Posts made by seb 0

What a beautiful downtrend
Maybe you should try to extend the stops

@roar
Indeed it looks far to be excellent. Since it probably is slightly above 50% chance of winning I'm gonna invest some money I can loose while improving my strategy
Thanks for showing me i had a too narrow view of the picture!

@roar
I agree, but it's the maximum I can do on this time frame with this strategy and 3 years is the maximum period for a 99% modeling quality, maybe im gona switch of broker.
Ok i didn't do it because i thought these parameters should be part of the fixed ones..

@roar
Alright this reassures me a bit : ) , here is my EA https://fxdreema.com/shared/OKVi1ScGd
The 6 inputs to optimize are the 2 TPs; 2 SLs and 2 MAs ... in the constants, did it on GBPUSD; H1; 10/10/2017 - today.

@roar
You were saying about optimization that : The "forward results scatter" should have a rising pattern, similar to the "optimization results scatter" so it would mean that the EA survives well on new data.
I managed to have some good optimization results with my last EA using the Forward result that had the best compromise result between Back-test and Forward-test. However I don't manage to get this rising pattern on that "forward results scatter".
Would you mind if i send you my EA so, by using your methodology you can tell me if you find a good "resilience" also or it inst worth it?

@tsheppo
Hi, can you paste the fxdreema shared link of your project please

@ajmd97fx
This is in the case of a (positive)profit inferior to 5% of the balance

@ajmd97fx
Sorry i should have done this configuration
In yours you should reverse the "<" sign if you want the difference to be higher than X% of the Balance

@nazafks
Hi, you could make increase a variable by 1 every candle by the condition the position is losable, when the condition is above N -> close the trade.

@ajmd97fx
Hey i think I think this is what you need to do

@khalids222
I am a bit lazy helping you building this strategy, even it is not so complicated.
However I can give you my advice : I don't think strategies only based on distance between one candle and another are worth it. Because,

• no trend is supposed to last especially on the forex, in a way that if candle 0 > candle 1 the probability that the next candle is also above is close to 50%. 2 bull candles on H1 is an up trend on M5.
• And the probability of the price to reach a point is inversely proportionate to the distance between it and the start point, in a way that the probability of the price to go from 1.1500 to 1.1520 is 2 times lower than the probability to go from 1.500 to 1.510.

So the only utility(or one of the utilities) of fixing different levels of stops is to avoid the unluckiness of loosing an entire trade because of 1 pips, by spreading the targets into 2 or 3 stops to decrease the risk.
I would so add other indicators, more information to take in count in the strategy.

@xbuilder
Hum something like that then ?
https://fxdreema.com/shared/h2X08rYSd

@xbuilder
https://fxdreema.com/shared/EynkkElgd
Is that what you are looking for ?
The percentage is the "0.5" in block 18, for 50%.

@xbuilder
That's wierd, % of account loss, or % of specific positions loss ?

@biztet
I think it's a bit too simple to spot true reversals, but it can be seen temporary resistance or support zones at least.
It is the strategy of Zackry after all so you should ask him.

I actually don't manage to import properly bars from tickstory to my mt5 broker.
These are the settings in tickstory (with H1 period) :

I then converted the .csv file to .json so it can be readen by the broker. I've placed the file in the terminal folder :
Launched my broker, and i can select the symbol appearing in the backtest window :

But it doesn't seem to find any data. I've tried to create a symbol by selecting this .json file but no graph is displaying, do you have an idea ?

@zackry
Good it is useful to you : )

Hmmmm... right lets let Zackry choose the project he prefers

@miro1360
I'm not sure this is correct
Conditions in block 15 and 16 can not be filled but, what Zackry want to be true, be true.
wick3<wick2<wick1 isn't necessary in this case.. or I didn't understand @Zackry well.
Test your EA then mine they dont have the same results.

Now works perfect https://fxdreema.com/shared/gk5Bhv8I
@Zackry