Never add to a losing position
I agree, but something I do is if I have a loser and I judge that price is going back up (vice versa for shorts) then I would buy same lot size so I get back to breakeven or very small loss in half the price distance it took to lose (and hopefully before I go to bed! 🙂 )
Works surprisingly often!
Yes, I confess I sometimes do that too, in manual trading. Hard to resist the temptation of maybe getting out of a stinker at breakeven. I think the question you have to ask is, ‘if I didn’t have an open position, would I buy here?’ If the answer is a confident Yes, then averaging down can be a good tactic. At least manually you have full control over the extra risk. But it would take some v clever coding to get it right in an algo.
ModaParticipant
Senior
Here a Dax version with BreakEven fonction, no modifications with the main MTF values
It’s far from the original version in terms of performances i just adapted it to my needs…
Thanks @nonetheless
Here’s my adaptation to @Moda version above – with added @JuanJ ML (to adjust sl / pt) and @Paul ML (to adjust StartBreakevenLong/Short) including @Vonasi MM. Improved win rate and less losing trades.
backtest results for the above.
Hi eckaw,
Im new and trying to learn to program in prc. What do I need to remove to get it working because I dont know what to remove from backtest code to make it work.
Here’s my adaptation
I can’t wait until tomorrow to try these last 2 versions … even if they are only half as good as the backtest … they should be winners!!?? 🙂
Thank You @Moda and @eckaw for being magnanimous and sharing with us all!
Thanks also to JuanJ, Paul and Vonasi and Nonetheless for the original MoD strategy, and Jan and Laurenzo for the original 69 MA definitions.
The ML version is very interesting, also reduces all the noise of the little profit trades. What if applied on the original trailing stop version and not on the less performing breakeven one?
How come i get these results on MoD-Dax-4.x ML
look at the DD from dec 27 2019, its huuuge… just me that thinks this is not ready to live?
just me that thinks this is not ready to live?
We all check out Systems on Demo Live / Forward Test for a certain number of trades / ups and downs period in price levels etc.
Unless there are other folk like me, who like the excitement of putting a System Real Live following good backtests / walk forward etc?
BUT I would always then watch each and every trade like a hawk and would only do above if I was sitting at my desk doing other tasks on one monitor whilst the trade develops on No 2 monitor in my peripheral vision.
I might then exit a trade / close the System down within a few seconds or when it reaches a logical exit point which may not be catered for by the coded strategy.
I don’t see above as any more risky than manual trading? But if you are not comfortable manual trading then don’t do above! 🙂
EDIT / PS
There’s an even bigger drawdown on 13 March … see attached.
my point was that it wasnt even a point in putting it on demo whit this huge drawdowns with no protection at all..
the point of automatic systems is to not have to look at the monitor all the time.
if u want to trade manual then just have the system as an alert?
with no protection at all..
Are you able to do anything to improve the System to get it more to what you would like?
Hi @eckaw, I think there’s an error at line 421 – should be HeuristicsAlgo4, no?
Also, testing should be really done with MM switched off, so line 43 needs to be MM2ndType = 0
Nice work with the ML application, are you sure there’s no crosstalk between them?
backtest results for the above.
I do not get the same results as you with the original ML code, did you optimize more? Spread?
Also, @vonasi produced a more sophisticated version of that MM code, if you’re interested – includes a fuse to quit the algo if x% DD:
Capital = 10000
MinSize = 1 //The minimum position size allowed for the instrument.
MM1stType = 0 //Starting type of moneymanagement. Set to 0 for level stakes. Set to 1 for increasing stake size as profits increase and decreasing stake size as profits decrease. Set to 2 for increasing stake size as profits increase with stake size never being decreased.
MM2ndType = 1 //Type of money management to switch to after TradesQtyForSwitch number of trades and ProfitNeededForSwitch profit has occurred
TradesQtyForSwitch = 50 //Quantity of trades required before switching to second money management choice.
ProfitNeededForSwitch = 3 //% profit needed before allowing a money management type change to MM2ndType.
DrawdownNeededToSwitch = 3 //% draw down from max equity needed before money management type is changed back to MM1stType.
DrawdownNeededToQuit = 35 //% draw down from max equity needed to stop strategy
Once MoneyManagement = MM1stType
Equity = Capital + StrategyProfit
maxequity = max(equity,maxequity)
if equity < maxequity * (1 - (DrawdownNeededToSwitch/100)) then
enoughtrades = 0
tradecount = 0
moneymanagement = MM1stType
endif
if equity < maxequity * (1 - (DrawdownNeededToQuit/100)) then
quit
endif
if not EnoughTrades then
if abs(countofposition) > abs(countofposition[1]) then
tradecount = tradecount + 1
endif
if tradecount > TradesQtyForSwitch and maxequity >= Capital * (1 + (ProfitNeededForSwitch/100)) then
EnoughTrades = 1
MoneyManagement = MM2ndType
endif
endif
IF MoneyManagement = 1 THEN
PositionSize = Max(MinSize, Equity * (MinSize/Capital))
ENDIF
IF MoneyManagement = 2 THEN
PositionSize = Max(LastSize, Equity * (MinSize/Capital))
LastSize = PositionSize
ENDIF
IF MoneyManagement <> 1 and MoneyManagement <> 2 THEN
PositionSize = MinSize
ENDIF
PositionSize = Round(PositionSize*100)
PositionSize = PositionSize/100
// Size of POSITIONS
PositionSizeLong = 1 * positionsize
PositionSizeShort = 1 * positionsize
