ProOrder stopped – stop below minimum distance (IG Index)

Forums ProRealTime English forum ProOrder support ProOrder stopped – stop below minimum distance (IG Index)

Viewing 15 posts - 61 through 75 (of 80 total)
  • #232031

    Has anyone tried coding on lower timeframes with higher timeframe datas and just using selling at market commands and not stops to combat this? And keeping those large minimum stops as a backup safety, and not a trading mechanism?

    Yes and No. But this starts with not using faster than 1 minute systems in the firsts place**. And please notice that this depends on what you start out with. Thus, if you use 1 minute systems anyway, you can easily add the slower time frames. The other way around is a kind of pain.
    So as per my previous post – yes, Market orders, but with changed systems in a fashion that they remained as profitable. This is for me not per means of faster timeframes.

    On this latter again : I see too many advices to drop to a faster timeframe; those advices somehow skip the idea of throwing out precious backtest data. But if you are used to “the past 5 years” what to do if you hop to “seconds” for the solution, and be left with a few weeks only. Thus this is only for those who don’t need backtesting (that will be more rare) or those who were used to using very little backtest data anyway.

    Then, I see no reason why the “large minimum stops” would need to remain for safety; I rather trust my own Market Order for that again, than praying whether IG will not  hugely glitch when the time comes (I experienced that and lost quite some unplanned money on it). And No, I am not eager to pay for guaranteed (IG) stops as well. Why would I, one I changed my systems to deal with that too, with Market Orders.

    **): Someone may notice some day that I sure do use “seconds” systems as well. But they are that from the ground up, plus they don’t require the backtest data. They are more technical, say like Scalping. An example of this morning below (yes I was sleeping there 😉 ). Anyway, that too is comprised of Market Orders only. And if I am well informed, this is not really scalping-like. But that thus works too after some sitting back.

     

    #232042
    Wim

    One of my algos suffers regularly from the stop-below-minimum-distance. I never understood why because I thought the minimum distance was way lower than the fixed SL distance used in the algo, 100 points for the DAX. But it appears that this minimum distance is quite often 300 points! When launching the algo the minimum distance is however announced as being 12 points. Checking the adjustment-box makes no difference for me. I have other algos on DAX and other instruments with smaller SL distances (e.g. twice the ATR) that do not suffer at all from stop-below-minimum-distance. Any ideas?

    #232044

    Checking the adjustment-box makes no difference for me.

    Are you 100% sure that you are ticking the Re-adjust Stops box for this particular Algo when you launch this particular Algo?

    #232045
    JS

    One explanation could be: if your SL distance depends on a “volatility indicator”, such as the standard deviation or the ATR, then your SL distance will “automatically adjust” to the current volatility…  (as opposed to a fixed -SL)

    1 user thanked author for this post.
    #232056
    Wim

    Are you 100% sure that you are ticking the Re-adjust Stops box for this particular Algo when you launch this particular Algo?

    I am 100% sure. Experiments have been done with a check and without a check. In both cases the same message was given as the reason for rejection.


    @JS
    – When using the ATR on M5 on the DAX, a SL distance calculated as 2 x ATR will normally be below 60. Way smaller still than the fixed SL distance of 100 used for this particular algo. It’s the minimum value of 300 that’s completely out of proportion and only gets signaled with this algo and none of the others (all running simultaneously on DAX and other instruments). I will try running this algo on another account.

    1 user thanked author for this post.
    #232062

    Does the Algo get Rejected when attempting to enter a trade or exit a trade?

    Easiest way to see this might be to run a backtest and do a cross check to the exact time when Rejection occured when the Algo was running with Live data.

    #232080
    Wim

    Does the Algo get Rejected when attempting to enter a trade or exit a trade?

    The algo places 2 pending orders, one buy at the high end of a range, one sell at the low end. It’s on M5, but it pauses for 1 candle after closing a trade, and it doesn’t add pending orders when in position. So, it places orders quite often. Total trades per day varies between 0 and 12 until now.
    I experimented yesterday with the backtest running. I launched the robot with backtest in position -> rejection; with the backtest during a pause (1 candle) -> rejection; with the backtest having outstanding orders -> rejection.
    This morning I launched the robot on another account. Big surprise, both (!) accounts are running the robot without rejections today, multiple winning trades already. On the newly engaged account there are no other robots on the DAX, only one robot on Nasdaq. On the original account there are a handful of other robots exploiting the DAX. I was curious whether that gave some interference. But after these morning results, I wouldn’t know. Must be IG that decided to keep the minimum distance reasonable today. Will track it for the rest of the week.

    #232123
    Wim

    Today, my robot got kicked out of both accounts at 9:00 Paris time, but not for not respecting the minimum SL distance. This time it was the non-respect of a minimum distance between actual price and price setting of the pending order (probably because the real DAX is closed today). Funny thing was that the imposed minimum price distance was 0 (!) on the original account and 10 on the recently selected account.
    I will relaunch the robots tomorrow early in the morning, first orders are possible after 9:00.
    Guess it’s time to migrate this profitable algo from the PRT playground to a platform that allows to check all kinds of minimum distances (the ridiculous and the normal ones) before placing an order, or that can check the go/no-go of an order and react accordingly without terminating the robot. Due to these numerous rejections/restarts I realised only 60% of the backtest profit of last quarter.

    #232124

    minimum distance between actual price and price setting of the pending order

    Was your pending order price > price + 10 (-10 for shorts)?

    Is there any way to Buy at same price point (as your pending orders) but without placing a pending order?

     

     

    #232129

    Guess it’s time to migrate this profitable algo from the PRT playground to a platform that allows to check all kinds of minimum distances (the ridiculous and the normal ones) before placing an order

    We’ll go together then.
    But Wim, it is not about being able to check, it is about another broker which does not imply this rubbish. And on top of that, deal with it yourself (read : with any other platform you can do that, but then you also NEED to do that (nothing will be arranged for you).

    En Wim, we nemen er ergens een flinke kop koffie bij, uiteindelijk gevolgd door veel bier. Als je een beetje zoekt weet je me wel te vinden. Haha. Groeten !

    #232130
    Wim

    Was your pending order price > price + 10 (-10 for shorts)?

    The pending order price for the Buy was at 6 points from the actual price at 9:00, the distance for the Sell was 14 points. So it was the Buy order that violated the minimum distance requirement of 10 on one account, but not on the other account (minimum requirement was 0 points there). This was the first time I encountered this rejection cause. Not sure whether it deserves a lot of our attention, it’s probably related to the 1st of May, real DAX closed, cfd DAX open, but with low ATR (lower than spread) and super high spread (7 instead of 1,4).
    It’s the minimum SL distance of 300 that bothers me most. It’s absurd, makes me consider launching orders without SL (but hate the idea).

    #232131

    Hi all, so sorry just to confirm, is this just an IG problem, or is it prevalent on IB as well?? Not that we can apparently get IB in Aus yet, but i was holding out on free market access, and not OTC, for trading with less problems and manipulation. Not sure if this will be the actual case but?

    #232133

    t’s the minimum SL distance of 300 that bothers me most. It’s absurd,

    300 would be only during high volatility and the enabling ‘Re-adjust stops’ is supposed to take care of high volatility periods.

    Weird that I don’t get Rejects relating to min distance. My SL & TP are as below and generally on DJI.

    Wim it would be useful for the investigation if you set yours up as below, tick the Re-adjust Stops box, but don’t tick the Guaranteed Stops box.

     

     

    #232137

    Hi all, so sorry just to confirm, is this just an IG problem, or is it prevalent on IB as well??

    No, nothing of that with PRT-IB.
    There’s a similar caveat if you trade Eurex futures but that can be solved (at a small loss of profit). By the time you are doing that for real (European Futures via IB) let me know.

    #232140
    Wim

    We’ll go together then.

    Already halfway on my side (on tea during daylight, and “ranja” or Laphroaig after sunset) . Polishing up my object oriented skills, my future is in Python or C#-like languages, skip C++-variants. Lots of Python-apis and numerous well known platforms have chosen C#-alikes. Most of these platforms can connect to a large selection of brokers, also prop firms.

    Wim it would be useful for the investigation if you set yours up as below, tick the Re-adjust Stops box, but don’t tick the Guaranteed Stops box.

    I will test this. Like mentioned before, ticking Re-adjusted Stops has been thoroughly tested with my current approach (see capture). Guaranteed Stops are never selected, already payed enough for that with guaranteed being the default when opening up an account in France 🙂
    My algo doesn’t use any indicator, just high, low, close, barindex and the clock. Levels are re-evaluated every 5 minutes and then supplied with a Buy and Sell pending order (which I will adapt according your test request). Changing to market orders will change the behavior because orders will be given at opening of new candles now. But I guess you want to see whether market orders will get rejected or not around the times that my pending orders get rejected.

    Question: Does this forum have a private messaging option?

Viewing 15 posts - 61 through 75 (of 80 total)

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