OHLC data for scalping strategy that doesn't trade intra candle, only on candle close...?

New comment
 

Hi all,

Clearly there are huge differences in results between back testing using OHLC and 'Every Tick based on Real'. My strategy is catching small pullbacks on the 1m XAUUSD in the NY session i.e. when price is really moving. But it only opens/closes trades after a candle close. I'm not concerned about all the tick movements within the 1m candle, just so long as the OHLC is correct. My orders are also a good few pips away from these points too.

Of course I will forward test, but do you think that back testing on OHLC (or even 'Every Tick') will give me a sufficient idea as to whether the strategy is worth pursuing? It's just that everything I try using  'Every Tick based on Real' turns to crap, even when manual review and logic suggests otherwise.

What is your opinion?

Regards,

Martin.

 
Your topic has been moved to the section: Expert Advisors and Automated Trading
Please consider which section is most appropriate — https://www.mql5.com/en/forum/172166/page6#comment_49114893
 
martymoon120: Clearly there are huge differences in results between back testing using OHLC and 'Every Tick based on Real'. My strategy is catching small pullbacks on the 1m XAUUSD in the NY session i.e. when price is really moving. But it only opens/closes trades after a candle close. I'm not concerned about all the tick movements within the 1m candle, just so long as the OHLC is correct. My orders are also a good few pips away from these points too. Of course I will forward test, but do you think that back testing on OHLC (or even 'Every Tick') will give me a sufficient idea as to whether the strategy is worth pursuing? It's just that everything I try using  'Every Tick based on Real' turns to crap, even when manual review and logic suggests otherwise. What is your opinion?

It depends on the strategy entry and exit rules.

If both entry and exit are both on the open price of bars, then the OHLC test will be almost identical to real tick testing.

If however, the exits are for example, are at a fixed (or variable) sized stop (or even a trailing stop), then the OHLC tests could be vastly different to the real ticks.

 
Are you familiar with the concept of whipsaws?

 
Thank you Fernando and Tobias. Yes I am aware of whipsaw. But thast's my point, I'm building my strategy based on taking trading decisions after candle close. So yes, my position closing and opening shouldn't be affected by intra candle tick movements. I just don't understand how a strategy - that seems perfectly logical to me and is a replica of my manual rules - can show up as losing 90% on 'Every tick based on Real' but +56% on OHLC. Such a big difference. I think I'll go ewith the latter :)
New comment