Hello Mr Popov,
I just wanted to say thank you for this new great improvement. Thanks a lot.

I'm using EA studio to build portfolios and what I usually do is once I have a collection of strategies I download them one by one (I would like to be able to download them all in a zip or something and not having to go downloading one by one) and once I have them all downloaded I backtest them in MT4 (I have created scripts to automate that). Only the best strategies in MT4 backtester are uploaded to a new collection in EA Studio that I use to build a portfolio EA. In my experience some of the original EA Studio strategies don't work well in the MT4 backtester and I prefer not to trade them (I can build so many strategies that I don't care rejecting those that don't work well in MT4). So, I guess with this new improvement most of the strategies built with EA Studio will work the same in MT4 backtester and I won't have to download and backtest them in MT4. That's great!

One question though: I understand the MT4 strategies that are built with EA Studio and that I download and backtest in MT4 are quite similar to EA Studio backtester. But the "problem" is I'm trading those strategies not as individual downloaded MT4 strategies, but as part as a portfolio EA whose logic is coded in your (azure?) web server. May I ask if this logic is also the same as the results I would get in the MT4 backtester? As I cannot use this Portfolio EA in the MT4 backtester we don't really know if they are performing the same as they would do if they were individual strategies. Do you think the Portfolio EA server could need fixes too to perform the same as they would do if they were individual strategies?. Alos, If we find a strategy performing well in the portfolio EA is it going to be a good idea to trade it as an individual MT4 EA or there could be diferences in results? I think the Portfolio EA feature is the best I've seen in a software to build forex strategies, although I would like to use it in the backtester.

Uf, sorry for posting such a long post, I just wanted to say thank you, but I started thinking about the Portfolio EA strategies being diferent from individual MT4 downloaded strategies. What do you think about that? Thanks!

Thanks a lot for your good work in your great tools. I'm working with EA Studio everyday and I have big expectations on it
Best regards!
Rj

Hello Popov,

I had a question about MT4's Strategy Tester.  I see from this thread you worked hard to make EA Studio back tests compatible with MT4's Strategy Tester.  EA Studio is a wonderful tool and I've returned to using it on a daily basis.  I've read mixed opinions about MT4's Strategy Tester.  Does this mean you do have confidence in MT4's back tester?

Thanks...

There is a difference in how the two programs process the data.

There will always be confusion between the two. Personally, I use MT4 as the 'bridge' to the broker, not for backtesting or optimization.

It's a good idea.  I added recently "Months on Profit %".
I want to add also "Time in position %" as stats and a criteria.

However every new stats and criteria makes the programs a little bit slower.

I am sure that the speed vs additional feature will be a better compromise. More stats means better decision making. So in case Max Equity drawdown could be in $ only as percentage does not do justice. I will give you an example. I start with a $1000 account and do a backtest and it turns out that over the last 2 years I would have lost 30% which is $1345. Now the numbers are correct as equity had gone up by so much that 1345 was 30 of the equity at that specific time. Now imagine I say 30% is not bad lets go ahead...now it has probability of going down and wiping my account...1000-1345. I would never trade that EA on a 1000 account. I might do it on a 2000 account to keep myself covered. The %age of just 30% means nothing in this case. Anyways my 2 cents but do put stats that help people make better decision. Yes 30% is fine to input while generating but very very tricky when looking at the results.

When one is new to the algo world..such stats throw new users out and like all human beings we blame someone else...in this case bad software...it said 30 percent but wiped out my $1000 account...rather than blame our lack of knowledge and understanding...

Also, please dont forget to answer Sleytus query as I have hijacked his question again.