Fractal breakout intraday Strategy EUR/USD 1H –
- This topic has 359 replies, 1 voice, and was last updated 1 year ago by
.
-
I answered to this question in this post : https://www.prorealcode.com/topic/fractal-breakout-intraday-strategy-eurusd-1h/page/6/#post-32855
The last one because it refers to the recent market behavior, assuming this period is close of the current day of course.
Walk Forward optimisation is the same as a classic one except that it divides the history into 2 periods, the In Sample one (which is the tested period where variables are optimized) and the Out Of Sample one (the period where these optimized settings are tested). This test can be repeated X times you want over the whole history. Please tell me if it is now clear 🙂
1 user thanked author for this post.
Hi Nicolas, yes I read the post, but obviously it’s not an exact science- So what I got out of your posts is that:
- It should be the most recent WF sample if WFE>50%
- WFE >50% but if larger that 100% we cannot expect this to be normal behaviour.
- if sample 1-5 is very different we need optimize again after x month
- if sample 1-5 is more or less a like, we have a very solid strategy
is this the rule of thumb?
Your WF video on youtube only has 295 hits- it would be a lot more if it was in English.. *wink *wink 😀
Have a nice weekend
Kasper
3 users thanked author for this post.
Hi!
If you test your strategy with different setups for your WF :
ancored/not ancored
repetitions
in sample%
out of sample%
and then take the best WF and do a regular opti on the same amount of in sample data as your most robust WF .
Then you atleast do the same as your test.
just an idea.
Regards
Henrik
1 user thanked author for this post.
Just an idea Nicolas … why don’t you get a French to text converter and then a french text to english translator and then english text to audio … hey presto english WF video and any other language you wish! 🙂
Or even French to English audio / any other language?
I know ‘it can get lost in the translation’, but you know brain storming and all that, it might trigger a less crazy idea??
GraHal
1 user thanked author for this post.
@all can we register in mother language walk farward video for the whole community to help Nicolas ?
Nicolas can you give us in English test your explanation of WF, to translate and register it to add to your video?
2 users thanked author for this post.
HELLO,
Following is a rough version of the strategy. It seems to work better in different markets. I found that in order to find the CP value appropriate to the market, the search can start from a crude code, I removed the trailing stop and the Donchian stop, then I put a simple stop and a take profit, and I have startan automatic search of the values: CP,SL,TP. The dax version that follows it is rough because it seems to work better than with the trailing stop and Donchian stop. In the coming days, I’ll used the rough version to look for other markets and other time frames. Thank you all for your attention
A greeting, Ale
6 users thanked author for this post.
…FBS_DAX_1H_V2
6 users thanked author for this post.
Looks really promising! Thank you very much!
1 user thanked author for this post.
HELLO,
here below results of Rough strategy Bund 1h
Regards
1 user thanked author for this post.
FYI, I compiled a lot of useful informations about Walk Forward tool in this new blog article: https://www.prorealcode.com/blog/prorealtime-walk-analysis-tool/
How it helps everyone..
9 users thanked author for this post.
Hello
Below the EUR/USD versions for the TF 30m, having lowered the TF I decided to add filters to raise efficiency. If you want to try changing the filter values, it will be fun.
In the first version there are two filters, while in the second I have eliminated the 1st filter and I kept the 2nd
1° filt: It is composed of two moving averages with a period of 20 and 50 that control each other
2°filt: The candle breaking the level must be 30 pips above or below the previous closure.
Thanks
Find exclusive trading pro-tools on 
