Try My EA! Need Group Feedback... | Page 19 | Forex Factory

Options

Go to
Permalinks
Print Thread

Related Instruments

Gold/USD

Similar Threads

USD/TRY and EUR/TRY 12 replies

Fibo Group Feedback 23 replies

Need feedback on my trading breakouts system 8 replies

One World Capital... Need feedback 2 replies

Need feedback on FXCM Educational workshops 3 replies

Attachments: Try My EA! Need Group Feedback...
Exit Attachments

Try My EA! Need Group Feedback...

  • #361
  • Dec 6, 2014 10:10am Dec 6, 2014 10:10am
Quoting kosmo
kashTrader, I believe you missed my post (as quoted below). Would appreciate your reply... a simple 'yes' or 'no' will suffice as I know you're very busy with other improvements for the EA. Thank you. {quote}

Hi Kosmo,

I should be able to add that but would have to wait as its lower priority. However, I have already built in a kill switch for my testing. You can ask the EA to stop trading after this current cycle has reached the desired Equity.

I am doing extensive testing over the weekend. It turns out that the original 30 approach gets boxed in 2/3 times this year alone. It spells disaster as you would end up loosing your entire account. I am working on the grid expansion as discussed with Pips. Also, am trying to address the trade closing scenario by closing hedged trades early. That way I envisige we would never have more than 10 trades to close. Also, the next iteration would be setting the server side TP.

I will get on to your request as soon as I can. But you can at the moment manually stop the EA from trading after the current cycle.

Thanks
  • #362
  • Dec 6, 2014 10:19am Dec 6, 2014 10:19am
Quoting kosmo
{quote} Hey Pipster, Please find attached my detailed statement. It looks like I've been too optimistic about the 10 sec cycle closure; from what I can see, it's more like 20-25 sec. I still don't mind as long as it's racking up profit. Surely, you can make more sense of the data, so I'd appreciate your comments. By the way, Kosmo is a woman, lol. {image}
Is this Demo or Live?
  • #363
  • Dec 6, 2014 10:39am Dec 6, 2014 10:39am
Quoting kashTrader
{quote} Hi Kosmo, I should be able to add that but would have to wait as its lower priority. However, I have already built in a kill switch for my testing. You can ask the EA to stop trading after this current cycle has reached the desired Equity. I will get on to your request as soon as I can. But you can at the moment manually stop the EA from trading after the current cycle. Thanks
Thanks, Kash. There's absolutely no rush. I'm happy enough that you have this feature on your list. Surely, I can stop the bot manually as I did yesterday, but having it automated would be very handy indeed.
mind over matter
  • #364
  • Dec 6, 2014 10:41am Dec 6, 2014 10:41am
Quoting kashTrader
{quote} Is this Demo or Live?
I wouldn't dare to let this loose cannon (with all due respect, Pipster) anywhere near my live account, yet. Demo, of course.
mind over matter
  • #365
  • Dec 6, 2014 11:10am Dec 6, 2014 11:10am
Quoting kosmo
{quote} Hey Pipster, Please find attached my detailed statement. It looks like I've been too optimistic about the 10 sec cycle closure; from what I can see, it's more like 20-25 sec. I still don't mind as long as it's racking up profit. Surely, you can make more sense of the data, so I'd appreciate your comments. By the way, Kosmo is a woman, lol. {image}
Sorry, girl! hard for me to tell!

Ok, after looking at your statement, it's even more amazing to me the amount of profit! It looks like the opened orders are closing within 3 seconds, so that's actually quite good. It's the pending orders that are taking the extra time, because there is so many of them. You also have commission on your account, which takes 0.06 per 0.01 lot away from profit, but won't affect gridlevels or buffers, yet didn't cause any negatives! Commissions only took away $23 from $494 gross profit, netting you $471!

It seems to me we may have stumbled upon something that we need to be careful not to overlook. You have and equity target setting of 2, so 0.2 percent. And you have so many pending orders that I think the EA has enough time leftover to comb through and look for open orders and close them, too. I didn't see any occurance of any late opened orders, though. Let me ask you to please continue running your version of the EA for at least the next week without any changes to how you're doing it now, including only running it during London.


~kashTrader... getting boxed in with 60 trades on my account would only make me lose about 30%, not the entire thing!
Too close for pips, switching to points.
  • #366
  • Dec 6, 2014 11:14am Dec 6, 2014 11:14am
Quoting tradeforlife
{quote} Your results looks impressive kosmo
Too early to get excited, tradeforlife. These are one day results. Once I see any kind of consistency, that's where the fun starts.
mind over matter
  • #367
  • Dec 6, 2014 11:26am Dec 6, 2014 11:26am
Quoting Thepipster
{quote} You also have commission on your account, which takes 0.06 per 0.01 lot away from profit, but won't affect gridlevels or buffers, yet didn't cause any negatives!
Thanks for your comments, Pipster. One correction: the commission is 0.03 per 0.01 lot, and 0.06 per 0.02.

Quoting Thepipster
Let me ask you to please continue running your version of the EA for at least the next week without any changes to how you're doing it now, including only running it during London.
That's exactly what I'm intending to do. If we're really on to something here, it's got to prove to be repeatable, first.

I'm really glad that Kash did take on board the daily target feature as I strongly believe this might be an additional game changer.
mind over matter
  • #368
  • Dec 6, 2014 12:12pm Dec 6, 2014 12:12pm
I have two running now my Alpari uk and Armada, the Alpari is set up as standard and is making a recovery! The armada account is set grid to 3.0 instead of 1.0!!
Just managed to have a look on my vps because been away all yesterday and see that my armada account is up 20% but still has a trade running its in + but not sure if it should be open!

I will keep them both running for a while and keep up update with results!

Keep up the good work
  • #369
  • Dec 6, 2014 1:16pm Dec 6, 2014 1:16pm
The way things worked out last week. I prefer whiskey over forex any night. In lay language: "Great Scotch!"
  • #370
  • Dec 6, 2014 3:10pm Dec 6, 2014 3:10pm
Quoting kosmo
{quote} Too early to get excited, tradeforlife. These are one day results. Once I see any kind of consistency, that's where the fun starts.
  • #371
  • Dec 6, 2014 4:40pm Dec 6, 2014 4:40pm
Hi Guys,

I have been doing a lot testing over today and last night. I have coded various version of the EA...
1. Expanding gird when both sides are taken out
2. Progressively making the gap in the grid larger as you move away from the initial positon

and so on....

However, after intense testing, I am back to square one. As always suspected (but with still faint hope) that this grid will give good returns almost all the time but will fail eventually blowing up your account. With the expanding versions you will surely blow your account now matter what configurations you try unless you have unlimited equity and are trading with extremely small lot size.

With the version proposed by Pips you will get different results. Some settings will not cause account crash but a large % loss. However, these % loss will happen often and ultimately blow up your account.

On the other hand there are other configuration which again will happen very infrequently but will be sufficient to blow up your account.

Unfortunately, I don't see a way around. Pips, in the period you have been trying you must have seen this happen. Infact one such unavoidable instance would happen between the 12 Nov - 4 Dec period.

Sorry guys.
  • #372
  • Dec 6, 2014 10:54pm Dec 6, 2014 10:54pm
Quoting kashTrader
Hi Guys, I have been doing a lot testing over today and last night. I have coded various version of the EA... 1. Expanding gird when both sides are taken out 2. Progressively making the gap in the grid larger as you move away from the initial positon and so on.... However, after intense testing, I am back to square one. As always suspected (but with still faint hope) that this grid will give good returns almost all the time but will fail eventually blowing up your account. With the expanding versions you will surely blow your account now matter...
I 200% agree with you friend. I have worked on some similar project in past and found the same what you has written.
My Pips Stolen By The Broker
  • #373
  • Dec 7, 2014 4:26am Dec 7, 2014 4:26am
So here's my trial of your system. I'm running Chandan655's ex4 file from post #32, your original ex4 file creates an error on my platform every time it tries to execute a trade, however the other file seems to work quite well.

http://www.forexfactory.com/puppet81/19
  • #374
  • Dec 7, 2014 12:03pm Dec 7, 2014 12:03pm
being able to backtest a system saves so much time and energy.. farward testing before backtesting is such a waste of time.. a system should first pass a backtest and then pass a farward test before going live, you have skipped step 1 which is very important and went straight to step 2, thats a recipy for disaster, code the EA again and give it the ability to be backtested.
I am what Many Dream to be but only a few can achieve, im a part of the 1%
  • #375
  • Dec 7, 2014 4:26pm Dec 7, 2014 4:26pm
Quoting kashTrader
Unfortunately, I don't see a way around.
@Kash,

With all due respect for your effort and initiative, I think the way things are going, I'm not surprised that you don't see a way around. Please don't take it the wrong way, but you're chasing an illusion here. Rather than try and make it a 100% winner, why not focus on perfecting what we already have on our hands - a tool with potential. Please, read on.

@everyone,

In my view, the logic of the EA itself doesn't need to be changed.
All that needs to be done is to eliminate any remaining execution bugs, so that it opens and closes the cycles without any hiccups.
There's no need to try and make it a 'set and forget' Holy Grail which never gets 'boxed out'.
The simple truth is that it's just a tool. The built in settings, although very basic, are good enough to start with.
It's our job to use this tool to our advantage, by selecting favourable market conditions to trade with it (simply put, the market needs to be moving).
Once the execution side is perfected, the next step could be to think of some additional features, such as different triggers or filters to activate/deactivate the EA. Then, we will have a more versatile tool, even though still just a tool...

Happy trading.
mind over matter
  • #376
  • Dec 7, 2014 5:00pm Dec 7, 2014 5:00pm
Quoting kashTrader
Hi Guys, I have been doing a lot testing over today and last night. I have coded various version of the EA... 1. Expanding gird when both sides are taken out 2. Progressively making the gap in the grid larger as you move away from the initial positon and so on.... However, after intense testing, I am back to square one. As always suspected (but with still faint hope) that this grid will give good returns almost all the time but will fail eventually blowing up your account. With the expanding versions you will surely blow your account now matter...

I feel your pain, brother.

Having been locked into similar boxes, it can be hard to see the light, but I do like where Kos is going with this...

Quoting kosmo
Once the execution side is perfected, the next step could be to think of some additional features, such as different triggers or filters to activate/deactivate the EA. Then, we will have a more versatile tool, even though still just a tool... Happy trading.
Kos, great eyes on the idea and to be 100% honest, it is completely different than my norm. Rather than trading against the cycle it looks to have the greatest success trending into the trend...

FUCKING BRILLIANT!!

I think I have a little bit I might be able to add, but will need to do it in my own specific way...might just give you a little better handle on things as well as make it a little more versatile towards other instruments...

Kind of exited about the idea, really...

Will see what I can whip up over the next short period of time and find a nice place for the posting!

Keep your chins up in this merriest of seasons!!

Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #377
  • Dec 7, 2014 5:07pm Dec 7, 2014 5:07pm
Quoting kosmo
{quote} @Kash, With all due respect for your effort and initiative, I think the way things are going, I'm not surprised that you don't see a way around. Please don't take it the wrong way, but you're chasing an illusion here. Rather than try and make it a 100% winner, why not focus on perfecting what we already have on our hands - a tool with potential. Please, read on. @everyone, In my view, the logic of the EA itself doesn't need to be changed. All that needs to be done is to eliminate any remaining execution bugs, so that it opens and closes...
Hi Kos,

Its not really about the Holy Grail but its about finding where the limitation of the current system exist. Pips pointed out that during this testing when using the current grid size he never got boxed in. However, the testing reveals that no matter what configuration you try you will always get boxed in. Also, if you use an expanding grid at one point you will get a margin call.

However, with the expanding grid, the situation happens very rarely. I am still to figure out how but I want to find out ho many times this has happened. Looking at the results it seems to me that it happens rare enough to be handleable. For example, we start with 1000 and run the system. We should be able to double up pretty quickly with the expanding grid. Once we hit 2000 we withdraw 1000 and restart. That we when we hit one of those events we loose 1000 but hopefully by then we have amassed 3-4K. So, we restart giving us a winner. The problem remains when we keep money in the account and let it grow.
  • #378
  • Dec 7, 2014 9:04pm Dec 7, 2014 9:04pm
Quoting Magix
{quote} Kos, great eyes on the idea and to be 100% honest, it is completely different than my norm. Rather than trading against the cycle it looks to have the greatest success trending into the trend... FUCKING BRILLIANT!! I think I have a little bit I might be able to add, but will need to do it in my own specific way...might just give you a little better handle on things as well as make it a little more versatile towards other instruments... Kind of exited about the idea, really... Will see what I can whip up over the next short period of time...
Pleasure to see you here and ready to roll, Magix! Thank you for taking it on board. From now on, it can only get better!

Glad you find the idea exciting. So do I. Apparently, it can be quite a constructive excercise to go beyond one's norm sometimes. You know how sceptical I am about autotrading... yet, here I am, lol. Maybe it's because I've already had some success trading in a similar way manually, that makes me so confident about what can be achieved here.

mind over matter
  • #379
  • Dec 7, 2014 9:09pm Dec 7, 2014 9:09pm
Quoting kosmo
{quote} Pleasure to see you here and ready to roll, Magix! Thank you for taking it on board. From now on, it can only get better! Glad you find the idea exciting. So do I. Apparently, it can be quite a constructive excercise to go beyond one's norm sometimes. You know how sceptical I am about autotrading... yet, here I am, lol. Maybe it's because I've already had some success trading in a similar way manually, that makes me so confident about what can be achieved here.

Kos...

Half in the bag, but 2/3rds the way through a thought and code...

We've played this a little different, thus the cause of a lot of previous misery...I think I can go dynamic with an H4 pogo and an M1 trigger...SL in play and rock this fackwagon idea.

A pilot is only as good as it's navigator...are you down?
Money Can't Buy Happiness. Poverty Can't Buy SHIT! You Choose!
  • #380
  • Dec 7, 2014 9:54pm Dec 7, 2014 9:54pm
Quoting Magix
{quote} Kos... Half in the bag, but 2/3rds the way through a thought and code... We've played this a little different, thus the cause of a lot of previous misery...I think I can go dynamic with an H4 pogo and an M1 trigger...SL in play and rock this fackwagon idea. A pilot is only as good as it's navigator...are you down?
There's really no rush, Magix... I think the EA is a good base for any add-on trigger or filter, so I'm very interested to see your magic woven into it. As long as its original logic is not lost in the process, it's all good.

Down? Let the bygones be bygones... Upwards and forwards!
mind over matter
1 trader viewing now