Averaging Down Orders improvment

Viewing 15 posts - 1 through 15 (of 20 total)
  • Author
    Posts
  • #43195 quote
    Ludde500
    Participant
    Average

    The past 6-12 months I´ve broadly tested on the demo account all existing “strategy” on ProRealCode. Most results has been pretty bad. Except one that consistently has showed good results (winning trades 67%)
    It have had a bit to large drawdowns but has always recovered and showed good results
    Probably the one with most simple tenchology, Averaging Down Orders
    The best results have been on USD/JPY 5M, but on most other index and currencies the result has been bad.
    I don´t know programming and wonder if someone would be interested in help improve it?
    The drawdowns would need to be limited to reasonable levels, and that it shouldn´t place order during certain times and days
    You would probably also discover other improvements that the system is missing

     

    //-------------------------------------------------------------------------
    // Main code : #averaging down orders
    //-------------------------------------------------------------------------
    defparam cumulateorders = true
    
    // indicators
    EMA3 = exponentialaverage[3]
    EMA10 = exponentialaverage[10]
    if barindex>1 then
    haopen=(haopen[1]+haclose[1])/2
    haclose=(open+close+low+high)/4
    endif
    
    // first order (EMA cross)
    if not shortonmarket and EMA3 crosses under EMA10 then
    SELLSHORT 1 SHARE AT MARKET
    endif
    
    // close position with bullish EMA cross only if we are in profit and we are not already averaging down (1 position only)
    if countofposition=-1 and haclose<tradeprice and EMA3 crosses over EMA10 then
    EXITSHORT AT MARKET
    endif
    
    // averaging down
    if shortonmarket and haopen>haclose and haopen[1]<haclose[1] and haclose>tradeprice then
    SELLSHORT 1 SHARE AT MARKET
    endif
    
    // monitor the average price of whole orders and close them accordingly
    if shortonmarket and haclose<positionprice and countofposition<-1 then
    EXITSHORT AT MARKET
    endif
    jpy.jpg jpy.jpg
    #43197 quote
    victormork
    Participant
    Veteran

    maybe try to add a profit target?

    if shortonmarket then
    set target pprofit x
    endif
    I would advise you not to use an average down code. Generally you will (over time) gain more on adding up than down. Remember that the 5M timeframe is pretty short on history so just because it’s working now it might not in the future.
    #43200 quote
    Ludde500
    Participant
    Average

    It should not work, but still it does.
    I have run 75 strategy both long and short, currency, index, commodities but no one has given better results.
    Over the past 6 months, USD / JPY has gone from 114.00 February 24 down to 100.00 and up to 118.50 and down again to 109.00 with profit.
    It has cleared both up and down ….

    #43235 quote
    victormork
    Participant
    Veteran

    Just keep in mind that just because it turned around this time it doesn’t need to do the same in the future.

    #43242 quote
    Nicolas
    Keymaster
    Master

    Seems obvious that an averaging down strategy perform better than other ones which are allowed to loose.

    #43256 quote
    Despair
    Blocked
    Master

    I had to ramp up the start capital to show you a 200K bars backtest. There are evil drawdowns in the equity curve (typical for averaging down systems). You (or your account) would hardly be able to endure those in real life.

    averaging-down.png averaging-down.png
    #43279 quote
    Ludde500
    Participant
    Average

    Big drawdowns are a problem, therefore I would like to test it with a intelligent stop loss system, if there is any?
    Do you know of any good stop loss system that i could impliment in the code without any major programming skills?

    #43326 quote
    Despair
    Blocked
    Master

    There is no way in fixing this with a stop loss without the performance deteriorating. This is the effect of averaging down. You get a super high winning % but one day comes the day your account blows up (or at least takes a hard blow).

    #43333 quote
    Nicolas
    Keymaster
    Master

    The more you add orders in your grid, the faster and easier will be the return to the average price to exit in profit or at breakeven.

    #43335 quote
    victormork
    Participant
    Veteran

    @Ludde500 Also just an advise for the future when you look at systems posted on this or other forums – just because a system has got a high winning percentage it doesn’t automatically mean that it’s good system. A high winning percentage is normally the product of risking a lot of money for a very small profit. Sure it feels nice to be right all the time but it’s not always the smartest way to go. Try to read up on how you can make most money based on the amount of money you risk in each trade. Once you know this it’s easier to determinate if a system is good or not.

    Nicolas thanked this post
    #43337 quote
    Despair
    Blocked
    Master

    I saw now, that stupid me forgot to attach the screenshot.

    averaging-down-1.png averaging-down-1.png
    #43339 quote
    Despair
    Blocked
    Master

    Ok, I did not forget it but it doesn’t work!? Why can’t I attach a screenshot?

    averaging-down-2.png averaging-down-2.png
    #43341 quote
    Nicolas
    Keymaster
    Master

    @Despair

    The screenshot is here, I think there is a bug somewhere, I’m currently updating some things ..

    Despair thanked this post
    #44109 quote
    Nicolas
    Keymaster
    Master

    Everything’s back to normal, sorry for disturbing 😐

    Despair thanked this post
    #44232 quote
    Henrik
    Participant
    Veteran

    Hi Viktormork!

    Do you have any god links for Reading in this subject?

    Regards Henrik

    “Try to read up on how you can make most money based on the amount of money you risk in each trade. Once you know this it’s easier to determinate if a system is good or not.”

Viewing 15 posts - 1 through 15 (of 20 total)
  • You must be logged in to reply to this topic.

Averaging Down Orders improvment


ProOrder: Automated Strategies & Backtesting

New Reply
Author
author-avatar
Ludde500 @ludde500 Participant
Summary

This topic contains 19 replies,
has 6 voices, and was last updated by Nicolas
8 years, 3 months ago.

Topic Details
Forum: ProOrder: Automated Strategies & Backtesting
Language: English
Started: 08/15/2017
Status: Active
Attachments: 4 files
Logo Logo
Loading...