did the paper account register the same drawdown today? also I was wondering what was your historical maximum daily drawdown for the year of test data?
Frosty's trading bot goes live part 2
- Thread starter frostengine
- Start date
did the paper account register the same drawdown today? also I was wondering what was your historical maximum daily drawdown for the year of test data?
walter,
Yes the paper account's results were the same.. well actually was off a couple dollars, but that was due to the difference in fills between the two. But the jist of the results was the same....
Yes the paper account's results were the same.. well actually was off a couple dollars, but that was due to the difference in fills between the two. But the jist of the results was the same....
Does the system have a disaster stop? How come there is such a big drawdown in one day?
U are right about not turning it off. There will always be losses, once you turn it off it will go your way again. But is tough!
U are right about not turning it off. There will always be losses, once you turn it off it will go your way again. But is tough!
Has your strategy changed a lot since your last run? Below is the ER2 chart for 6/20, which is the day your bot lost $1200. The dominant trend was down through the day, and since you indicated that the bot is "trend-following", I wonder where your entry/exit points were. Can you circle them on the chart or mentioned the times the bot bought and sold?
linear brings up a great observation - if you are following the trend frosty and Wed's move was down, up a tad, down more, repeat process is your bot best positioned for straight up/down days or what kind of conditions? That's an important observation linear made b/c if the trend following bot can't make money on a day like Wed, I'd be curious to know when it will make money if the goal is to follow the trend.
Just some food for thought.
Just some food for thought.
it really does'nt matter 'how' it makes money. just that it does make money. and you have the statistics behind it to support future performance. you could have a trend following system that only makes money on days beginning with 'T' and it still isint a bad thing
as long as the statistics support itBad luck on your first day.
Looks to me like your system is analysing either too short a trend length - or your stops are too tight.
I'm keeping my fingers crossed for you though - I have been through this pain myself (not with real money fortunately).
I did have some success finding a system - but to keep the drawdowns manageable in the worst scenarios I could find in back-data - meant that it only made about 12% a year on average backtested. I can beat that easily with my discretionary system.
I plan on going back to it - but I want a bit of time off to clear my head so I can think up alternative strategies, I found I got stuck in a rut trying to flog the last % out of a system that wasn't delivering.
Looks to me like your system is analysing either too short a trend length - or your stops are too tight.
I'm keeping my fingers crossed for you though - I have been through this pain myself (not with real money fortunately).
I did have some success finding a system - but to keep the drawdowns manageable in the worst scenarios I could find in back-data - meant that it only made about 12% a year on average backtested. I can beat that easily with my discretionary system.
I plan on going back to it - but I want a bit of time off to clear my head so I can think up alternative strategies, I found I got stuck in a rut trying to flog the last % out of a system that wasn't delivering.