EAs and real accounts - page 5

 
barry:
Newdigital,

I think it is not so easy to set the timefilter according to our brokers. Here are the problems I see:

1 You didn't say what broker data you used to set the 8 & 17 hours, so we don't know the difference in hours between your broker and ours.

I tested StepMAExpert_v1.42 with North Finance broker. If i remember Alpari should be 7 & 16 hours in this case.

2 I use Alpari data for backtesting, so when I optimise, I optimise to Alpari time.

Yes but I did not backtest this EA (StepMAExpert_v1.42). I forward tested it with North Finance demo during the several months. It was not backtesting results. So just change the time in timefilter acording to your broker.

3 Then I demo test the EA on MIG or InterbankFX or Neuimex or whatever broker we choose, and their time may be different from Alpari, so my demo testing values (on say InterbankFX) will need to be different from the values I optimised (using Alpari data). 4 Also, I think that some brokers change for Daylight Saving Time and others do not change, so we may need different settings for DST and non-DST times. This may also influence backtester results (if the broker changes for DST but the EA uses the same hours all year round).

Sometimes EAs are performing differently even if we adjust the right timefilter time or just right time or timezone ect. To say a true every EA should be optimized concerning the settings with the broker we are using. But only Alpari was provided the data for traders for backtesting back to 2001. Other brokers do not care about it (one or 3 month data is not enough to optimize the settings). So we had to optimize the settings using Alpari.

I think that this time issue may be a big issue for making our EA's profitable. For example, you are getting good results forward testing SBS and TPE EA's but I get very poor results. Maybe it is because I didn't convert the TimeZone or Start and End times correctly? Do you agree that we need to work out how to get the times right in our trading platforms, and how to confirm that we have them right?

I think the total leaders thread is just for the people who came to elite section for the first time. I am not using this thread. More usefull thread is weekly performance and the statements. The statements are the main in this case.

I am using TPE and SBS EAs with real money account and want to say the following:

1. I am testing SBS and TPE EAs with alpari demo (few months) and with SIG real (few weeks). Those EAs were very profitable during the 2 or 3 weeks and then EAs were lossing money. Now those EAs are recovering deposit slowly. But it is still profitable from the beginning of the testing period. The problem is that I attached those EAs to my real account 3 weeks ago when EAs started to lose money (I am forwad testing SBS and TPE on Alpari demo during the few months already but on my real account just for few weeks).

So, SBS may be ok but TPE lost a lot (more than -200 pips) on real account (not alpari).

2. How i estimated the TimeZone settings? I backtested those EAs (TPE and SBS) using alpari data. Then I see that if TimeZome=1 so EA is placing the pending orders at 1 am Alpari MetaTrader time. If TimeZone=2 so EA is placing the pending orders at 2. So this time is Alpari server time and I changed it to my broker (+1 hour): 0 in Alpari is 1 am in my broker (TimeZone).

3. Reliability of SBS and TPE working on real account for long way run. I am testing TPE and SBS EA on Alpari demo during the few months. Besides I am using those EAs with my broker for few months (Alpari + 1 hour). So I may compare the trades with Alpari demo and my real:

- SBS EA is ok: this EA is trading the same with Alpari demo and with my SIG real. Just few pips differencies. It was profitable with Alpari when I attached it to my SIG real account. And now the trades are almost identical. SBS is still profitable.

- TPE is showing more different results compare Alpari demo and my SIG real. Sometime TPE opened buy on GBPUSD and in the same time sell with my real account. As this EA is having big stop loss level so I am lossing money because of that. But it is just sometimes. Not always.

Of course, SBS is more reliable EA than TPE. I mean in long way run.

4. Disconnection/bad connection problems.

When we are backtesting SBS and TPE EAs so we may see that EAs are placing the pending orders at 1am Metatrader time fi we have TimeZone option = 1. Right? Right. According to backtesting results. It may be ok on demo account as well. Not fully identical with backtesting by may be ok. But it may be different with real account. Why?

One example. I adjusted TPE EA with GBPUSD to be TimeZone=1. It is Alpari broker. My real SIG broker is having the time "Alpari + 1 hour" so TimeZone is 2 for me. It's ok. But as we know the main discunnection problems for any broker is coming at GMT night. Why? No idea. It is the same with my home computer and the same with the server where I am testing elite EA located somewhere in very West Europe (to far from me). So TPE EA should start placing pending orders at 2 am for me. It started at the time. But because of "trade context is busy" or just because of something not real written in journal TPE EA may finish to place the pending orders at 2:59 am. Just start at 2am and finish at almost 3. It is real account. But backtesting? Backtesting is ok: start at 2 and finish at 2. Demo account? Ok as well: start at 2 and finish at 2:15 for example.

5. EAs which are risky ones by definition but may be very profitable. Not for long way run. There are EAs which can not run in long way. Why? Very risky. I am talking about my real account only. Which EAs? MAChannel and Scalp_net_v1.3.

Generaly talking the EAs which are working on M1 timeframe and which based on MA indicator crossing are having the high risk. But they may be very profitable.

 

Newdigital,

Thanks for your very complete reply to my questions. I will do the same as you are doing to change times to be appropriate for my broker.

I checked the times on various broker platforms today. When it was 6am at North Finance, it was:5am at Alpari,5am at MIG, and 3am at InterbankFX. Those differences may change when DST finishes.

Which broker is SIG that you are using?

 
barry:
Newdigital,

Thanks for your very complete reply to my questions. I will do the same as you are doing to change times to be appropriate for my broker.

I checked the times on various broker platforms today. When it was 6am at North Finance, it was:5am at Alpari,5am at MIG, and 3am at InterbankFX. Those differences may change when DST finishes.

Which broker is SIG that you are using?

www.liteforex.org

It i account in cents.

 

Newdigital & Igor,

A few more questions on times - sorry to keep going on about it:

In both Traders Power Expert v1.2_eurusd and Simple Daily Range Break v1.21_eurusd, TimeZone is set to 1, and ND you said that sets up trades at 1am on your North Finance platform. IBFX time is 3 hrs earlier than North Finance, so in order to set up my trades at the same moment as your trades, I need to set TimeZone 22. But Igor has cautioned against changing the TimeZone setting, because it changes the daily OHLC. What do you think is the answer here?

In 20Pips v2.1_audusd, StartTime is set to 22, so I should set that to 19 on IBFX.

On StepMAExpertv1.45 on gbp, SessionStart is 8 and SessionEnd is 24. On IBFX I will set them to 5 and 21.

On e-Monday, lHourClosePos is 10, sHourClosePos is 13 and HourOpenPos is 11. On IBFX I'll use 7, 10 and8.

On qq v2_eurusd, chas1 is 10 and chas2 is 21. I'll use 7 and 18 on ibfx.

On StepMA Expert v1.45 on chf, SessionStart is 0 and SessionEnd is 24, which I think means entry is allowed anytime, so no need to change anything.

On StepEnvelopev1.1 on the eur, TimeOpen is 0 and TimeClose is 23. On ibfx that would be TimeOpen is 21 and TimeClose is 20. I hope that's not going to confuse the system!

How am I doing so far - do my changes seem right?

Thanks again,

Barry

 
barry:
Newdigital & Igor,

A few more questions on times - sorry to keep going on about it:

In both Traders Power Expert v1.2_eurusd and Simple Daily Range Break v1.21_eurusd, TimeZone is set to 1, and ND you said that sets up trades at 1am on your North Finance platform. IBFX time is 3 hrs earlier than North Finance, so in order to set up my trades at the same moment as your trades, I need to set TimeZone 22. But Igor has cautioned against changing the TimeZone setting, because it changes the daily OHLC. What do you think is the answer here?

In 20Pips v2.1_audusd, StartTime is set to 22, so I should set that to 19 on IBFX.

On StepMAExpertv1.45 on gbp, SessionStart is 8 and SessionEnd is 24. On IBFX I will set them to 5 and 21.

On e-Monday, lHourClosePos is 10, sHourClosePos is 13 and HourOpenPos is 11. On IBFX I'll use 7, 10 and8.

On qq v2_eurusd, chas1 is 10 and chas2 is 21. I'll use 7 and 18 on ibfx.

On StepMA Expert v1.45 on chf, SessionStart is 0 and SessionEnd is 24, which I think means entry is allowed anytime, so no need to change anything.

On StepEnvelopev1.1 on the eur, TimeOpen is 0 and TimeClose is 23. On ibfx that would be TimeOpen is 21 and TimeClose is 20. I hope that's not going to confuse the system!

How am I doing so far - do my changes seem right?

Thanks again,

Barry

barry,

IBFX broker is having very different compare with alpari and north finance broker. So those EAs should be backtested a little with IBFX data before use it. And/or test on demo account.

I spoke with IBFX live chart support and they told me that it is impossible for them to set account for me without expiration date so what is why I am not testing EAs on IBFX.

You changes are right but it is necessary to test with IBFX before using it with this broker as this broker is having very different data feed.

 

I checked the trades on TPE with demo and real account

https://www.mql5.com/en/forum/173397/page14

and it is idential now. It was some problem with this before but now is ok.

 
 
 

Which EA has mild risk?

I understand that EAs that earn the most pips usually come with the highest risk, and the EAs with the lowest risk earn the least pips. What I am looking for is an EA that is in between these two extremes. Does anyone have any good recommendations for such an EA? The answer to this question is a bit hard to find in the 'Sticky" threads, thanks.

 
Yoda_Glenn:
I understand that EAs that earn the most pips usually come with the highest risk, and the EAs with the lowest risk earn the least pips. What I am looking for is an EA that is in between these two extremes. Does anyone have any good recommendations for such an EA? The answer to this question is a bit hard to find in the 'Sticky" threads, thanks.

Just something about it was described on this thread.