Apparently this issue is still in present in v5.0.19.
Aren't the results suppose to match 100% regardless of which data bars are selected, on condition that the same data is used in both visual/non-visual modes? Anyway, running backtests on tick data for every strategy iteration can be impractical.
It doesn't seem logical (at all) to me that checking/unchecking the visual mode box should affect results when the all the underlying data remains the same. Given this, it is highly unlikely that this is a user or cBot issue, suggesting an issue within cTrader's backtesting logic.
As you know, backtesting is crucial for traders to evaluate cBot logic accurately. Solving this inconsistency is central for traders to have confidence in their backtesting and, as a result, being confident about their cBot development process.
Last but not least, this issue has been reported multiple times. When will this finally be addressed?
Search was removed. Now you can open log file as txt files and search there
Best regards,
Panagiotis
Panagiotis,
That I've noticed. Is this considered an improvement?
With the built-in search function only the search results that matched the search were showing. There was also a filter (trading, information, warning, error).
I don't think that such features are available in notepad or similar app.
Regarding backtesting: I see the option of having a built-in search/ filter feature of much higher applicability than an option to save the log as a .txt file.
In fact, I see both these options as complementary and not as incompatible.
Please note that, among other, it is not very practical to have to open external applications to check backtest data on every backtest iteration.
I can already tell you that, in my case, this will be a huge drawback when backtesting, since I am still expecting thousands of iterations to come.
Search was removed. Now you can open log file as txt files and search there
Best regards,
Panagiotis
Panagiotis,
That I've noticed. Is this considered an improvement?
With the built-in search function only the search results that matched the search were showing. There was also a filter (trading, information, warning, error).
I don't think that such features are available in notepad or similar app.
Regarding backtesting: I see the option of having a built-in search/ filter feature of much higher applicability than an option to save the log as a .txt file.
In fact, I see both these options as complementary and not as incompatible.
Please note that, among other, it is not very practical to have to open external applications to check backtest data on every backtest iteration.
I can already tell you that, in my case, this will be a huge drawback when backtesting, since I am still expecting thousands of iterations to come.
Apparently this issue is still in present in v5.0.19.
Aren't the results suppose to match 100% regardless of which data bars are selected, on condition that the same data is used in both visual/non-visual modes? Anyway, running backtests on tick data for every strategy iteration can be impractical.
It doesn't seem logical (at all) to me that checking/unchecking the visual mode box should affect results when the all the underlying data remains the same. Given this, it is highly unlikely that this is a user or cBot issue, suggesting an issue within cTrader's backtesting logic.
As you know, backtesting is crucial for traders to evaluate cBot logic accurately. Solving this inconsistency is central for traders to have confidence in their backtesting and, as a result, being confident about their cBot development process.
Last but not least, this issue has been reported multiple times. When will this finally be addressed?
That's a pity. It would be great to be aware of some key levels through the mobile app chart.
Are there any plans to implement this in the future?
The teams are considering this but it's a huge project. It would take time
That's good to hear! I understand.
It would be also interesting to see mobile notifications via cBots available, as an alternative/complement to the emails. Anyway, I'll soon provide several suggestions for future improvements/features and I will include these.
ncel01
27 Mar 2024, 19:23
( Updated at: 27 Mar 2024, 19:24 )
Hello,
What's the outcome here? As far as I can see the issue persists.
I'd also suggest that for indices (maybe for some other asset classes as well) the spread shown is based on the price difference (ask-bid) rather than in pips.
Why? Unlike forex, the pip size for indices among different brokers can reach a ratio as low as 1:100. Furthermore, it is not intuitive to look at a spread based on pips if not knowing the pip position.
Sometimes it is a good idea to think outside the box when it comes to asset classes other than FX. The fact is that trading encompasses a wide range of assets, not only forex.
We are looking at this but we were not able to reproduce yet. Can you advise the app and broker as well?
Best regards,
Panagiotis
Hi Panagiotis
Yes, of course.
I am using cTrader cross-broker Android v. 4.8.54456.
I am not aware this happens with other instruments but, at least for one of the US indices and/or Germany 30, the spread is not mentioned or it is wrong.
I can see that this is happening (at least) with Pepperstone, TopFX and Varianse.
ncel01
17 May 2024, 11:12
RE: RE: Problesm with Visual vs non Visual Backtesting
PanagiotisCharalampous said:
Panagiotis,
That's good to hear!
Thanks for informing.
@ncel01