Hi Audrey,

I rented your EA today and was doing some backtesting using TDS (99.9%). Thanks for your hard work, but here is my problem.

I tried to mimic your test results on GBPUSD with standard setting.

However, there seems to be a huge drawdown that kills the account at some point. The settings are identical with yours and I don't get why I get different results because you state that the standard sets pass the 20 year test.

And it looks like it hits the stoploss when this happens because if I look at the "Type of transaction" on the result page, every profit/loss is "t/p", but this huge drawdowns are either "s/l" or "close at stop".

Please see attached images.

Please help

Hi, yount79!

Thanks for renting. My set files are an example, if advisor in the test received a strong drawdown, try to reduce the size of "Fixed trailing size" or independently optimize your sets.

Also, the test results are heavily influenced by the size of the spread. In my tests it does not exceed 20 pip.

What is your spread in the test?

It is even more important to use in real trading, demo and test ECN account, because on other accounts the trailing stop will not work because of the broker's limitations.

And most importantly - test the advisor solely on the quotes of your broker, downloaded historical data from other sources have nothing to do with trading on your broker and the results of the test and the real will be very different.

My broker is true ECN. My GBPUSD spread is about 0.6pips and I put 0.8pips in the tester, just to give it some room.

So, if a position is still open at "End trading time", what will happen to it? Does EA just closes it with a giant loss??

or TP, or SL, or loss of deposit. To avoid worrying about a large drawdown, create sets with your own risk and stop-loss conditions on your own. All in your hands. Marauder is a constructor.

So, how do you get the quotes from your broker for say past 10 years???

My brokers have it:)

f your broker does not have deep historical data, but there are over the past 2-3 years, then this is quite enough for optimization and tests.

Yes, if the time of your broker corresponds to my message #33.

You can use any other time if you want, but you need optimization and test the advisor with a new time.

Hello 

I have just rented and am using on small live account on IC markets with the extreme set files. So far very good, 8 wins in a row with very little DD. Andrey is very helpful. I am looking forward to seeing how it goes by the end of the month. 

Michael

Hello, Michael!

Thanks for the comment and rent.

See my comments #77-80.

Thanks i see now, and also what what will be DD for Super extreme set? 

In Super Extreme set the maximum test DD was about 40-70%.

@yount79 close at stop, at the end of testing, is common: https://www.mql5.com/en/forum/153578

out of my mind, a few suggestions for your new version:

1. a right sided panel: this way you can shift your chart left and have a better readability

2. a default GMT sync of the EA: this would make it a lot easier for your differently time-zoned customers to identify the proper trading windows for each pair

and

3. a 2/3-digits support for more pairs to backtest (USDJPY?)