Differences in system´s profit

Forums ProRealTime English forum ProOrder support Differences in system´s profit

Viewing 7 posts - 1 through 7 (of 7 total)
  • #220450

    Hi all,

    I just noticed that in several of the autos I have active in real, the profit that appear in the Proorder Table doesn´t match the profit that appear in the detailed report (the one with all the operations and other stuff). It happens just in a few of the systems that I have active but is quite disturbing because I don´t really know wich one is the right one.

    It´s easier to show the issue than try to explain it so I am going to attach two images with that difference in one of my systems in real

    Does it happen to anybody else? Anybody know the reason of this difference and which one is the right one?

    Thanks

    #220452

    I am trying to send the second image but for some reason it gives me error. I keep trying..

    #220472

    The part of PRT’s database where your trades are stored, gets corrupted when Systems are thrown out (killed from the server – see below).
    On a daily basis the historical (!) data may even change.

    For me the Detailed Reports have become totally useless. I have mentioned this (via ticket and all) over and over and over.

    I know of no way to check the real figures – so I just don’t know them. 🙁 🙁 🙁
    The only thing you can do is check at the global level at the Broker. But that does not help you much in keeping track of the performance of your Systems …

    Another issue is that the Detailed Reports don’t calculate consistently. This is a different issue than mentioned “On a daily basis …” (above). So I have a separate screen with a bunch of important Detailed Reports, but I am always manually refreshing them in order to get the data right or at least better. Refreshing : change from AutoTrading (filter) to All Trading and back.

    The fact that a System is thrown out causes yet another anomaly, because the trade involved (when an open position was there when the System was killed) becomes a Manual trade instead of an Auto trade.

     

    1.  The detailed trade data gets corrupted (like 20 trades started at the same date/time while in reality it was only one).
    2. The totals are accumulated wrongly.
    3. The maximum of 1500 trades in the Detailed Report makes you blind anyway (but first have 1500 trades in there).
    4. The 500 trades maintained in the List of Executed trades (in the Orders list)- same thing but earlier.
    5. The 1-2 days of history kept in the Cancelled/Rejected tab makes you totally blind after 2 days when something went wrong. There is nothing to communicate over with PRT support (though they can see more and will try their best to sort it out for you**).

     

     

    It is playing with fire.

    **): Which does not count for PRT-IG. There you just have bad luck.

    I have become numb about it all. Already the fact that 1500 of trades limits your further vision on the financials is … good for amateurs. Possibly we are just that.

    3 users thanked author for this post.
    #220482

    PeterSt,

    Thank you for such a detailed answer, I can see you have been struggling with this issue quite a lot already. I have sent a ticket also about this and maybe with enoungh people asking the PRT-team will do something about it (the back of my mind: “yeah, sure!”)

    One question though… You mention “when the systems are thrown out” or “killed from the server”, does this happen when we stop manually the system? when some error happen and the system stop? or when?

    In my experience in the majority of systems the backtest is not exactly the same as in real but similar enough. Maybe that´s the best way to track their performance…?

    #220485

    Hello TempusFugit,

    Yes, good that you sent a ticket as well. Please notice that this should be done via the Help menu – but maybe you did that already. Only then it will arrive directly at PRT’s internal support (other than your local support for your country). But again, if you are with PRT-IG then it will be difficult. In that case it is best to tell your local support about the ticket (via Help) and ask for a response.

    Your idea about observing all via Backtests, will not work well. Things can work out too different indeed. I myself am close to 100% the same happening in Backtest vs Live, but still I would never use the Backtest means for observing the real performance. Also notice that when your position quantity varies for a System, it will be hard/impossible to reflect the Commission “payed” in Backtesting. This would mostly count for IB. For IG you’d have the spread and that too can’t be dealt with reliably (though it will depend on the instrument whether it has fixed times for different spreads (Nasdaq) or not at all (Forex)). And then there is slippage …
    So No.

     

    You mention “when the systems are thrown out” or “killed from the server”, does this happen when we stop manually the system? when some error happen and the system stop? or when?

    Dozens of reasons (most spelled out by me in various topics in this forum). But like I showed in my previous post. Things get volatile (like yesterday) and IG’s systems get confused (this includes their procedural systems and methods). In that very example (2 occasions) it is harmless because those systems were thrown out while they wanted to take a position. That failed, and no harm was done to the Statistics (Detailed Report data).
    When you manually stop a System yourself while the System is in a position, the same thing happens. It will end as a Manual trade. Notice that per parameter setting in ProOrder, you can choose to let the position stay when a System is killed. But no matter what – if you deal with it further and make a profit of 1000, it has become a manual trade and won’t appear in the Stats of your System.


    For this special occasion, so that everybody can see and MUST see what a mess it is, this is all from minutes ago – so just taken from Live (PRT-IB) :

    1st screenshot is from the video-take of a trade which ends at 2125 profit (not accomplished yet here, but you will trust me).

    2nd SS is what the Orders List makes of it; that should be reliable, but is not at all. If I want to communicate with PRT over “a” situation, I just can’t because this is not reality.
    Oh, I thus have videos of a lot of this misery, but PRT is not allowed to watch videos, unless I make private YouTube of it.

    3rd SS is what I promised : The Detailed Report and its closed orders, showing a bunch of orders all started at the same date/time. This is just not so at all – it was one order taken a position only once. But hey, if you add up the numbers, it comes at the correct 2125.

    The 4th SS I promised too : this is a compilation of what the Detailed Report of my portfolio shows in the top version by default (notice the “Today” filter – which makes no sense at all).  Then I change the “AutoTrading” filter to all (middle version) and then back to Autotrading (bottom). The $2125 version is the correct one but it shows under the All filter. The number of trades is wrong and will resemble from what the 3rd SS shows. It really is one (half) trade for today only.

    And No, I did not perform any manual trades. Would I have done that too, then the puzzle becomes too complicated.

    There is more anomaly going on, but I ran out of attachments for this post. This is about the hovering over the entry arrow and exit cross, which makes no sense at all. E.g. the Exit shows a profit of $3900. And so …

    ONLY because I have a video of it, I can see what reality should be, and what I should be able to see back at the Broker’s report from this day. Otherwise I would have no single clue about what reality has been. As you have seen, ALL is inconsistent with each other and not even a bit.


    A last thing for this post :
    With over 2000 tickets on my conto, I know a bit of what can happen (badly) from Manual trading. But what I did not think of really, is that with Autotrading all goes “behind your back” hence automatically when you are not there to observe it. If I had not been there with this example (and no video), what would I have thought that my profit (or loss) actually has been ?

    ProRealTime’s behavior is very irresponsible, and I have told them many times. I am a financial provider myself (ERP System), and if I would present such a thing only one time to one customer, I would be sued instantly. Why ? well, because this is about money and what happens to it. It has to be JUSTIFIED. Nothing can be justified here. ProRealTime rather works on V12 and now V13. Seriously …

    Oh, right. I am numb to it. But some times the beast in me wakes up again. So … one should say that I better stop using PRT. No, Not because I can deal with it. But what about others ??
    So I hope this warns / helps a few. The hope that ProRealTime reads this and could commercially be terrified … No. That hope I lost long gone.  🐱‍🚀

    2 users thanked author for this post.
    #220517

    PeterSt,

    I can see from your post that the issue is worst than I thought

    Yes, I sent the ticket via the Help menu and today I received an answer. They tell me that the differences are because those systems were stopped manually at some point, as you were explaining in your post.

    And also:
    “I have alerted our technical team and they have noted this as an improvement for the next version of the Detailed Report that we will be releasing”

    Maybe all your efforts and tickets will serve a purpose eventually. Thanks for that 😉

    1 user thanked author for this post.
    #220518

    Maybe all your efforts and tickets will serve a purpose eventually. Thanks for that 😉

    Our efforts. 🙂

Viewing 7 posts - 1 through 7 (of 7 total)

Create your free account now and post your request to benefit from the help of the community
Register or Login