Does backtesting even work?

[@trustfultrading]

Hope this was helpful 😀

[@XaviAndreu]

I love watching your videos, thank you very much!

[@weiSane]

Tobey I missed your dry humour in your videos 😂. I hadn't watched in a while.

[@MQLwithBrian]

I use backtest to 1. Check recovery factor and profitability in the long run. 2. Check if EA don't blow the account. 3. Check custom functions if working 4. Check for max drawdown. Then solve problems to limit risk. 5. Test it live for small amount of money and solve problems as the EA run until the EA is safe to use for bigger capital

[@trustfultrading]

Nice! Do you use the default broker data?

[@MQLwithBrian]

Yes. And spread in live data is lower than in historical data.

[@OneCreator87]

Thanks! Waiting for your tutorial on meaningful backtesting :)

[@trustfultrading]

Thank you!

[@klaasvk8365]

'glasses on steroids' HAHAHAH this guy

[@trustfultrading]

;)

[@angelchaveschinchilla7787]

Good video, I'm developing a Expert Advisor with good statistics in backtesting. I would like to know a little more about those concepts that you think are important for a good backtesting. I hope you upload a video soon. Thanks!

[@GascanNBK]

do you have a public git repository or bitbucket for your source code?

[@trustfultrading]

No I don't, sorry

[@richardbokele830]

Thanks Tobby.. For me It works very well with higher time frames and with strategies that open positions after previous candle close

[@damienbikoun7058]

Yes, I found it very accurate to use higher tf, 1h + and strategies that initiate trade at candle's close.

[@Island_Algorithms]

Hi Tobi, some great points made. I personally use the backtester to make sure my trades open when the conditions are met and to also ensure they close when specified. Once I am happy with that I will then run a full a back test with 100% data. If that seems to go well then the only way I know for sure I am happy with my EA is to run it in a demo account and monitor it under certain scenarios like - market open times, high impact news events, ranging markets etc. That way I know what to modify or add to my EA.

[@trustfultrading]

Totally agree. Letting it run on a demo account is also a "backtest". Its just very slow but you can trust the results 100% because there is no way to trick yourself or overoptimize :)

[@trudonkatende]

Hi toby, thanks you for this topic of backtesting, very helpful video. I'm still waiting about multi currency EA video😄

[@trustfultrading]

I have not forgotten ;) Thanks for watching!

[@reallifeanarchist7578]

Feel like a creepy chart stalker now. Thanks ;)

[@trustfultrading]

You're welcome

[@ah89971]

What I found it to work is to use high quality tick data that allow you to tune the slippage. So you get the correct ticks and slippage with assumption that the other fees are constants. Those are the things that make your live trades different from the backtesting. Getting the correct slippage is impossible but enough to get the average. The slippage at news is totally different story.

[@trustfultrading]

Good point!

[@tamih4256]

terrible example. the neighbour leaves house following a routine schedule which u observed to be around 730-8pm. the market does not have a routine schedule or movement, it's random. So it will be like your neighbour randomly leaving the house any time of the day, and after observing him/her out through the window for 1 year, you still wouldnt be able to guess when's the next time your neighbour would randomly leave the house

[@trustfultrading]

If the market were 100% random, trading would make no sense and would not work. What we are looking for is the deviation from randomness. Of course, the example is very simplified, but the point is to discover a pattern or anomaly by observing. For example, the probability that the neighbor leaves the house between 7-8 is 6% (instead of the expected 100/24 = 4.166%). The deviations are of course much smaller in reality and not as obvious as shown here in the video ;)

[@Frank-si2jd]

Yes, you’re completely right on this, but in my opinion you can only work with the 1 year of actual real tick data from your broker. Of course if you work with a good broker. Downloading external data for more years is often not complete as well and will provide distorted outcomes. So for the 1 or 5 minute chart used in day trading, this high(er) quality tick data of “just” 1 year should be enough. Of course for swing traders this will not work.

[@jmcanbal]

BT or Optimization?

[@trustfultrading]

Both. Optimization is just many BTs in a row.

[@mikalgaren9082]

1:22 ...binoculars

[@maffysmods9718]

Congratulations Toby on making a great video! It's always impressive to see individuals like you sharing their knowledge and experience with others in the community. Your insights on backtesting and algorithmic trading with Metatrader 5 are valuable and can help others improve their own trading strategies. Keep up the great work!

[@veiculoavenda6678]

congratulations ! Luciano, brazil

[@Cola_77]

Dein Ref Link zu Darwin Zero ist unterbrochen

[@trustfultrading]

Danke, aber ich weiss nicht warum... Beim anderen Link ist auch ein ?.. Hmm. Mit dem Code bekommt man trotzdem die 20€ discount. Aber vielen Dank für die Info!

[@soellhussain1519]

Make a bot on macd

[@salvator3e906]

Terrible example dude 😂😂🤣🤣

[@trustfultrading]

Thanks for watching, can you explain why?

[@salvator3e906]

@@trustfultrading Never mind, your tutorials have helped me a lot, please keep it up, thank you❤