It was like 1h ago. But I think it was a bug of some kind.
- How long have you ran the 10s nonethe?
Also i got different results on demo and on live with same .itf
but I guess its prt problems like he said.
By reading through the thread is it recommended to change the timezone to English or doesnt it matter I didnt get that part.
It’s the Dow Jones so it runs on NY time. I’m in London so Wall St hours are 14.30 to 21.00. Change the line that says Ctime = to whatever the Wall St hours are where you live.
So if I understand you right this is something I have to do on all my algos? Isnt there a general option? And also this part, if it something like this Ctime = time >= 000000 and time < 230000 I do have to change it also right? (I just went to +1 England instead. I guess that will fix all algos to right tz.)
If you are running your platform on your country time / France / UTC +2 then you need all times in codes to be changed to UTC +2.
So if a code was written by a UK coder with times in it then he would have used UTC +1 as that is what UK time is currently through the Summer.
So you would need to add 1 hour on UK / UTC +1 times to make the times in the code good for a Platform running on UTC + 2.
https://www.timeanddate.com/time/aboututc.html
Yeah or keep coding in London C-times and do +1 in the program. Thanks
Yeah or keep coding in London C-times and do +1 in the program
What and run your Platform on UTC +1 also??
One bigger losing trade today on the 10s. Im trying to tweak it, if someone gets better result, please share it. (overall it seems to win more trades than losing but the losing ones are often so much higher so it doesnt profit.
One bigger losing trade today on the 10s. Im trying to tweak it, if someone gets better result, please share it. (overall it seems to win more trades than losing but the losing ones are often so much higher so it doesnt profit.
I know that short timeframe systems are greedy as they promise high earnings in a short time, but it is no wonder that overnight they can easily become trash; don’t focus on them so much.
However, it remains interesting to continue developing them for the sake of knowledge and (why not) the search for the right one.
I feel like its quite good since it winning so many trades but i dont know.
@ you have to be very careful with such short TFs because there is no real back test to speak of. All we know is that it worked last week, that’s it.
I would strongly recommend regular re-optimising – at least once a week, maybe more.
I feel like its quite good since it winning so many trades but i dont know.
It is winning so many trades because the backtest shows you only the previous 5 days and the algo is optimized only on them.
I also can create a 100% system with 500% gains by optimizing the entries on what happened last week, but that does not mean that it will work in the future (it’s not secure for any kind of system, but for short timeframes ones is mostly common).
@nonetheless i’m not discrediting your work, i know that you also have this thought about short timeframe systems.
