Why is that? "every tick" vs "every tick based on real ticks"

[Jean Francois Le Bas]

I backtested an EA on MT5

when I select "every tick based on real ticks ", I got a losing backtest :

but when I choose "only ticks".
I got a good result :

the EA has a lot of positive reviews, so it's actually working live I think

so why the discrepancy with real tick backtest ? I don't understand. the broker is [broker name removed], it's a good broker, not a scammer

thanks for the help

Jeff

[Sergey Golubev]

Jean Francois Le Bas:
I backtested an EA on MT5

when I select "every tick based on real ticks ", I got a losing backtest :

but when I choose "only ticks".
I got a good result :

the EA has a lot of positive reviews, so it's actually working live I think

so why the discrepancy with real tick backtest ? I don't understand. the broker is [broker name removed] , it's a good broker, not a scammer

thanks for the help

Jeff

Backtesting with "every tick based on real ticks" is the backtesting with most closed to reality concerning the data from the broker (your broker, because it may be different results with different broker ... because the brokers are not unifying the data/pric/datafeed between each other).

The article and some explanation from the post:
Testing trading strategies on real ticks and the explanation is on this post.

[Jean Francois Le Bas]

Sergey Golubev #:

Backtesting with "every tick based on real ticks" is the backtesting with most closed to reality concerning the data from the broker (your broker, because it may be different results with different broker ... because the brokers are not unifying the data/pric/datafeed between each other).

The article and some explanation from the post:
Testing trading strategies on real ticks and the explanation is on this post.

I understand that using "ticks based on history" is much better because it doesn't rely on perfect spreads, and trading conditions as MT4 does

but the problem is that the EA seems to work live (as shown by the reviews). So there is a problem here : if the backtest shows the EA losing but yet it works live ? see what I mean

or is it [broker name removed] ? But I don't know why [broker name removed] would make an EA that works losing ? or to pocket the profits themselves, while showing us losses ?

I will try different brokers and come back to you

EDIT : ok so I tried [broker name removed] and the EA is profitable using "every ticks based on history"

EDIT : I tried [broker name removed] and it's losing as well...

so does it means [broker name removed] and [broker name removed] manipulate the trading ?? I thought they were good brokers

so it seems the problem comes from [broker name removed]  

I will try more brokers

Jeff

[Jean Francois Le Bas]

Or is it broker3 that manipulates the prices to make the backtest look good ?

but then why the good reviews ? it seems to work live

[abusive language removed] I need help there

[Fernando Carreiro]

Please note that broker discussions are not allowed on the forum. I have removed their names from your posts. Only generalised broker discussions are allowed.

Regarding your issue, it is not the broker that is at "fault", but the EA. I use the word "fault" here lightly.

There is no manipulation or scamming. The Forex market is a non-centralised exchange. Each broker has their own liquidity providers and they will be different broker to broker.

EAs need to be coded to adjust to different contract specifications, spread (and other trading costs), and the actual market conditions which can be very different, especially at the tick level.

[Fernando Carreiro]

Jean Francois Le Bas #: [abusive language removed] I need help there

Please refrain from using abusive language.

[Fernando Carreiro]

Please also read the following ... Tips to avoid being scammed …

The brokers you mentioned are NOT scammers, but you need to improve your knowledge and skill to be able to understand the differences and how to properly use EAs.

[Jean Francois Le Bas]

sorry about that

not that offensive :)

so as the defense for the brokers, after looking more into it, I think the EA is a tick scalper, which would explain why it is failing on some brokers, but I thought tick scalpers would not work live at all ?

see there is a problem here, because either the backtest on broker3 (the broker where is works) is wrong and it doesn't work live, or the backtest is true and tick scalpers work live

hmm so many questions :)

[Fernando Carreiro]

Jean Francois Le Bas #:  so as the defense for the brokers, after looking more into it, I think the EA is a tick scalper, which would explain why it is failing on some brokers, but I thought tick scalpers would not work live at all? see there is a problem here, because either the backtest on broker3 (the broker where is works) is wrong and it doesn't work live, or the backtest is true and tick scalpers work live

The author of the EA may have optimised the EA for his own broker on which they trade it live.

It may even be that their broker is the very one on which you had a positive result.

If the EA provides input parameters that allow optimisation, then you may need to optimise the EA in the Strategy Tester for your own broker.

[Jean Francois Le Bas]

Fernando Carreiro #:

The author of the EA may have optimised the EA for his own broker on which they trade it live.

It may even be that their broker is the very one on which you had a positive result.

If the EA provides input parameters that allow optimisation, then you may need to optimise the EA in the Strategy Tester for your own broker.

I understand

my question is : can I trust the backtest results, even if it's a tick scalper ? will it work live ? I don't have $600 to waste

and not so long ago MT4 was not reliable with tick scalpers so please understand my wondering

[Fernando Carreiro]

Jean Francois Le Bas #: I understand. my question is : can I trust the backtest results, even if it's a tick scalper ? will it work live ? I don't have $600 to waste and not so long ago MT4 was not reliable with tick scalpers so please understand my wondering

You will have to decide that for yourself by analysing how the EA works and how "sensitive" it is to changes in conditions. You will need to do more research.

My personal opinion, is that you should first improve your own skill, knowledge and experience before using EAs.