1 Min Trading Strategy

Forums ProRealTime English forum ProOrder support 1 Min Trading Strategy

  • This topic has 219 replies, 29 voices, and was last updated 6 years ago by avatarLeo.
Viewing 15 posts - 1 through 15 (of 220 total)
  • #40559

    Good Day,

    I have seen many people asking for a 1 min trading strategy so I decided to make one.

    I personally  don’t like HFT type strategies as the spread can really eat away at profits compared to a nice trend following strategy.

    The strategy is based on candle formations, along with a Long/Near Moving Average Trend Check and a RSI[2] check.

    The trigger is made by a Bullish (or Bearish) candle formation and if the MA and RSI checks are valid a position is opened in the direction of the trend.

    Some candle patterns are counter intuitive and hence some should be commented out (this would require some testing).

    It is currently configured for long only trades (although the code is included for short trades).

    I don’t think it is exceptionally effective, but maybe someone here can make something worthwhile out of it.

    Results on 1min for thr EUR/USD with spread set to 0.8:

     

     

    #40607

    hello !

    Thank you for sharing !

    Have you tested with WF ? Why not activate the sell ?

    thanks !

    #40608

    I try to test it but I have the division by zero problem on this strategy 🙂

    #40614

    I checked the optimisation (good) and tried the SellShort (not great, but maybe we can improve?) and I’ve set it going Live … it’s okay I’ll watch it as soon as it triggers.

    It would be useful and interesting if  (somehow easily?) we could GRAPH which of  the candle patterns has formed at Trade entry.  I guess you’d need Bull 1, Bull2, Bull 3 … etc and then use DRAWTEXT? And then GRAPH Bull 1 OR Bull 2 … etc (if even allowed?).  Naah not worth the effort, just a few thoughts anyway?

    Many Thanks for your inspiration, time and effort and sharing with us all juanj
    Regards
    GraHal

    PS  I’ll let you know if I get the divide by zero error TheAccountant.

    1 user thanked author for this post.
    #40621

    this is the backtest with 200k :

    #40633

    As promised TheAccountant … above Bot (with no changes) triggered a Live Trade with NO divide by zero error.

    1 user thanked author for this post.
    #40635

    I have been thinking about the whole idea of a 1 min (scalping type strategy) and the important factors to consider should be as follow:

    1. In Direction of Primary Trend i.e. Long above and Short below 100-250MA (Long MA)
    2. In Direction of Near Trend i.e Near MA pointing in direction of Trade
    3. Strong Momentum i.e. RSI and or Stachastic in Overbough/Oversold area
    4. Good Trigger i.e. Reliable Candle Stick Pattern
    5. Small Target (i.e. <= ATR)

    Also worth pointing out would be to use a market with a small spread <= 0.8

    I believe my strategy over complicates the above.

    Instead of using a whole library of candle patterns lets identify 1 to 3 ‘reliable’ ones such as the recent Hammer-Negated pattern posted by Francesco.

    Also lets use confluence of Momentum Indicators such as Stochastic and RSI to identify strong Momentum

    And finally using ATR to identify a likely Target

    I don’t have the time right now to try this but will look into it sometime this weekend.

    #40640

    Good and logical ideas juanj!

    it would seem sensible to use those candle patterns that are statistically proven to be the most reliable in predicting market direction after the pattern formed.

    I’ll find some links for above and post on here.

    Cheers
    GraHal

    #40653

    Okay so I got distracted by the idea and ended up coding something in line with the above logic.

    Code is for Long only but pretty sure we can adapt it to be bi-directional.

    I really like the results and think we are onto something here! EUR/USD (spread 0.8):

     

     

    1 user thanked author for this post.
    #40666

    It’s weird I do not have the same results at all ! It’s not bad but it’s different ! 🙂

    I can test on 200k but it will be necessary that we have the same results on 100k first lol

    #40671

    Have you got 0.8 spread set (same as juanj) on the backtest engine?

    Also I don’t think juanj has stated exactly what instrument he is testing on … CFD or spreadbet etc??

    I couldn’t get same results as juanj either.

    Let’s keep at it now we are gaining momentum? So many of these collaborative / improvement projects seem to fall over?

    Thanks
    GraHal

    #40675

    Hi, I suspect it could have something to do with the stop/target being specified in $profit and $loss instead of Pprofit and Ploss. Perhaps we should standardize to points x*pipsize.

    TP = x*pipsize

    SL = TP*2

    Set Target pProfit TP

    Set Stop pLoss SL

    For the record it is standard CFD. 

    #40678

    “Have you got 0.8 spread set (same as juanj) on the backtest engine?”

    yep 🙂 and Im on CFD too.

    #40680

    You both with IG?

    I am with IG and my PRT Platform is direct from IG.   Attached is latest version etc.

    #40692

    Hi,

    This is what I have from my side.

Viewing 15 posts - 1 through 15 (of 220 total)

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