DAX 30 contado 25€ TF1m
- This topic has 170 replies, 24 voices, and was last updated 6 years ago by
.
-
Sorry Juanan, but I don’t think we really can help you because the principles of your strategy are too personnal: you take position in a certain configuration at 11:13 am because you found good results in this configuration 10 times in the last 6 months (more than 100 days), but this is not statistically significative, and this hour doesn’t make sense (hour of open or close of markets would be different of course). So your backtest is perfect because you have kept all good configurations in the past, but there is no reason for the market to repeat such scenario at this very moment. Or this is a belief, and we can not discuss beliefs. Anyway, to evaluate the strategies, the only thing to do now is to let them in live demo and not modify them to adapt to new conditions. In a few weeks, it will be possible to know if the strategy is interesting.
1 user thanked author for this post.
I agree absolutely with Aloysius. It looks for me also like a product of data mining – you found random patterns which simply occur if you look at sufficient data. There’s something in statistics called “Infinite monkey theorem”…
1 user thanked author for this post.
Hi Juanan,
Thanks for showing the code. I have tried “filtrado200k” in DAX TF 1m and it seems very good. I am going to study the code, knowing that you have used several filters, and will put it in demo for some time. I have checked in 100k units and got 49-0 since July17. Very impressive!!!
As Despair said, you have found, or I should said “put together” a random pattern that only time will prove right. In any case, stick around in the forums and lets see in some weeks how the code goes.
Great work, considering you are new on programming. Thanks for contributing,
Juan
1 user thanked author for this post.
Hi Juanan,
Since the SL is 25-28, I have decided to try it in real and today I just have the first winning trade (12€) in DAX 1€. I have a very good feeling about the code.
Buen trabajo,
Juan
1 user thanked author for this post.
I’m not completly sure…so i’ve found other code is using a diferent multiplier for the diferent months and days..so it’ll be confirm my theory than every seasons have diferent action…he aplies 1,2 or 3 valor number to every month and 1,2 or 3 for every day..adding valors we have 1 unique and depending of this number the action is one or other..
Jan,Febr,Mar..August,septembre,october are same=1
April,may,june, nov and dec=2 and july=3
following this theory now is not the same season than last 100K…mid august,sept,october
if the theory is good will have next 2 months failing so i need test and test and test..maybe i can found something in other codes…
today was ok for 1 order +12 points but maybe tomorrow lose 28—..need weekly balance or monthly
1 user thanked author for this post.
Find exclusive trading pro-tools on 
