Possible EAs for LIVE (By Siitari)
Lietotājs ir dzēsis šo sistēmu.
Diskusija
Oct 19, 2010 at 16:34
1,013 Views
11 Replies
Biedrs kopš
111 ieraksti
Oct 26, 2010 at 09:29
These are some of the EAs that I'm trying out to see what kind of performance they have. Idea is to start a live account based on these results.
Currently on account and most likely to be put on LIVE account:
Forex Crescendo
- slow, steady
Forex Retribution
- very slow, steady(could use bigger risk?)
Automated Pips
- slow, steady(could use bigger risk?)
Forex Complexity
- faster but more risky(not able to use lots smaller than .1 for some reason)
XPA
- extremely slow, swings back 'n forth
Currently on account but not sure will survive LIVE(need to study more):
TraXXon(EA BOSS)
- riskier
EA Kain
- riskier
Smart Scalper
- still studying
Of course risks are closely related to settings(SL, TP, lot size/risk %) so it's hard to compare.
Changed risk settings on Supreme Complexity(.6 lots is a bit too much for account of this size) and using custom start date beginning after that change and replaced SteadyWinner(makes many winning trades but losing trades wipe the profit) with Smart Scalper.
I have tested a lot of other EAs too and... well, most are crap. I think I should put MD on test again since some versions of it are profitable.
I also have Blessing3 on test and there's a huge potential there but it comes with big risks too(thought with substantially smaller risks it still can perform rather well).
Currently on account and most likely to be put on LIVE account:
Forex Crescendo
- slow, steady
Forex Retribution
- very slow, steady(could use bigger risk?)
Automated Pips
- slow, steady(could use bigger risk?)
Forex Complexity
- faster but more risky(not able to use lots smaller than .1 for some reason)
XPA
- extremely slow, swings back 'n forth
Currently on account but not sure will survive LIVE(need to study more):
TraXXon(EA BOSS)
- riskier
EA Kain
- riskier
Smart Scalper
- still studying
Of course risks are closely related to settings(SL, TP, lot size/risk %) so it's hard to compare.
Changed risk settings on Supreme Complexity(.6 lots is a bit too much for account of this size) and using custom start date beginning after that change and replaced SteadyWinner(makes many winning trades but losing trades wipe the profit) with Smart Scalper.
I have tested a lot of other EAs too and... well, most are crap. I think I should put MD on test again since some versions of it are profitable.
I also have Blessing3 on test and there's a huge potential there but it comes with big risks too(thought with substantially smaller risks it still can perform rather well).
Less effort, better results.
Biedrs kopš
1430 ieraksti
Oct 26, 2010 at 09:33
Are you saying that you have all the EA's mentioned on this one account?
You do realise that that isn't the best way to test the EA's? Unless of course what you are planning to do is put them all together, live on one account.
You do realise that that isn't the best way to test the EA's? Unless of course what you are planning to do is put them all together, live on one account.
11:15, restate my assumptions: 1. Mathematics is the language of nature. 2. Everything around us can be represented and understood through numbers. 3. If you graph these numbers, patterns emerge. Therefore: There are patterns everywhere in nature.
Biedrs kopš
111 ieraksti
Oct 26, 2010 at 09:35
That's the plan :)
I know, I know, it's insane and so on.
This account is to test how these all work at the same time on a single account.
I know, I know, it's insane and so on.
This account is to test how these all work at the same time on a single account.
Less effort, better results.
Biedrs kopš
111 ieraksti
Oct 26, 2010 at 09:37
Supreme Complexity is my main concern ATM since I'm not able to use smaller lots than .1, for some reason. I think it's coded in. Dunno.
Less effort, better results.
Biedrs kopš
1430 ieraksti
Oct 26, 2010 at 09:41
OKay, well a good starting point might be to backtest them all together. You can use Report Manager to combine backtests from MT4.
https://www.mqlsoft.com/download/reportmanager
https://www.mqlsoft.com/download/reportmanager
11:15, restate my assumptions: 1. Mathematics is the language of nature. 2. Everything around us can be represented and understood through numbers. 3. If you graph these numbers, patterns emerge. Therefore: There are patterns everywhere in nature.
Biedrs kopš
111 ieraksti
Oct 26, 2010 at 09:42
(labots Oct 26, 2010 at 09:45)
Thanks, I will have to try that one.
EDIT: Must say that I've found backtests to be more of way to just see if EA works at all. Forward tests are far better and I've got time ^^
EDIT: Must say that I've found backtests to be more of way to just see if EA works at all. Forward tests are far better and I've got time ^^
Less effort, better results.
Biedrs kopš
1 ieraksti
Oct 26, 2010 at 11:16
I have Crescendo, Retribution and Automated Pips and Forex Autopilot running on one live account at the moment.
The account has been running since 4th October and is up 23.6% in that time. By far the biggest earner is Forex Autopilot, although that one has the most aggressive settings. The others don't seem to set stop losses, so I keep them on a tighter reign.
I was thinking of getting rid of Retribution as it hadn't done anything, then it took 7 profitable trades this morning :)
Good luck with the trading.
The account has been running since 4th October and is up 23.6% in that time. By far the biggest earner is Forex Autopilot, although that one has the most aggressive settings. The others don't seem to set stop losses, so I keep them on a tighter reign.
I was thinking of getting rid of Retribution as it hadn't done anything, then it took 7 profitable trades this morning :)
Good luck with the trading.
Biedrs kopš
111 ieraksti
Oct 26, 2010 at 11:25
Retribution is a slow gainer, looks for those 'good' opportunities. It's nothing like they promise on their site but seems profitable so far...
I'll have to take a look at Autopilot, thanks for the info.
Nice gain btw and wish you good luck in the future too.
I'll have to take a look at Autopilot, thanks for the info.
Nice gain btw and wish you good luck in the future too.
Less effort, better results.
Biedrs kopš
1430 ieraksti
Oct 26, 2010 at 11:39
(labots Oct 26, 2010 at 11:39)
Siitari posted:
Thanks, I will have to try that one.
EDIT: Must say that I've found backtests to be more of way to just see if EA works at all. Forward tests are far better and I've got time ^^
True, but you'll be a little frustrated if it wipes the account after three months of forward testing and you realise a consolidated backtest might have indicated this could happen.
11:15, restate my assumptions: 1. Mathematics is the language of nature. 2. Everything around us can be represented and understood through numbers. 3. If you graph these numbers, patterns emerge. Therefore: There are patterns everywhere in nature.
Biedrs kopš
111 ieraksti
Oct 26, 2010 at 11:47
(labots Oct 26, 2010 at 11:48)
:) Note taken.
So far it seems that they work largely at a bit different strategies and don't really open up lot of positions in the same direction at the same time(all the different EAs together). So basically it should work out just fine.
Just need to run them at lesser risk all the time if the unfortunate would happen. Or when it is going to happen :)
So far it seems that they work largely at a bit different strategies and don't really open up lot of positions in the same direction at the same time(all the different EAs together). So basically it should work out just fine.
Just need to run them at lesser risk all the time if the unfortunate would happen. Or when it is going to happen :)
Less effort, better results.
Biedrs kopš
1430 ieraksti
Oct 26, 2010 at 11:54
In general when mixing EA's here is what I have found.
EA's that perform well during volatile periods mix well with EA's that perform well during less volatile periods. While one is raking in the profits the other one is going through it's lean period and vice-versa.
Mixing EA's that trade the Asian session and importantly that get their trading finished before London opening mix well with EA's that intra-day trade London and New York.
Ea's that use Martingales, pyramid staking or that in general hold positions for longer periods of time will eventually wipe your account if used together in my opinion. Let's face it, most of these EA types are good enough at wiping accounts on their own. Mix them at your peril!
EA's that perform well during volatile periods mix well with EA's that perform well during less volatile periods. While one is raking in the profits the other one is going through it's lean period and vice-versa.
Mixing EA's that trade the Asian session and importantly that get their trading finished before London opening mix well with EA's that intra-day trade London and New York.
Ea's that use Martingales, pyramid staking or that in general hold positions for longer periods of time will eventually wipe your account if used together in my opinion. Let's face it, most of these EA types are good enough at wiping accounts on their own. Mix them at your peril!
11:15, restate my assumptions: 1. Mathematics is the language of nature. 2. Everything around us can be represented and understood through numbers. 3. If you graph these numbers, patterns emerge. Therefore: There are patterns everywhere in nature.
Biedrs kopš
111 ieraksti
Oct 27, 2010 at 07:42
(labots Oct 27, 2010 at 07:44)
So far Automated Pips and Retribution have the highest winning rates but smallest gains overall, just few pips at a time. I will have to up the risk on these two.
Changed the start time to the beginning again. There's Supreme Complexity at work with too high risk :) It still has got too high risk since it doesn't trade smaller than .1 lots.
Changed the start time to the beginning again. There's Supreme Complexity at work with too high risk :) It still has got too high risk since it doesn't trade smaller than .1 lots.
Less effort, better results.
*Spams netiks pieļauts, un tā rezultātā var slēgt kontu.
Tip: Posting an image/youtube url will automatically embed it in your post!
Tip: Type the @ sign to auto complete a username participating in this discussion.
