Matrix Grid Trading Community Edition | Page 107 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

IntelliJ IDEA Community Edition with MQL 4 replies

Max Draw Down EA from Matrix Grid Thread 22 replies

Matrix grid community 0 replies

Grid, Burst Grid, Grid Risk and Grid Profit 1 reply

Attachments: Matrix Grid Trading Community Edition
Exit Attachments

Matrix Grid Trading Community Edition

  • #2,121
  • Mar 13, 2015 12:28pm Mar 13, 2015 12:28pm
Hello All. These are my results from a small live account trading USD/CAD. I only had 400$ free from my standard acc. for this, so that is what I used. Josh
edit. This is this week.
Attached Image (click to enlarge)
Click to Enlarge Name: UC.jpg Size: 69 KB
  • #2,122
  • Mar 13, 2015 1:08pm Mar 13, 2015 1:08pm
I would like to make a point about this EA. It works very well, however, I have been experimenting with several EAs that are similar in nature. These would be Blessing(margin_SR1.1) and SWB(many versions) and a few others. I have tried out many modifications and I have not found a solution to their biggest problem. They all work well, but eventually lead to a death trade...even if you use really big accounts. However, all of them you can back test and you see the train wreck. Not in this one as you cant back test it. I have tried using various hedge options, break even trade options etc. So far no luck, but I am still working on it. The problem is that neither solves the issue of a trend change. Despite the fact this EA has a good trend "indicator", the way it digs itself out of a counter trend trade is by buying more in that direction hoping for a reversal. Reducing the timeframe to 1 minute is a very good idea, as it has a higher degree chance of a snap back. But sometimes, it will not come...ergo the DD and potential wipeout. Using multi lots, also good, but if its a deeper pullback, can be a double edge sword if there is no effective snap back, especially on 1 min TF on s strong move. The best way to test it to back test most volatile events for the past few years and see how it does. I still think this is probably the best performer I have seen to date, but only to be used live with caution. I hope next version will find a way to use hedging in some smart way. Best, Joshua
  • #2,123
  • Mar 13, 2015 1:14pm Mar 13, 2015 1:14pm
Quoting JoshuaHope
I would like to make a point about this EA. It works very well, however, I have been experimenting with several EAs that are similar in nature. These would be Blessing(margin_SR1.1) and SWB(many versions) and a few others. I have tried out many modifications and I have not found a solution to their biggest problem. They all work well, but eventually lead to a death trade...even if you use really big accounts. However, all of them you can back test and you see the train wreck. Not in this one as you cant back test it. I have tried using various hedge...

Fully agree...
Solutions ONLY!!!
  • #2,124
  • Mar 13, 2015 2:33pm Mar 13, 2015 2:33pm
Quoting JoshuaHope
I would like to make a point about this EA. It works very well, however, I have been experimenting with several EAs that are similar in nature. These would be Blessing(margin_SR1.1) and SWB(many versions) and a few others. I have tried out many modifications and I have not found a solution to their biggest problem. They all work well, but eventually lead to a death trade...even if you use really big accounts. However, all of them you can back test and you see the train wreck. Not in this one as you cant back test it. I have tried using various hedge...
I agree with you Joshua.

Regards,
Karl
  • #2,125
  • Mar 13, 2015 4:25pm Mar 13, 2015 4:25pm
I would like to make a point about this EA. It works very well, however, I have been experimenting with several EAs that are similar in nature. These would be Blessing(margin_SR1.1) and SWB(many versions) and a few others. I have tried out many modifications and I have not found a solution to their biggest problem. They all work well, but eventually lead to a death trade...even if you use really big accounts. However, all of them you can back test and you see the train wreck. Not in this one as you cant back test it. I have tried using various hedge options, break even trade options etc. So far no luck, but I am still working on it. The problem is that neither solves the issue of a trend change. Despite the fact this EA has a good trend "indicator", the way it digs itself out of a counter trend trade is by buying more in that direction hoping for a reversal. Reducing the timeframe to 1 minute is a very good idea, as it has a higher degree chance of a snap back. But sometimes, it will not come...ergo the DD and potential wipeout. Using multi lots, also good, but if its a deeper pullback, can be a double edge sword if there is no effective snap back, especially on 1 min TF on s strong move. The best way to test it to back test most volatile events for the past few years and see how it does. I still think this is probably the best performer I have seen to date, but only to be used live with caution. I hope next version will find a way to use hedging in some smart way. Best, Joshua


yes i agree but had a very good run
Funds

  1. Balance:
  2. Unrealized:
  3. Equity:

  1. Deposits:
  2. Net Transfers:

  1. $4,687.92
  2. -$78.26
  3. $4,609.66

  1. $3,000.00
  2. $3,000.00


Today (Sat)-0.1%-$32622/160.66
This Week18.8%$7282,045296/28115.54
This Month33.4%$1,1543,019520/50526.71
This Year53.7%$1,6102,993597/58235.67
All Time53.7%$1,6102,993597/58235.67

  • #2,126
  • Mar 13, 2015 5:49pm Mar 13, 2015 5:49pm
Quite a good run. If thats real money ellenbrook, epic run :-) Yes, the key to running it solo would be to keep taking out profits, until it blows out. Hopefully you can do well with clever money management. Now, the real question remains of how to solve the blowout issue, if its even possible. But that is going to take some doing and some hard coding. My guess is that some sort of a dynamic hedge might do the trick. But then, working with hedge can be tricky. Options are to use hedge to seek profits or to use it as a balancing act(relates to multiplier used in hedge), with a trigger either at a specified level or % DD hit. Even bigger problem, how to stop the hedge, keeping in mind 1 min TF volatility.
Best Joshua
  • #2,127
  • Edited Mar 14, 2015 6:44am Mar 13, 2015 10:46pm | Edited Mar 14, 2015 6:44am
My suggestions for EA improvement:

- Add the ability to use 2 EAs on one chart.
- Optional closing without profit in countertrend (it is obligatory now).
- Optional closing of all trades immediately if trend has changed. Not sure if this would be profitable, some testing is needed.
- MinimumStep. In period of low volatility AutoStep can be too small, in such cases it is perhaps better to open new orders with some minimal step (15-20 pips).
- Fix the bug: sometimes trend changes to unknown and EA gives false error alert that current version is old.

Quoting GoogleTrader
{quote} Someone here already tried. And seems this idea slowly killing deposit.
Well, it may sound counterintuitive but cutting drawdown is likely the way to go if the loss is small. I am testing 5% DD cut per pair (not overall DD) since february and it is not killing deposit, it makes good profit actually, 5-6 losses per month but EA generates substantially more profit in between. Manual interventions before very important news are necessary but that is not a big problem. Why only 5%? Look at it this way: with 0.01 lot per 2000 and step 15 - DD 5% in pip value is about 165 pips. DD 10% would be 2 times bigger in % value but only 1,45 times bigger in pip value (about 240 pips) and so on. I.e. there is some benefit in low DD cuts compared to big DD cuts but you also should not overdo - anything below 4-5% can be not profitable. The problem of not cutting DD is that major trend reversals without big retracements do happen, maybe 1 time per year but they are inevitable. So i also suggest to add the option of cutting DD at some % value per pair if possible. It must be worth testing.
  • #2,128
  • Mar 14, 2015 4:38am Mar 14, 2015 4:38am
It Seems that we´ll have a new version soon
  • #2,129
  • Mar 14, 2015 4:51am Mar 14, 2015 4:51am
Quoting bypasssbo
It Seems that well have a new version soon

  • #2,130
  • Mar 14, 2015 4:56am Mar 14, 2015 4:56am
Quoting fxtrader1081
{quote} I believe I am in the minority here but I believe the road we have gone down with max dd is really the only way. This is because as you said, sometimes it will never come back. With that being said, all of mine that I have let run have eventually come back as long as the lot size was small enough. The reason I believe in this way is because the ea makes a lot of money until a major move in the wrong direction and before that move it usually doesn't have a lot of drawdown. I have 2 demos with 0.01 per 1k on my profile that had max dd of 20...
I am with you on this. However, I think there is still room for live testing different lot sizes, TP settings, and from my recent experience MultiLotFactor settings (especially with TrendChange). I would also prefer to see MaxDD applied to each pair separately but appreciate this may not be possible unless incorporated into MGT CE.
  • #2,131
  • Mar 14, 2015 5:05am Mar 14, 2015 5:05am
Quoting JoshuaHope
Quite a good run. If thats real money ellenbrook, epic run :-) Yes, the key to running it solo would be to keep taking out profits, until it blows out. Hopefully you can do well with clever money management. Now, the real question remains of how to solve the blowout issue, if its even possible. But that is going to take some doing and some hard coding. My guess is that some sort of a dynamic hedge might do the trick. But then, working with hedge can be tricky. Options are to use hedge to seek profits or to use it as a balancing act(relates to multiplier...
unfortunately not a live account in the past did blow up an account i was to greedy grrr with that new account i have all the 5 currency on but when his a very high bank announcement i do stop the EA 30 minutes before then if my draw down to high about $ 300 then i stop the EA to make more trades 2 days ago even i did stop it did have _ $ 1000 was very scary but did recover after 2 days so i hope all go fine if so will go live next month but do not know if my setting are okay??????????


Attached Image (click to enlarge)
Click to Enlarge Name: Capture.PNG Size: 240 KB
  • #2,132
  • Mar 14, 2015 5:16am Mar 14, 2015 5:16am
Quoting JoshuaHope
I would like to make a point about this EA. It works very well, however, I have been experimenting with several EAs that are similar in nature. These would be Blessing(margin_SR1.1) and SWB(many versions) and a few others. I have tried out many modifications and I have not found a solution to their biggest problem. They all work well, but eventually lead to a death trade...
Maybe the solution here is to actually prevent a trade if the trend strength is above a certain % as it would indicate exhaustion with a likely impending reversal. I have two charts per pair on my setup (M1 and M15) and manually suspend trading before high impact news and when I see strong resistance ahead. This isn't always possible of course if you are stuck in a trade that is already going against you, so in the absence of a workable hedge option my suggestion would be to add a Max Draw Down setting for each pair as I think this is far better than letting a losing trade run until it wipes your account. When it is trading well it trades better than I think anyone possibly could manually so it just needs its Achilles heel addressed.
  • #2,133
  • Mar 14, 2015 7:06am Mar 14, 2015 7:06am
Quoting Deltraderboy
{quote} Maybe the solution here is to actually prevent a trade if the trend strength is above a certain % as it would indicate exhaustion with a likely impending reversal. I have two charts per pair on my setup (M1 and M15) and manually suspend trading before high impact news and when I see strong resistance ahead. This isn't always possible of course if you are stuck in a trade that is already going against you, so in the absence of a workable hedge option my suggestion would be to add a Max Draw Down setting for each pair as I think this is far...
I am not sure that would work. If you set up 50-52 - dont trade=ranging, 52-65-uptrend/downtrend, 65+ probable reversal? Interesting, but you would still have reversals and you would be kept out of many "good" trades. Should be tested nevertheless.
You could cut all trades at a defined DD point, as you have suggested but that would severely hurt the profitability of your account.(should be tested as well)

A possible solution to this Achilles heel, is to fire up a hedge as soon as your DD hits, lets say 20%. Second, you have to define a good multiplier to the hedge. - and this balance is the key. The hedge must not be too sharp or too weak to account for both scenarios - reversal or more downtrend while keeping your DD in check.
This sort of a hedge would not be in place to make profits, but to BE as soon as possible,while keeping the DD in check. After the BE the EA can start working in the right direction. If this could be worked out, a better version of hedging would need to be developed to make profit in these areas, but for now, this would be a good solution. Thats my 2 cents.

Best Josh.
  • #2,134
  • Mar 14, 2015 8:35am Mar 14, 2015 8:35am
Anyone can share, which days the trend has changed.

Thanks
  • #2,135
  • Mar 14, 2015 9:40am Mar 14, 2015 9:40am
Don't know if this is too far off topic but I was just musing about how to establish the grid with greater success.
Seems we're concentrating heavily on lot sizing, DD and the like, but maybe there could be improvements on where the trades are placed rather then just an arbitrary spacing.

So I was wondering about S/R. If trend is up place a buy at recent support. If price goes against you then 2nd order is placed at next support (obviously many different ways to figure support/resistance)
  • #2,136
  • Mar 14, 2015 10:10am Mar 14, 2015 10:10am
Quoting spyderman
Don't know if this is too far off topic but I was just musing about how to establish the grid with greater success. Seems we're concentrating heavily on lot sizing, DD and the like, but maybe there could be improvements on where the trades are placed rather then just an arbitrary spacing. So I was wondering about S/R. If trend is up place a buy at recent support. If price goes against you then 2nd order is placed at next support (obviously many different ways to figure support/resistance)
Spyderman, you should take a look at blessing ea. You can choose the way it makes the entries, PSAR,MACD, Stochastics overbought/sold etc. I agree that improvements can be made to optimize it in that regard, but it doesn't solve the bigger issue. There is a version that someone had modded to include support/resistance levels into the mix as you have described. It is Blessing_SR_Margin6xx. V1.1. I use it as a test platform to play out some ideas. Also, the cool thing about it that it has DD shutdown, margin shutdown etc.
Looks like this.
Best, Josh
Attached Image (click to enlarge)
Click to Enlarge Name: Picture.jpg Size: 71 KB
  • #2,137
  • Mar 15, 2015 2:09am Mar 15, 2015 2:09am
Quoting Deltraderboy
{quote} I am with you on this. However, I think there is still room for live testing different lot sizes, TP settings, and from my recent experience MultiLotFactor settings (especially with TrendChange). I would also prefer to see MaxDD applied to each pair separately but appreciate this may not be possible unless incorporated into MGT CE.
To keep in the spirit of a request, I send you to the ea thread. I have responded there. If I receive permission from GT to conduct conversations here on the ea then I will close that thread. Sorry for the inconvenience.

http://www.forexfactory.com/showthre...43#post8137743
  • #2,138
  • Mar 15, 2015 5:12pm Mar 15, 2015 5:12pm
Update:

As I stated my live cent account did well along with one demo out of 6 lol. Well at least one of the successful accounts was real money. As I said I might do, I am posting the statement from the live account with personal info removed. Don't get too excited. It's a cent account which started at 4800 and is now 6500 so I have made a whole $17. It is real money though so there is some value in it. I will be making a number of changes this week after watching the max dd and how it works.

I will remind people of my main set which is:

InternalTrendConfirmation=20
TakeProfit=10
Lots=depends
MultiLotsFactor=1.2
MultiLotsFactor_TrendChange=1.5
AutoStep=false
AutoInitialStep=false
Step=10

The following 3 accounts will be running that as follows

The live account will stay at 0.01 per 2500
The only successful demo of 0.01 per 1k with a max dd of 30% will stay as is
All 0.01 per 500 accounts have been deleted and I will be going with one account of 0.01 per 500 with a max dd of 50%

Then we will have a different (extreme) set which you can see in the description but it basically has a big step with a big multiplier.

I am considering maybe to add a set based on arbinka's set which is just defaults I think.
  • #2,139
  • Mar 15, 2015 5:36pm Mar 15, 2015 5:36pm
[quote=fxtrader1081;8138515]Update: As I stated my live cent account did well along with one demo out of 6 lol. Well at least one of the successful accounts was real money. As I said I might do, I am posting the statement from the live account with personal info removed. Don't get too excited. It's a cent account which started at 4800 and is now 6500 so I have made a whole $17. It is real money though so there is some value in it. I will be making a number of changes this week after watching the max dd and how it works. I will remind people of my main set which is: InternalTrendConfirmation=20 TakeProfit=10 Lots=depends MultiLotsFactor=1.2 MultiLotsFactor_TrendChange=1.5 AutoStep=false AutoInitialStep=false Step=10 The following 3 accounts will be running that as follows The live account will stay at 0.01 per 2500 The only successful demo of 0.01 per 1k with a max dd of 30% will stay as is All 0.01 per 500 accounts have been deleted and I will be going with one account of 0.01 per 500 with a max dd of 50% Then we will have a different (extreme) set which you can see in the description but it basically has a big step with a big multiplier. I am considering maybe to add a set based on arbinka's set which is just defaults I think.[/quote


thanks for the very good information
  • #2,140
  • Mar 15, 2015 11:19pm Mar 15, 2015 11:19pm
Quoting BillYon
{quote} -1.0 Negative 1.0
The negative value doesn't seem to work. At this moment I have 3-4 open transaction on EU, GU and UC.
The settings are more or less the default ones.
Attached Image (click to enlarge)
Click to Enlarge Name: Untitled.png Size: 18 KB
Using the RobotFX EAs for a while https://www.robotfx.org/
1 trader viewing now