Experts: Daily Zone Recovery EA mt5 for GOLD - page 2

New comment
 
Ramil Kireev #:

Hi! Thanks for reaching out.

As I mentioned above, the EA timeframe doesn't matter — it uses M1 and D1 data internally. Just make sure M1 history for GOLD is fully downloaded in your terminal.

To help you, I need more details:

  1. What exactly is "not working properly"? — no trades at all, errors in the log, wrong entries?
  2. Screenshot of the Journal tab from the tester (or the Experts tab if running live)
  3. Screenshot of your tester settings (symbol, date range, modeling mode)
  4. Are you running it in the Strategy Tester or on a live/demo chart?

If you're getting zero trades in the tester — most likely the M1 history is missing. Go to Symbols → select GOLD → Properties → check that M1 bars are available for the test period.


Running live Demo account to test it on real time basis.
On the chart i have just added 200 EMA red line

Do let me know if any other info i can share for analysis

 
allwynnco #:


Running live Demo account to test it on real time basis.
On the chart i have just added 200 EMA red line

Do let me know if any other info i can share for analysis

Thanks a lot for the screenshot and details - now it's much clearer!

On your screenshot I can see the EA is correctly attached to XAUUSD M15 ( DailyZoneRecovery - XAUUS  in Navigator) and Algo Trading is now ENABLED. 

A few important points:

  1. The EA uses M15 only for display, internally it trades on M1 + D1 (yesterday High/Low + ATR). So your chart timeframe is correct.
  2. This EA does NOT trade every hour. It waits for price to reach the Daily Zone. It can be 0-2 trades per day, sometimes no trades for a day - that's normal.

Please do this:

  1. Double-click the EA on chart -> Common tab -> make sure  Allow Algo Trading  is checked.
  2. Keep it running on the Demo and check the  Experts  tab (not Journal) - the EA prints there why it enters or skips a trade.
  3. If you want a quick check, run it in Strategy Tester -> XAUUSD ->  Every tick based on real ticks  -> last 1-2 months. If M1 history is fully downloaded you should see trades.

Leave it running for 24-48 hours on your Demo, and if still no trades, please send me a screenshot of the  Experts  tab and your EA Inputs - I'll check immediately.

Thanks again for testing!

 
Ramil Kireev #:

Thanks a lot for the screenshot and details - now it's much clearer!

On your screenshot I can see the EA is correctly attached to XAUUSD M15 ( DailyZoneRecovery - XAUUS  in Navigator) and Algo Trading is now ENABLED. 

A few important points:

  1. The EA uses M15 only for display, internally it trades on M1 + D1 (yesterday High/Low + ATR). So your chart timeframe is correct.
  2. This EA does NOT trade every hour. It waits for price to reach the Daily Zone. It can be 0-2 trades per day, sometimes no trades for a day - that's normal.

Please do this:

  1. Double-click the EA on chart -> Common tab -> make sure  Allow Algo Trading  is checked.
  2. Keep it running on the Demo and check the  Experts  tab (not Journal) - the EA prints there why it enters or skips a trade.
  3. If you want a quick check, run it in Strategy Tester -> XAUUSD ->  Every tick based on real ticks  -> last 1-2 months. If M1 history is fully downloaded you should see trades.

Leave it running for 24-48 hours on your Demo, and if still no trades, please send me a screenshot of the  Experts  tab and your EA Inputs - I'll check immediately.

Thanks again for testing!

Thanks for the info. Will test it in live Demo account
 

I spent some time going through the source code and there are a few things here that I genuinely find interesting.

The part that stands out most to me is the way the EA builds its logic around the previous day’s high and low instead of stacking a lot of indicators on top of each other.

The breakout/retest logic in Strategy 1 and the move-away-and-return behaviour in Strategy 3 are especially interesting because there is a clear price-action reason behind the initial entry.

I also like the way the recovery side has been approached.


The grid does not appear to use the usual aggressive lot multiplication. It retrieves the lot size of the original position and continues using that same lot size for the recovery entries.

That obviously does not remove the risk of a recovery system, but in my opinion it is a much more sensible starting point than immediately increasing volume after every adverse move.

The basket management around the weighted average entry also caught my attention.


Recalculating the effective average price as recovery positions are added makes much more sense than treating every grid position independently.

I did notice one small error-handling issue in CheckSignalS1() .


If one of the CopyHigh , CopyLow or CopyClose calls fails to return the requested price data, the function can return 0 .

Later in the processing logic, 0 is also used as a valid state where both BUY and SELL conditions are allowed.

It looks more like a small oversight than a problem with the actual strategy, and it should be easy to correct by giving the error/no-data condition its own return value.

I would also consider putting a hard limit on the number of recovery positions.


Even with fixed lots, an unlimited recovery chain can eventually build substantial exposure if the market keeps moving in one direction.

Another thing worth looking at is the fact that the basket TP, SL and new recovery entries are processed on a new M1 bar.

With some of the very small percentage targets used in the defaults, especially Strategy 2, price could touch a basket target and move away again before the next check.

None of these points take away from the concept though.


They are exactly the type of things that can be refined when experimenting with an open-source EA.

Overall I think there is a good idea underneath this code, particularly the previous-day zone logic combined with fixed-lot recovery.


I would be very interested to see the three entry strategies tested individually over a long period to see which one is actually producing the edge.


Thanks for making the source available.


There is enough here to make it worth studying rather than just compiling it and running the default settings.

 
Wilna Barnard #:

I spent some time going through the source code and there are a few things here that I genuinely find interesting.

The part that stands out most to me is the way the EA builds its logic around the previous day’s high and low instead of stacking a lot of indicators on top of each other.

The breakout/retest logic in Strategy 1 and the move-away-and-return behaviour in Strategy 3 are especially interesting because there is a clear price-action reason behind the initial entry.

I also like the way the recovery side has been approached.


The grid does not appear to use the usual aggressive lot multiplication. It retrieves the lot size of the original position and continues using that same lot size for the recovery entries.

That obviously does not remove the risk of a recovery system, but in my opinion it is a much more sensible starting point than immediately increasing volume after every adverse move.

The basket management around the weighted average entry also caught my attention.


Recalculating the effective average price as recovery positions are added makes much more sense than treating every grid position independently.

I did notice one small error-handling issue in CheckSignalS1() .


If one of the CopyHigh , CopyLow or CopyClose calls fails to return the requested price data, the function can return 0 .

Later in the processing logic, 0 is also used as a valid state where both BUY and SELL conditions are allowed.

It looks more like a small oversight than a problem with the actual strategy, and it should be easy to correct by giving the error/no-data condition its own return value.

I would also consider putting a hard limit on the number of recovery positions.


Even with fixed lots, an unlimited recovery chain can eventually build substantial exposure if the market keeps moving in one direction.

Another thing worth looking at is the fact that the basket TP, SL and new recovery entries are processed on a new M1 bar.

With some of the very small percentage targets used in the defaults, especially Strategy 2, price could touch a basket target and move away again before the next check.

None of these points take away from the concept though.


They are exactly the type of things that can be refined when experimenting with an open-source EA.

Overall I think there is a good idea underneath this code, particularly the previous-day zone logic combined with fixed-lot recovery.


I would be very interested to see the three entry strategies tested individually over a long period to see which one is actually producing the edge.


Thanks for making the source available.


There is enough here to make it worth studying rather than just compiling it and running the default settings.

Hi Wilna,

Thank you so much for taking the time to go through the source and for such a detailed and constructive feedback - I really appreciate it!

You described the idea exactly as intended - Daily High/Low zones + fixed-lot recovery with weighted average price. I deliberately avoided lot multiplication to keep the exposure more predictable than classic martingale.

Thank you for the sharp observations:

  1. CheckSignalS1() error handling  - you are absolutely right. Returning  0  on a  CopyHigh/CopyLow/CopyClose  failure coincides with a valid state where both BUY and SELL are allowed. Thank you for spotting this, it is indeed an oversight and should have a separate error/no-data value.

  2. Hard limit for recovery  - also a very valid point. Even with fixed lots an unlimited chain can build substantial exposure in a strong one-directional market, so a hard cap is a sensible risk consideration.

  3. M1 bar processing  - good catch as well. The basket TP/SL and recovery entries are checked on a new M1 bar to reduce noise, but as you noted, with the very small percentage targets of Strategy 2 price could touch the target and move away before the next check. That's a useful nuance to keep in mind when experimenting with the settings.

Thanks again for the thorough review and for highlighting both the strengths and the points worth refining. I am glad you found the concept worth studying.

Best regards,
Ramil

12
New comment