Forex Factory (https://www.forexfactory.com/forum.php)
-   Trading Journals (https://www.forexfactory.com/forumdisplay.php?f=49)
-   -   BTM - Expert Advisor (EA) (https://www.forexfactory.com/showthread.php?t=961302)

Froggy Jan 23, 2020 4:44am | Post# 2201

Only a few thoughts ... (1) I'm really pissed off that this thread was kicked off from Trading Systems Forum. Without single word of explanation. (2) Regardless what I implement into JF4XTrader EA people will start to requesting something else. (3) The most of people will disappear. Especially the most of new feature requesters will disappear. (4) Is there any reason to continue with this thread? regards
Joeff,

I haven't been around for a while, however I know that your dedication and generosity is useful to a lots a people including me. I have been using one of your indi with success!
Again thank you

kette Jan 23, 2020 5:01am | Post# 2202

Only a few thoughts ... (1) I'm really pissed off that this thread was kicked off from Trading Systems Forum. Without single word of explanation. (2) Regardless what I implement into JF4XTrader EA people will start to requesting something else. (3) The most of people will disappear. Especially the most of new feature requesters will disappear. (4) Is there any reason to continue with this thread? regards
if success does not go, people go for a reason, it must quickly bring success and profit
the reality is different because we don't know anything, we just guess (speculate)

Alberto_Jazz Jan 23, 2020 5:07am | Post# 2203

I'm running the system with parameters very similar to Vicente's report, I think this is the best configuration.

At the moment I don't know how to improve the system. As JooeF pointed out, the system works well in trending periods, but loose in ranging markets (nothing new for trend following systems).

The question is the same: how to avoid ranging markets?

JoeeF idea to work with equity curve is a classic solution.

I seem this is the state of the art at the moment.

JoeeF Jan 23, 2020 5:40am | Post# 2204

{quote} 1. This is not correct. The right place is TSystem Forum, no way! Explanation? Robots and automatics take decision probably, they don't have heart & soul! 2. People always have 1 idea a day, then the day after have even changed that idea. So go your own way, imho. 3. People come and go here and on social media. Specially who wish to have fast and for free money....let them disappear, you have always a lot of friends here to count on! 4. Is up to you, take your decision. This is a terrible job, even the best system has no chance to succeed on extreme & erratic wild market conditions. It's hard to go ahead in this job with the same confidence, the same strenght every day. But you have the only thing that matters in trading : the patience of saint!
I won't stop development of JF4XTrader EA. I've got a few new versions including usage of AI & ML.

The only thing I'm considering is support of this thread. Here are proper and "more" proper people on FF ... And one of the "proper" members was complaining/hating against this thread ...

Take care.

JoeeF Jan 23, 2020 6:26am | Post# 2205

I'm running the system with parameters very similar to Vicente's report, I think this is the best configuration. At the moment I don't know how to improve the system. As JooeF pointed out, the system works well in trending periods, but loose in ranging markets (nothing new for trend following systems). The question is the same: how to avoid ranging markets? JoeeF idea to work with equity curve is a classic solution. I seem this is the state of the art at the moment.
I've developed all required framework to improve this idea. And a huge testing is in progress right now ...

On the other hand I'm trying to implement some sort of AI and ML into decision process ...

Take care.

Brixen Jan 23, 2020 6:30am | Post# 2206

1 Attachment(s)
{quote} I won't stop development of JF4XTrader EA. I've got a few new versions including usage of AI & ML. The only thing I'm considering is support of this thread. Here are proper and "more" proper people on FF ... And one of the "proper" members was complaining/hating against this thread ... Take care.
Only you JoeeF that have the patience of a saint can fully understand these famous old words.....
Click to Enlarge

Name: Luke1.PNG
Size: 606 KB

Thgoldenstar Jan 25, 2020 3:06am | Post# 2207

My previous week with original settings (include TP1 in range) result:
1. For 28 FX pairs:
Gross Profit : 512.64 pips
Gross Loss : (259.85) pips
Profit factor : 1.973
Win rate. : 66.7%

2. For Commodities & Indices pairs:
Gross Profit : 32.29 pips
Gross Loss : (88.38) pips
Profit factor : 0.365
Win rate. : 23.8%

1 week result can not prove everything, however with large spread, different in contract size, commodities & indices pairs may not suite to trade this time frame

Toppi Jan 25, 2020 10:57am | Post# 2208

1 Attachment(s)
Hi! Why EA remove from charts.

I have read all and try all but still the same problem. When I put EA on empty charts its remove automatically after some seconds
Name:  2ad902341698a6cca17039d18eb6138e.png
Views: 439
Size:  16 KB

Ched Jan 26, 2020 7:31am | Post# 2209

2 Attachment(s)
{quote} You've changed settings. It's interesting ... "it survived ..." - This is important. We need robust system ... Your input is highly appreciated. Please keep posting. Take care. {image}
Thanks for explaining the cause for today's longs (tl) and today's longs for symbol (tlfs) aren't identified properly, I did see that but did not fully understand it. At least I was looking in the right place.
Also noted : (1) The system tends to produce streaks of wins and losses, so that when it starts to lose, it's best to stop trading until it starts winning again.
Last week survived with small profit (probably should have disabled it and gone fishing) by the looks of this weeks results.
This weeks results and open positions with same configuration.
Click to Enlarge

Name: Trades 20-24 Jan 2020.jpg
Size: 202 KB
Click to Enlarge

Name: open trades.jpg
Size: 68 KB

Nomenix Jan 26, 2020 9:51am | Post# 2210

1 Attachment(s)
This week was bad for me too, in all 5 configurations used.
I have identified another repetitive pattern like in the picture below.

Click to Enlarge

Name: Untitledxx.png
Size: 21 KB

In some cases the SL is smaller then 10 pips and TP is bigger then ATR. Not so good.
A fix could be something like:

Do not trade if SL < x pips
Do not trade if TP > x pips

At the time of the transactions, the bands were perfectly aligned and TP1 inside.

JoeeF Jan 26, 2020 11:41am | Post# 2211

Hi! Why EA remove from charts. I have read all and try all but still the same problem. When I put EA on empty charts its remove automatically after some seconds {image}
Every ignorant that doesn't bother to read the first post or doesn't follow the first post will be banned. No hesitation ...

JoeeF Jan 26, 2020 11:51am | Post# 2212

{quote} Thanks for explaining the cause for today's longs (tl) and today's longs for symbol (tlfs) aren't identified properly, I did see that but did not fully understand it. At least I was looking in the right place.
This bug should be solved in version v1.06 ...

https://twitter.com/joeef69/status/1...984147456?s=20

Take care.

JoeeF Jan 26, 2020 12:08pm | Post# 2213

6 Attachment(s)
Also noted : (1) The system tends to produce streaks of wins and losses, so that when it starts to lose, it's best to stop trading until it starts winning again. Last week survived with small profit (probably should have disabled it and gone fishing) by the looks of this weeks results. This weeks results and open positions with same configuration. {image} {image}
Definitely yes. A raw system with wide settings is break even ...
Click to Enlarge

Name: Image1.png
Size: 18 KB
Click to Enlarge

Name: Image2.png
Size: 15 KB
Click to Enlarge

Name: Image3.png
Size: 25 KB
Name:  Image4.png
Views: 337
Size:  7 KB
Click to Enlarge

Name: Image5.png
Size: 19 KB
Click to Enlarge

Name: Image6.png
Size: 16 KB

JoeeF Jan 26, 2020 12:19pm | Post# 2214

6 Attachment(s)
Also noted : (1) The system tends to produce streaks of wins and losses, so that when it starts to lose, it's best to stop trading until it starts winning again. Last week survived with small profit (probably should have disabled it and gone fishing) by the looks of this weeks results. This weeks results and open positions with same configuration. {image} {image}
{quote} Definitely yes. A raw system with wide settings is break even ... {image} {image} {image} {image} {image} {image}
Anyway, when you would implement equity crossover rules ...
Click to Enlarge

Name: Image1.png
Size: 18 KB
Click to Enlarge

Name: Image2.png
Size: 16 KB
Click to Enlarge

Name: Image3.png
Size: 24 KB
Name:  Image4.png
Views: 332
Size:  7 KB
Click to Enlarge

Name: Image5.png
Size: 20 KB
Click to Enlarge

Name: Image6.png
Size: 16 KB

JoeeF Jan 26, 2020 12:22pm | Post# 2215

{quote} {quote} Anyway, when you would implement equity crossover rules ... {image} {image} {image} {image} {image} {image}
You need to find a proper settings of course ...

JoeeF Jan 26, 2020 12:26pm | Post# 2216

{quote} You need to find a proper settings of course ...
Here's a huge space for ML I think ...

JoeeF Jan 26, 2020 12:33pm | Post# 2217

This week was bad for me too, in all 5 configurations used. I have identified another repetitive pattern like in the picture below. {image} In some cases the SL is smaller then 10 pips and TP is bigger then ATR. Not so good. A fix could be something like: Do not trade if SL < x pips Do not trade if TP > x pips At the time of the transactions, the bands were perfectly aligned and TP1 inside.
Thank you for your observations.

Yes proper settings are important.

JoeeF Jan 26, 2020 12:38pm | Post# 2218

This week was bad for me too, in all 5 configurations used. I have identified another repetitive pattern like in the picture below. {image} In some cases the SL is smaller then 10 pips and TP is bigger then ATR. Not so good. A fix could be something like: Do not trade if SL < x pips Do not trade if TP > x pips At the time of the transactions, the bands were perfectly aligned and TP1 inside.
It's easy to implement something as this. However, which value of X will you use?

The danger of overfitting will be high.

Nomenix Jan 26, 2020 1:06pm | Post# 2219

You could calculate % of ATR to use limit of SL and TP.

A value under 15 pips for SL is a no go for me because SL it is almost always hit.
A value for TP more then 50-75% of 20 days ATR the same. Depending of the evolution of market until then.

JoeeF Jan 26, 2020 1:27pm | Post# 2220

You could calculate % of ATR to use limit of SL and TP. A value under 15 pips for SL is a no go for me because SL it is almost always hit. A value for TP more then 50-75% of 20 days ATR the same. Depending of the evolution of market until then.
This sounds good ...

Take care.


© Forex Factory