(binned per thread starter’s request) MM (Money Maker) Detective
- #4,623
- Jul 9, 2019 2:05am Jul 9, 2019 2:05am
- Joined Jun 2012 | Status: Trader | 1,106 Posts
{quote} {quote} @Robot Trader -- I was thinking of removing all TF restrictions and take you up on your offer of OP'ing another thread devoted to the lower TF's. And, so, I just now checked your profile to see what experience you had in hosting threads and came across your "Trash Traders Thread". Too funny !!!! Why didn't you tell me? There were times I could have used a few laughs.
Some people say I'm lucky, but the more I practice the luckier I get
- #4,624
- Jul 9, 2019 2:07am Jul 9, 2019 2:07am
- Joined Jun 2012 | Status: Trader | 1,106 Posts
Trend Following System vs Return to the Average by Sergei Tarasov (Timing Solution) Food for Thoughts {file}
Some people say I'm lucky, but the more I practice the luckier I get
For the lovers of LTF, I strongly suggest to expand the ADR boundary further.
@R4M: I like to look at candles or price to improve reversal timing. We can avoid catching the falling knife on several cases.
Should you want to look at filtering with candle patterns, this is what I would use. Check this link for the consecutive candle indicator (https://www.forexfactory.com/showthread.php?p=12305180#post12305180)
It does some nice filtering.
- In the blue circles, you take only the Green or Red signal (blue marks)
- For continuation trades, you consider the magenta or or red marks
That will help control number of positions.
@R4M: I like to look at candles or price to improve reversal timing. We can avoid catching the falling knife on several cases.
Should you want to look at filtering with candle patterns, this is what I would use. Check this link for the consecutive candle indicator (https://www.forexfactory.com/showthread.php?p=12305180#post12305180)
It does some nice filtering.
- In the blue circles, you take only the Green or Red signal (blue marks)
- For continuation trades, you consider the magenta or or red marks
That will help control number of positions.
- #4,626
- Jul 9, 2019 3:13am Jul 9, 2019 3:13am
- Joined Jun 2012 | Status: Trader | 1,281 Posts
{quote} @jdva -- earlier today I revisited that issue and changed some code. In the next update -- which I haven't yet released -- I hope that has been fixed. I'll post another update before the next market open.
!
I never lose - either I win or I learn...
- #4,628
- Edited 4:58am Jul 9, 2019 4:12am | Edited 4:58am
- Joined Dec 2017 | Status: Trader | 4,378 Posts
Lower Time Frames
I wanted to share a nice result from testing at a lower time frame. The chart below is GBPNZD-H4 using optimized parameters:
The BD strategy continues to work well under a variety of different pairs and time frames -- which is good, right? But, as always, there is a trade-off. There is no single set of parameters that will work under all conditions. And this is also the case for the 100's or 1000's of EAs I've worked with in the past. Each pair and time frame must be passed through the optimizer -- this is how we "calibrate" an EA. Also, it's not a one-time task -- it should be repeated from time-to-time.
What I'm thinking of doing at first is instead of trading 28 pairs I'll trade GBPNZD at 3 time frames that I will calibrate for -- D1, H4 and H1. The reason is because it seems GBPNZD works particularly well with the BD strategy -- at least when it comes to back testing. And trading at 3 GBPNZD time frames might work better than 28 pairs all at D1. I don't know whether that will be the case -- but that's what I'll try.
Also, since @Robot Trader has so graciously offered to create yet another thread devoted to lower time frames then I'm also hoping he'll be the keeper of good settings as they get reported. Assuming 28 pairs x Five Time Frames x Four Parameters is a lot of combinations. What I'm thinking is over time we can create a "Calibration Table" with the best settings. And, then, when the BD indicator or EA is attached to a chart (pair / time frame) it will open and read a file that contains the calibration table and apply the best settings -- something like that. And, collectively, we'll attempt to keep the calibration table updated with the latest and greatest parameter settings. This isn't going to happen over night, but with enough people participating we might be able to fill-in most of the table within a month or two. Also in @Robot Trader's thread people could share how they are trading at lower time frames with the help of additional indicators. For example, I like the BB% indicator @Robot Trader showed in the screen shot of his M1 chart. And I found the source code for it.
EDIT: Sorry -- it was @skyway who shared the screen shot with the BB% indicator.
I wanted to share a nice result from testing at a lower time frame. The chart below is GBPNZD-H4 using optimized parameters:
The BD strategy continues to work well under a variety of different pairs and time frames -- which is good, right? But, as always, there is a trade-off. There is no single set of parameters that will work under all conditions. And this is also the case for the 100's or 1000's of EAs I've worked with in the past. Each pair and time frame must be passed through the optimizer -- this is how we "calibrate" an EA. Also, it's not a one-time task -- it should be repeated from time-to-time.
What I'm thinking of doing at first is instead of trading 28 pairs I'll trade GBPNZD at 3 time frames that I will calibrate for -- D1, H4 and H1. The reason is because it seems GBPNZD works particularly well with the BD strategy -- at least when it comes to back testing. And trading at 3 GBPNZD time frames might work better than 28 pairs all at D1. I don't know whether that will be the case -- but that's what I'll try.
Also, since @Robot Trader has so graciously offered to create yet another thread devoted to lower time frames then I'm also hoping he'll be the keeper of good settings as they get reported. Assuming 28 pairs x Five Time Frames x Four Parameters is a lot of combinations. What I'm thinking is over time we can create a "Calibration Table" with the best settings. And, then, when the BD indicator or EA is attached to a chart (pair / time frame) it will open and read a file that contains the calibration table and apply the best settings -- something like that. And, collectively, we'll attempt to keep the calibration table updated with the latest and greatest parameter settings. This isn't going to happen over night, but with enough people participating we might be able to fill-in most of the table within a month or two. Also in @Robot Trader's thread people could share how they are trading at lower time frames with the help of additional indicators. For example, I like the BB% indicator @Robot Trader showed in the screen shot of his M1 chart. And I found the source code for it.
EDIT: Sorry -- it was @skyway who shared the screen shot with the BB% indicator.
- #4,629
- Edited 5:10am Jul 9, 2019 5:00am | Edited 5:10am
- Joined Jun 2012 | Status: Trader | 1,106 Posts
Some people say I'm lucky, but the more I practice the luckier I get
- #4,631
- Jul 9, 2019 7:15am Jul 9, 2019 7:15am
- Joined Jun 2012 | Status: Trader | 1,281 Posts
Lower Time Frames I wanted to share a nice result from testing at a lower time frame. The chart below is GBPNZD-H4 using optimized parameters: {image} The BD strategy continues to work well under a variety of different pairs and time frames -- which is good, right? But, as always, there is a trade-off. There is no single set of parameters that will work under all conditions. And this is also the case for the 100's or 1000's of EAs I've worked with in the past. Each pair and time frame must be passed through the optimizer -- this is how we "calibrate"...
I never lose - either I win or I learn...
- #4,634
- Jul 9, 2019 9:52am Jul 9, 2019 9:52am
- Joined Jun 2012 | Status: Trader | 1,281 Posts
I never lose - either I win or I learn...
Amazing piece of work R4M
This stuff is neatly picking SR zones all over the place. Thank you for the hard work. I really appreciate it.
Entries are spot on. Exciting times ahead. This thing will pip all day and all night.
Unfortunately, exits are premature. Exit on opposite band is not working correctly. In a long trade, it is seeing the lower band as the exit. Vice versa for short trade. I am getting really impatient with you on this....lol... I thought you are smarter....
You need to add a function for setting minimum distance between trades. This should be a % of ADR. See the trades below. The bot just batched the trades around the same price. I would like it to check that the nearest trade below or above is at least y pips away estimated using x% of ADR.
This stuff is neatly picking SR zones all over the place. Thank you for the hard work. I really appreciate it.
Entries are spot on. Exciting times ahead. This thing will pip all day and all night.
Unfortunately, exits are premature. Exit on opposite band is not working correctly. In a long trade, it is seeing the lower band as the exit. Vice versa for short trade. I am getting really impatient with you on this....lol... I thought you are smarter....
You need to add a function for setting minimum distance between trades. This should be a % of ADR. See the trades below. The bot just batched the trades around the same price. I would like it to check that the nearest trade below or above is at least y pips away estimated using x% of ADR.
- #4,639
- Jul 9, 2019 12:57pm Jul 9, 2019 12:57pm
- Joined Dec 2017 | Status: Trader | 4,378 Posts
Amazing piece of work R4M This stuff is neatly picking SR zones all over the place. Thank you for the hard work. I really appreciate it. Entries are spot on. Exciting times ahead. This thing will pip all day and all night. Unfortunately, exits are premature. Exit on opposite band is not working correctly. In a long trade, it is seeing the lower band as the exit. Vice versa for short trade. I am getting really impatient with you on this....lol... I thought you are smarter....
Unfortunately, exits are premature. Exit on opposite band is not working correctly.
Yes -- I was having some difficulty getting this right. I mean, the logic and coding is simple enough and the indicator was making marks where I expected. And from my BT diagnostics it was executing the correct code. I'll revisit this.
By the way -- are you using the latest version?
Are you referring to the indicator or the EAs? Could you send me a screen shot that shows an incorrect Exit.
Also, it may be the way I'm making marks is confusing. For Exits, I'll simply mark an 'x' if it meets one of the 3 Exit conditions.