Backtesting for beginners

So I’ve never done this. What’s an EASY way to get started backtesting a strategy? Can I do that with my broker’s web platform, or am I in need of special software that requires I download and import data for the pair and timeframe I’m interested in? MT4 or MT5 perhaps?

Help a newb out!

1 Like

I don’t know how much of a beginner you are but personally I believe back testing is a complete waste of time.

I think you are better off studying what good traders do and practice practice practice on a demo account as much as possible to get a feel and understanding on reading charts.

Hope this helps. Good luck

Blackduck

There are several ways to do that. you can use your platform MT4/5 or some others.

MT4 strategy tester is a way to do backtesting manually, however, it’s not really practical because strategy tester is designed for testing EAs and don’t produce reports when backtesting manually as it does when running EAs.

another way is just by pressing F12 on your keyboard when your chart is opened. the chart moves forward candle by candle and you can test each candle without seeing the rest of the chart.

other ways are some software like forextester or soft4fx that are specifically designed for this purpose. if you want to backtest your strategy efficiently, it cost you some money to buy those softwares, however, they have a demo version so you can test them to see if they suit you.

you can read this if you want more information

1 Like

I concur. Back testing is difficult to get right. Forward test the strategy instead by either trading it on a demo account or by using extremely small position sizes until you are satisfied that the strategy either works or it doesn’t.

2 Likes

You can use TradingView’s replay mode to backtest, but you have to have a monthly subscription in order to use it. You can start a 30-day free trial first to check it out to see if you like it though.

I use forexthrive.com to backtest and it’s pretty decent. I give it a 7.5 out of 10 LOL

MT4 strategy tester can’t be trusted!

MT5 is better, but it also depends on the feed. If your broker is messing up the data, you are going to have fun :slight_smile:

If you want to see, if something really works test it using futures.

This is where I started with coding an expert advisor for back-testing on MT5:

Hey, you guys rock!

Did you give it a try and just didn’t see any benefits?

Great link, really.

Other than the benefit of seeing how something worked in the past, I guess you need to know what kind of market you were testing in to see if applies to what’s going on the market now. So why do I see so many claims that “This has been backtested over the last 10 years, and it generates 15% returns easily!” Does a statement like that tell you anything more than the strategy or indicator worked back then, but it means nothing going forward?

Just checked that out. That’s cool!

You jumped ship to MT5? It’s finally ready for prime time? Many users were complaining it was slow and bloated.

1 Like

Thanks. That is nice of you.

A great way to test a strategy rather than back testing is to invest in Forex Tester version 3. The beauty with this is you can practice day night weekend whenever and you can test on anything from a day chart to a tick chart at whatever speed you want.

Great tool. Well worth the $$$$$.

Happy trading.

Cheers

Blackduck

1 Like

@dudebro I am not using Metatrader. but if I have to it will be MT5

Statements such as these tell you absolutely nothing. Did they run a Monte Carlo Simulation to assess risk? What was the outcome of that? What sort of curve fitting did they do to obtain these results? Was the performance consistent when using tick data? If not, why not? What are the margin requirements used to obtain these results? How skilled are they in backtesting? All these questions and more remain unanswered.

EAs perform very differently under real market conditions when compared to their backtested results. If backtesting was a reliable measure of performance then this would not be the case. Backtesting leads you into a deep rabbit hole. Best to avoid it altogether, IMO.

1 Like

the metatrader 5 platform is best suited for backtest

Manual backtesting - hard work for sure - but youll learn so much about the market as you do it.

Sorry but i completely disagree that its a waste of time.

How on earth can u have confidence in a strategy if you dont backtest it?

With simple mechanical systems backtesting is straight forward, the more bells whistles and discretion used yes it becomes almost impossible to backtest.

1 Like

Because there is a margin of error you think it should be avoided?

Hmm

Are you using special software for backtesting?

@dudebro

No I did mine manually in Excel, it’s a hard slog.

Tnak you for this topic. I don’t have to creat mine now.

I can’t even grasp how that’s done. :thinking:

you mean you recorded the trades you took in Excel, right?

I’m thinking more than just trades, the actual currency pair data, right, @Johnscott31?