Backtesting Results Attached - Not Selling Anything

joecgoodman · #1 (**permalink**) 03-17-2007, 01:36 AM

$4000 grew to $37,911.05 so far this year (2007) (about 54 trading days). Minimal drawdowns. Compounding used.
Impressive, yes. But more data needed.
The attached backtesting was done manually. Very, very tedious.
I don't know mql. Hence, this posting.
---
My system consists of common sense entry and exit rules. No curve fitting.
Wins half the time. Winners twice as big as losers.
Generates an average of 2 signals per day.
Uses a dynamic stop which is capped at 20 pips max.
Perhaps best of all, it should work fine in trending or range markets.
More testing will either prove or disprove that.
---
The possibilities are endless. Consider this:
The attached backtest results are on only 1 pair and 1 timeframe (EURUSD: 5 min).
I can see running this on 3 pairs X 3 time settings; 9 systems running at one time.
Total system drawdowns would diminish to near zero .
Might go years without a losing week.
This is possible with a small loss / high-frequency system.
---
Anyone want to code it?
If so, I have detailed screen prints with text on the pictures that make the rules crystal clear.
I'm working on the TradeStation code, but I'm slow.
And I'll need MetaTrader to automate trading with it anyway. Please correct me if I'm wrong on that.
I'm a seasoned trader, but green green green when it comes to MetaTrader.
joegoodman@gmail.com
---END

RickW00716 · #2 (**permalink**) 03-17-2007, 02:41 PM

Quote:

Originally Posted by joecgoodman

$4000 grew to $37,911.05 so far this year (2007) (about 54 trading days). Minimal drawdowns. Compounding used.
Impressive, yes. But more data needed.
The attached backtesting was done manually. Very, very tedious.
I don't know mql. Hence, this posting.
---
My system consists of common sense entry and exit rules. No curve fitting.
Wins half the time. Winners twice as big as losers.
Generates an average of 2 signals per day.
Uses a dynamic stop which is capped at 20 pips max.
Perhaps best of all, it should work fine in trending or range markets.
More testing will either prove or disprove that.
---
The possibilities are endless. Consider this:
The attached backtest results are on only 1 pair and 1 timeframe (EURUSD: 5 min).
I can see running this on 3 pairs X 3 time settings; 9 systems running at one time.
Total system drawdowns would diminish to near zero .
Might go years without a losing week.
This is possible with a small loss / high-frequency system.
---
Anyone want to code it?
If so, I have detailed screen prints with text on the pictures that make the rules crystal clear.
I'm working on the TradeStation code, but I'm slow.
And I'll need MetaTrader to automate trading with it anyway. Please correct me if I'm wrong on that.
I'm a seasoned trader, but green green green when it comes to MetaTrader.
joegoodman@gmail.com
---END

Very impressive results!
I am curious to know the logic behind your system. I am not a coder but I have a friend who might be interested.

Have you tried trading it live yet?

tdion · #3 (**permalink**) 03-17-2007, 11:28 PM

Hi Joe,

This looks interesting. I might code it sometime this week. Thanks for sharing.

Tom

Wackena · #4 (**permalink**) 03-19-2007, 11:41 AM

Quote:

Originally Posted by joecgoodman

$4000 grew to $37,911.05 so far this year (2007) (about 54 trading days). Minimal drawdowns. Compounding used.
Impressive, yes. But more data needed.
The attached backtesting was done manually. Very, very tedious.
I don't know mql. Hence, this posting.
---
My system consists of common sense entry and exit rules. No curve fitting.
Wins half the time. Winners twice as big as losers.
Generates an average of 2 signals per day.
Uses a dynamic stop which is capped at 20 pips max.
Perhaps best of all, it should work fine in trending or range markets.
More testing will either prove or disprove that.
---
The possibilities are endless. Consider this:
The attached backtest results are on only 1 pair and 1 timeframe (EURUSD: 5 min).
I can see running this on 3 pairs X 3 time settings; 9 systems running at one time.
Total system drawdowns would diminish to near zero .
Might go years without a losing week.
This is possible with a small loss / high-frequency system.
---
Anyone want to code it?
If so, I have detailed screen prints with text on the pictures that make the rules crystal clear.
I'm working on the TradeStation code, but I'm slow.
And I'll need MetaTrader to automate trading with it anyway. Please correct me if I'm wrong on that.
I'm a seasoned trader, but green green green when it comes to MetaTrader.
joegoodman@gmail.com
---END

Joe, I will send email to you with my email address. If you want me to try to code system, please send data by return email.

Wackena