... came across your thread. if you have finished the final version, could you please post it on the thread ? thanks