Genetic Generated Strategies

Hey there,

 

I’ve had the demo now for a couple of weeks, and it expired today. I has some question I wanted to clear up.

 

I’ve figured out the gist of it without RTFM, set indicators robustness tests, etc.

 

My problem lies in what seems to be the generating. I did the testing on usual old EUR/USD and BTC/USD extensively, used all kinds of different indicators, and made sure I had correct data downloaded.

 

When I would generate strategies, it would tell me the win rate based on the historical data, it would tell me the win/lose ratio, it would tell me how many trades it made and how long the average win/loss was.

 

But as soon as I exported the strategy, opened it up in MT4, and ran the strategy tester, the results would be completely (and believe me when I say completely) different. It would say one thing in StrategyQuant, and when tested would do something else entirely. Occasionally it would spam out and constantly replace and delete trades, other times it would trades extremely often (which is good) but it would sell where it should buy (which is bad) and it would buy where it should sell (also bad.)

 

So, my question is, where does the discrepancy lie? Is it the strategy tester in MT4? I don’t want to run something in a live environment before extensive testing. It’s difficult creating a strategy, but having them modify settings automatically is wonderful, but what happens in the program after export isn’t generally what happens after import.

 

Any help would be appreciated.

 

Kind regards,

Alex.

Hello,

 

there are some essentials you need to consider when developing a strategy for MetaTrader

– you strictly need to use MetaTrader backtesting engine in SQ if you plan to trade it in MetaTrader

– make sure you use the same data for testing, developing and trading the strategy (timezone setting, etc)

– if you use limit/stop orders you must use tick data 

– strategies using market orders (trading on close) and generally simple strategies match SQ -> MT4 more easily compared to complex strategies

– you need to do an individual trade analysis to find what is causing the difference, whether it is caused by an indicator for example

 

you can post your strategy or send to [email protected] we can check more deeply