No or incorrect display of drawdown in V12
- This topic has 77 replies, 7 voices, and was last updated 1 week ago by
.
-
Ok – they said give the support Report a few days before they would give feedback. I heard nothing but left it for just over 2 weeks before chasing … maybe it’s a complex issue and they are working on it …
I sent a polite request for an update this afternoon and received the following.
> > Good day,
>
> Thanks for your email.
>
> Apologies for any inconvenience this issue may be causing and how long it’s taking to get it resolved. The team investigated but unfortunately couldn’t find the exact issue, they asked that next time the problem happens, please save the workspace under the name “FOR SUPPORT” as this will help them greatly to further investigate.> Patiently waiting for your response once the above is done so we can get your issue resolved.
>
> Regards,So, two weeks later and I get that response in a matter of minutes, only when I chase. If I had not chased, I guess the issue would have just been forgotten about and their request to me would not have been passed on.
I have no idea what they mean by “save the workspace” …. they have plenty of screenshots showing the issue. We have already shown that the issue is not related to my system or the instrument being used or the timeframe. And is happens randomly across all my different systems, so is not related to the coding of any particular system.
I have responded asking them to clarify what they want and expressing my honest opinion that the way these support requests are handled is neither professional not accountable and it makes me very concerned when I trust my money to these systems.
@, did you get anything back from them?
@Nicolas is there any light you can shed on this? Anything would be gratefully apprecated.On my side I’ll update anything I hear that might be of use.
1 user thanked author for this post.
I had a meeting the day before yesterday on this subject, among others, and so far it’s impossible for us to reproduce the same situation with the current versions. Does this problem still exist?
There are a large number of fixes and updates coming in version 12 for users, but at the moment they’re on hold because they’re being thwarted by another deployment in progress. It’s highly likely that these cases (which seem to be sporadic and few in number, if you’ll forgive me) will be eradicated with this forthcoming patch.Concerning the support request to create a workspace when the problem occurs, this is interesting, as it allows you to fix the exact moment (with which windows are open, which codes, etc.) when the problem occurs. Thank you in advance for creating this workspace and sending it to support for follow-up. All the best.
Thanks Nicolas, that is very helpful. The problem does still exist. It is random but consistent in that it always happens every x number of backtests. It might happen two times in five backtests, or just once in fifteen or so.
I understand the workspace request now. However there is no commonality here. It happens across my two complely separate IG accounts and is not reliant on timeframe of the chart. Nor is it reliant on any particular system or type of system. The experience of other users in this thread indicates that it is not reliant on the instrument being charted either. Please correct me if I am wrong, but I cannot see how a workspace file would help if that is the case. Indeed it might be a misdirection, as a workspace file from another user in this thread experiencing the same issue would be different in all the above respects. The nature of the problem though is 100% consistent – always the double counting issue which is described in detail in this thread.
My suggestion, if you agreem, is that we wait now for rollout of the update and all resources are focused on that as soon as it is possible to do so. Hopefully that will fix it. If not, we can look at it again and I will of course supply any information that will help.
1 user thanked author for this post.
This problem was already present in PRT-Complete-v11. Relaunch the backtest or even close PRT and repeat the backtest then might resolve the problem temporary or shift it to other statistics (like drawdown or number of trades). It’s like backtesting is a service provided for free if server resources are available. When server resources are in short supply (part of) the backtest is “shut down, please retry later”.
This one is funny too. The standard detailed Backtest report has all the categories filled. But when I switch to the multi-period version not one of the categories is filled. I can switch back and forth as many times as I can, one of them stays filled, and the other one stays empty. Fun with reports so to say. By the way – the export to pdf of the empty one is empty also, very consistent.
1 user thanked author for this post.
I can’t figure out what is going on at all … I got another response from IG but from Trading Services rather than Chart Services who, I thought, handled these types of support issues. This one gave me the impression PRT are still working on it but was very vague and did not answer my question regarding saving the Workspace (although Nicolas did answer it in this thread). They also directed me to the PRT Contact page in case I want to contact them direct about the issue … which seems really odd as they have never done this in the past and I didn’t think it worked this way for IG clients. Their response gave me the distinct impression they are don’t really want to follow it up.
What is clear is that there are multiple problems with V12. Nicolas has said that there is an update with many fixes pending – why this can’t be communicated through the normal support channels is beyond me. It leads to a huge amount of frustration and wasted time. Furthermore, Nicolas mentioned there does not appear to be many users experiencing the double counting issue. But that assumes everyone it is affecting would have noticed it. People backtesting less frequently or who are less experienced might not be noticing it, innocently taking incorrect results as being correct. Surely there should be a meaningul status page somewhere that notes potential bugs under investigation. It would save time and could save people trading off the back of incorrect results. IG do have such a page but I can’t recall it ever show when there are problems with PRT – not even when they are very obvious problems affecting all users. Right now it simple says PRT Charts are “operational”. OK that is true … but they are not quite operating how they should be. I wrote to IG many months ago suggesting they use this page more proactively for the benefit of both users and to stop their support from getting multiple ticket requests on the same thing. Nothing came of it …
I’m going to send the attached image to IG. Maybe it will help. Maybe there will be some more useful feedback …. maybe I’ll win the lottery tonight.
1 user thanked author for this post.
Somebody else experiencing the same issue with ‘double accounting’.
https://www.prorealcode.com/topic/etrange-comportement-dans-tableau-de-bord-2/#post-231728
1 user thanked author for this post.
Somebody else experiencing the same issue with ‘double accounting’.
Regularly (yesterday at least 3 times), as already posted in this thread. I guess that makes at least three of us: Batty, Netizen38 and me Wim. Pretty sure there are more of us out there . . .
1 user thanked author for this post.
Thanks Grahal, I also noticed increasingly indidences of the double counting yesterday. I sent an update request to IG on Saturday, with the additional screenshot. I referred to Nicolas’s helpful comment about the update that is pending but not yet scheduled and asked if they could get more information on this and whether it is likely to sort issues with the Detailed Report including this doublce counting issue. I referred them again to this thread. All just trying to be helpful and push this along.
The only response I got was a returned email in multiple languages stating “this email address (helpdesk@ig.com) is no longer monitored” and referring me to go to https://www.ig.com/sg/contact-us. I tried it again and got the same auto response. I think this is the Singapore office!
In the end I resorted to sending it to helpdesk.uk@ig.com. So far I haven’t had a response. Most likely it seems because these email addresses are not to the right people.
I will call them later today and try to find out what on earth is going on.
1 user thanked author for this post.
Find exclusive trading pro-tools on 
