- Is the advisor suitable for real life?
- MT5 EA Testing results vastly different
- How do you algorithmise the detection of MA feed clusters?
depends on the historical data that you are using. I have proven my trades on live account and then on strategy tester are over 95% the same in any period, however, I do not trust any of the statistics in the reports regarding drawdown and equity. I have to run the new ea on demo with an indicator that shows my equity in realtime, for few months before I can trust any new ea. They equity/DrawDown statistics are NEVER anywhere the same as what says in strategy tester.
With any other broker, I wouldnt trust strategy tester results until I was able to modify my tick data and compare with live trading on that new broker. If the trades are over 90% the same between live trades and backtest trades, then, I would trust that broker and backtest too.
"The equity/DrawDown statistics are NEVER anywhere the same"...
By this, I mean that in strategy tester, a typical example would be that in strategy tester drawdown statistic will say 5 - 15%, which is very good, right? but when running an equity indicator during that same date period, my floating P/L will be constantly, over 40 - 60% negative value of P/L/(Account Balance). That is VERY BAD. and a major inconsistency.
But thats not to say that strategy tester can not be used during testing of a new ea or a new strategy. You just have to be acutely aware that it is just 1 cog in the testing process... By that I mean that testing in strategy tester is still good to use for many reasons, including to test your code. And there is still a lot of numbers in the report that are still good to trust, such as the profit factor and number of trades etc etc...
I would say the tester is for testing an EA/ indicator whatever…. NOT for evaluating the EA… tester is just to make sure that EA is doing what it is designed to do
if that was true, then it would not be called a backtest. However, I would agree that using strategy tester is not good for testing to find worst drawdowns. Which is the main reason that we want to do a backtest, right? But even without that statistic, the backtest report shows many other numbers in the report that ARE accurate and essential to know, AND that are very hard to get from any other testing method.
if that was true, then it would not be called a backtest. However, I would agree that using strategy tester is not good for testing to find worst drawdowns. Which is the main reason that we want to do a backtest, right? But even without that statistic, the backtest report shows many other numbers in the report that ARE accurate and essential to know, AND that are very hard to get from any other testing method.
so your answer to the ops' question is no, for a short answer? haha. but i agree completely. all good points, however, if strategy tester was not for backtesting, then the backtesting would not be called that :) You should not tell users that it is not for backtesting when that is eactly what it is made for, no matter how badly it fails to do that. I use it for 90% of just ea testing, and not for backtesting, however, that is only due to past experience, but it is still a backtest, because that is what it is for. a spork is still a fork no matter how much it fails to do its original task :)
if that was true, then it would not be called a backtest. However, I would agree that using strategy tester is not good for testing to find worst drawdowns. Which is the main reason that we want to do a backtest, right? But even without that statistic, the backtest report shows many other numbers in the report that ARE accurate and essential to know, AND that are very hard to get from any other testing method.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use