Welcome, Guest. Please login or register.

October 01, 2020, 01:43:41 PM

Login with username, password and session length

Shoutbox

Recent

Members
  • Total Members: 95
  • Latest: prasono
Stats
  • Total Posts: 749
  • Total Topics: 70
  • Online Today: 18
  • Online Ever: 144
  • (August 14, 2019, 12:20:12 PM)
Users Online
Users: 0
Guests: 12
Total: 12

12 Guests, 0 Users

Author Topic: SFE Price Action EA  (Read 11247 times)

Attractor

  • Administrator
  • Sr. Member
  • *****
  • Posts: 300
    • View Profile
    • SmartForexExpert
Re: SFE Price Action EA
« Reply #45 on: April 02, 2020, 03:39:42 PM »
Thank you for implementing the %risk money management.  :)
Is it possible to know the max number of concurrent trades for a specific pair assuming a very strong impulse and all timeframes get active?
Any specific reason why it is not possible to select specific timeframes anymore?

You are welcome :)

The system has in a rude description 4 internal susbsystems, and 4 timeframes. And a sybsustem can open between 1 and 3 trades because have another internal division, but this division don't affect in all the subsysteams and not all the subsystems works in all the timframes.  And only the best symbols have activated all at the same time.
You know I base the risk recommendation in the long time backtest, and usually, when there are a lot of activity is sinonimous of profit, but for reply the question, I calculate that in the big days, the number of trades that can be opened in the 'best' like eurusd, usdch, euraud, is around 10-15 'straight or break' trades, and another 10-15 'reversal' trades.  But this is very very uncommon.
735/5000

Any specific reason why it is not possible to select specific timeframes anymore?

In all these list of different items (subsystems and subsystems of the subsystem and in the different timeframes), and with all the backtest and forward live experience,  the approach is different since the version 3.1. In the beginning, all were activated and in higher timeframes, less symbols were good for use all. Instead of this, the approach now is disconnect the items that don't show a good general performance in general, in any symbol. Then, now, any selected symbol has to work well in the reamining items despite the timeframe, then, don't exist the timeframe selection division.



nok

  • Jr. Member
  • **
  • Posts: 18
    • View Profile
Re: SFE Price Action EA
« Reply #46 on: April 03, 2020, 07:32:07 AM »
Great explanation, thank you.

nok

  • Jr. Member
  • **
  • Posts: 18
    • View Profile
Re: SFE Price Action EA
« Reply #47 on: May 30, 2020, 09:23:03 PM »
Hi Joel,
from the history I can see that this expert changed over time from catching trends towards a focus on strong reversals.
I wonder how the previous versions would have done in the recent years/months.
Would be interesting to see backtests up to now for version 1.4, 1.5, 2.5, 2.6, 3.0 and 3.1.
It would take some time to create the backtests and I know you are always busy with the experts. But maybe it could help to unterstand how the market changed in the recent years.

Attractor

  • Administrator
  • Sr. Member
  • *****
  • Posts: 300
    • View Profile
    • SmartForexExpert
Re: SFE Price Action EA
« Reply #48 on: June 02, 2020, 12:47:07 PM »
Hi Joel,
from the history I can see that this expert changed over time from catching trends towards a focus on strong reversals.
I wonder how the previous versions would have done in the recent years/months.
Would be interesting to see backtests up to now for version 1.4, 1.5, 2.5, 2.6, 3.0 and 3.1.
It would take some time to create the backtests and I know you are always busy with the experts. But maybe it could help to unterstand how the market changed in the recent years.

Hi Nok,

yes, this would be interesting but takes time and probably we see what happens without do this.

First, you can see, the impulse movements are a lot less, the volatility is very low. Seems is beginning to raise again, but has been very low for 2-3 years.

Second, the impulses fails and fails, in the sense that even when the movement develops, almost all the times has taken an agressive pullback that hitted the sl before continue. The problem is that when you receive a pullback of that kind, you don't have any way to detect if it's only a pullback or is a strong reversal, so, move the SL more far is not a solution.

For this reason, the weight of the 'straight' systems has been downgraded with the versions, and I'm sure that this has controlled the losses.
I don't know if that big pullbacks are for a fight of EAs or what, but has did more difficult this kind of systems, for this reason I'm trying to explore more kind of strategies in my new EAs.