Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Code for "if ADX (-1) < ADX (-2)" 18 replies

Looking for Super ADX/Power ADX 1 reply

Fast Averaging versus Slow Averaging 1 reply

Averaging 7 replies

Martigale and Averaging Down 4 replies

Attachments: Impulse ADX with averaging
Exit Attachments

Impulse ADX with averaging

Here what is possible with this EA when using compounding option scale_lots set to PERCENTS - EURUSD in 1 month from 100 usd to 17000 usd:

Attached Image (click to enlarge)
Click to Enlarge Name: eurusd_2020_high_return.png Size: 27 KB


Here backtest:

Attached File
File Type: xlsx eurusd_2020_high_return.xlsx   77 KB | 64 downloads


Here .set file:

Attached File
File Type: txt eurusd_hr.txt   3 KB | 42 downloads


Idea is to find out best parameters and manually re-test in Strategy Tester from 2017 each day how this would work... MM plan here cash out on 15000 usd. So with 100 usd we would have 150 attempts. Question: how many days % is ok get such impressive returns.
Price always goes somewhere - sooner or later...
Stocks - manual trading Return This Year: 24.8%
Thinking about this:

For standard averaging strong TREND is killer. If the price moves too fast in 1 direction standard averaging will fail.

Idea: EA starts to trade at the London opening. When GBPUSD price runs 10 pips up (or down) - open deal the same direction with TP target 20 pips and SL 10 pips (maybe other TP/SL but risk-reward still 1:2). When SL hits open deal in opposite direction TP 20 and SL 10. If again SL hit then open again new deal in the new direction.

For example:
1. deal BUY - hit SL (volume 0.01) Balance: - 1 if tp balance: 2
2. deal SELL - hit SL (volume 0.01) Balance: -2 if tp balance: 1
3. deal BUY - hit SL (volume 0.02) Balance: -4 if tp balance: 2
4. deal SELL - hit SL (volume 0.03) Balance: -7 if tp balance: 2
5. deal BUY - hit SL (volume 0.05) Balance: -12 if tp balance: 3
6. deal SELL - hit SL (volume 0.07) Balance: -19 if tp balance: 2
7. deal BUY - hit SL (volume 0.11) Balance: -30 if tp balance: 3

For this approach (martingale I guess) TREND is a friend and the RANGED market is a killer.

Here need to analyze how often and how long the price gets trapped in 20 pips range...?
Price always goes somewhere - sooner or later...
Stocks - manual trading Return This Year: 24.8%
Quoting gepards
Thinking about this: For standard averaging strong TREND is killer. If the price moves too fast in 1 direction standard averaging will fail. Idea: EA starts to trade at the London opening. When GBPUSD price runs 10 pips up (or down) - open deal the same direction with TP target 20 pips and SL 10 pips (maybe other TP/SL but risk-reward still 1:2). When SL hits open deal in opposite direction TP 20 and SL 10. If again SL hit then open again new deal in the new direction. For example: 1. deal BUY - hit SL (volume 0.01) Balance: - 1 if tp balance: 2...
this idea is working... here backtest results for gbpusd 2017:

Attached File
File Type: xlsx gbpusd_martingale_level2.xlsx   239 KB | 67 downloads


Entry very strange but still working - open buy at 7 AM... thats it.. TP 95 and SL 45, raise with 2. Max levels 2.
Price always goes somewhere - sooner or later...
Stocks - manual trading Return This Year: 24.8%
Quoting gepards
{quote} this idea is working... here backtest results for gbpusd 2017: {file} Entry very strange but still working - open buy at 7 AM... thats it.. TP 95 and SL 45, raise with 2. Max levels 2.
The "strangeness" of the entry brings up an interesting question: Would it work even better with SIMULTANEOUS buy AND sell orders issued at 7 AM....?
Quoting RoboSeer
{quote} The "strangeness" of the entry brings up an interesting question: Would it work even better with SIMULTANEOUS buy AND sell orders issued at 7 AM....?
Awesome question!! Tnx!

That's the desired collaboration I wrote in post nr1. So if you RoboSeer ever decide to use any EAs from this thread (seems we could have several versions of EAs..) on real account - you can get a free key, just write to me a private message with your real account number.

Everybody is welcome to collaborate. Let's make huge collective brains to discover the right approach!

I will implement the idea of 2 parallel BUY SELL orders. Let's see...
Price always goes somewhere - sooner or later...
Stocks - manual trading Return This Year: 24.8%
Quoting RoboSeer
{quote} The "strangeness" of the entry brings up an interesting question: Would it work even better with SIMULTANEOUS buy AND sell orders issued at 7 AM....?
Made something.. but seems another approach not exactly what you said.. will continue.. but ok here something:

Open 2 orders SELL and BUY. When one hits SL then close and continue other. When both hit SL then open the next level.

Attached Image (click to enlarge)
Click to Enlarge Name: 2orders.png Size: 33 KB


Here backtest for 2017 - 2020:

Attached File
File Type: xlsx 2orders_gbpusd_2017_2020.xlsx   163 KB | 47 downloads


Here SL 70 TP 160 and all starts at 5 AM. Max levels 8. Raise 2.
Price always goes somewhere - sooner or later...
Stocks - manual trading Return This Year: 24.8%
Quoting gepards
{quote} Awesome question!! Tnx! That's the desired collaboration I wrote in post nr1. So if you RoboSeer ever decide to use any EAs from this thread (seems we could have several versions of EAs..) on real account - you can get a free key, just write to me a private message with your real account number. Everybody is welcome to collaborate. Let's make huge collective brains to discover the right approach! I will implement the idea of 2 parallel BUY SELL orders. Let's see...
Thank you for your kind offer, I do appreciate it. I really like your idea of using max. 2 levels as it significantly limits the maximum loss on days when the market does not trend.

I suggest further testing this concept on more volatile Forex pairs and other instruments (incl. gold, oil, index futures).
very good results
  • Post #29
  • Quote
  • Edited at 4:15pm May 30, 2020 4:02pm | Edited at 4:15pm
Here one more result this time with EURUSD on timeframe M5, levels no more than 3 (good - stable martingale):

Attached Image (click to enlarge)
Click to Enlarge Name: 2orders_eurusd.png Size: 43 KB


Here 2017 - 2020 report:

Attached File
File Type: xlsx 2orders_eurusd_2017_2020_levels_3.xlsx   144 KB | 49 downloads


Set file - and added this version of EA in post Nr. 1:

Attached File
File Type: txt 2017_eurusd_tp110_sl55_2orders.set.txt   3 KB | 61 downloads


Please optimize/test with your broker data and other instruments - indexies, comodities, currencies pairs etc

While optimize and run backtests keep in mind this all works with open prices. And it can run so on live too. If use timeframe M5 (this case with EURUSD) it means EA will check SL and TP every 5 minutes on new candle opening. Yes... it might go quite far in any direction within 5 minutes. But seems for this EA it is not always bad... just statistic...

Anyway will implement some protection with SL - to add some hard StopLoss with X pips distance just in case (to avoid flash crashes for example).
Price always goes somewhere - sooner or later...
Stocks - manual trading Return This Year: 24.8%
do you have it for MT4?
Quoting Beren
do you have it for MT4?
nop. just Mt5 for now. MT5 far better for optimizations/backtests/forward tests.

If we get something stable on MT5 maybe we can consider port to MT4 version too. But for now, EA is not stable.
Price always goes somewhere - sooner or later...
Stocks - manual trading Return This Year: 24.8%
  • Post #32
  • Quote
  • Last Post: Sep 27, 2020 3:33pm Sep 27, 2020 3:33pm
Just updating status... have tried various approaches and seems found/made something stable with averaging.

Here backtest results:

Attached Image (click to enlarge)
Click to Enlarge Name: walker_profit.png Size: 23 KB


Attached Image (click to enlarge)
Click to Enlarge Name: walker_stats.png Size: 39 KB

It survived February/March 2020. And it survived 2010 - till now (more than 10 years). Didn't implemented in EA lots rising on account growth. Will do that manually or later. Will run this on live account.

I call this EA "WalkerPips". Algorithm is my "know how"... didnt see that approach anywhere but maybe nothing new. Who knows... just found good setups in optimization. It is not based on ADX anymore so not belonging to this thread. Maybe will make new thread (ar journal) if will see that EA works fine on live account.

Max levels it opened is 9. Backtest 1764 trades in 10 years indicates some stability.

Made EA so that if it is forced to go in deep averaging levels (and big stress for me) then increases reward too. Therefore on market crash it spiked 90% drawdowns but then went out with significant profits.

For now thats all folks - seems this thread is dead and let it be so.
Price always goes somewhere - sooner or later...
Stocks - manual trading Return This Year: 24.8%
0 traders viewing now