Khalil Abokwaik ATR reversal EA | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Related Instruments

Gold/USD

Similar Threads

Abokwaik replies to PMs 96 replies

Atr 5 replies

Correct number of weeks for ATR and hour ATR 1 reply

ATR Trailing Stop (With Modifications) 3 replies

Does ATR mean anything in Forex? 7 replies

Attachments: Khalil Abokwaik ATR reversal EA
Exit Attachments

Khalil Abokwaik ATR reversal EA

  • #41
  • Jul 11, 2026 2:48am Jul 11, 2026 2:48am
I have tried the ATR Reversal EA on a demo account but it has not made any trades (GBP/CHF). Auto trading green and a smiley face on H1 time frame. Any ideas about why no trades.
  • #42
  • Jul 11, 2026 3:37am Jul 11, 2026 3:37am
Quoting Minky01
I have tried the ATR Reversal EA on a demo account but it has not made any trades (GBP/CHF). Auto trading green and a smiley face on H1 time frame. Any ideas about why no trades.
Try it on XAUUSD,M1 with the settings shown in post #1, to confirm that it works for you.
It takes blood, sweat and enough tears to wash away the blood.
  • #43
  • Jul 11, 2026 5:43am Jul 11, 2026 5:43am
07.10. first day of pre-testing of ATR Reversal EA on M5. It ended the day nicely.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: ATR EA beállítás 5M.jpg Size: 27 KB
Click to Enlarge Name: ATR EA előre 07.10.jpg Size: 42 KB
Click to Enlarge Name: ATR ker. 07-10.jpg Size: 60 KB
  • #44
  • Jul 11, 2026 6:24am Jul 11, 2026 6:24am
Quoting pusztafie
07.10. first day of pre-testing of ATR Reversal EA on M5. It ended the day nicely. {image} {image} {image}
Thanks for sharing.

What was the acount balance 15000,
You are risking 0.1% = 15000 x 0.1% = 15 per trade as money management is ON

Seems like upi have changed a lot of settings in the parameters as compared to the original parameters of the EA as per Emmzett's suggestion.

Is this a backtest
OR

Live test on the demo account

OR

Real account ( I cant be a real account as it works only on Demo)

Which broker is it and how much spread?

If its demo please can you share the complelete Back test report in the HTML?
  • #45
  • Edited 7:19am Jul 11, 2026 6:45am | Edited 7:19am
Quoting mazibee
{quote} Thanks for sharing. What was the acount balance 15000, You are risking 0.1% = 15000 x 0.1% = 15 per trade as money management is ON Seems like upi have changed a lot of settings in the parameters as compared to the original parameters of the EA as per Emmzett's suggestion. Is this a backtest OR Live test on the demo account OR Real account ( I cant be a real account as it works only on Demo) Which broker is it and how much spread? If its demo please can you share the complelete Back test report in the HTML?
You can find the back test in post 29, you will also get an answer about the broker. He started trading with 0.07 lots with 0.1% risk on the 15000 demo account. The first day of the 3 month pretest.
The spread is 20-24 points under normal circumstances.
  • #46
  • Jul 11, 2026 4:02pm Jul 11, 2026 4:02pm
Quoting pusztafie
07.10. first day of pre-testing of ATR Reversal EA on M5. It ended the day nicely. {image} {image} {image}
Hi Everyone...

I did a backtest from July 6th to July 10th with pusztafies settings on the 5M chart, only I changed the ATR TF to 1H, I can't do more tests right now but wanted to post this.

Attached Image (click to enlarge)
Click to Enlarge Name: 1hr tf.jpg Size: 65 KB
  • #47
  • Jul 12, 2026 1:00am Jul 12, 2026 1:00am
Quoting cfudge
{quote} Hi Everyone... I did a backtest from July 6th to July 10th with pusztafies settings on the 5M chart, only I changed the ATR TF to 1H, I can't do more tests right now but wanted to post this. {image}
hello, very good test results, but I see that the bot is for a demo version. or is it not so. If it is limited in time and only for a demo and does not work in real time, where can it be bought. If not, if anyone knows the logic of actions, please explain it and I will try to do it for mt4. Thanks and have a nice day
  • #48
  • Jul 12, 2026 1:06am Jul 12, 2026 1:06am
Quoting emmzett
{quote} Try it on XAUUSD,M1 with the settings shown in post #1, to confirm that it works for you.
Thanks emmzet
  • #49
  • Edited 3:21am Jul 12, 2026 2:47am | Edited 3:21am
Quoting arteks
{quote} hello, very good test results, but I see that the bot is for a demo version. or is it not so. If it is limited in time and only for a demo and does not work in real time, where can it be bought. If not, if anyone knows the logic of actions, please explain it and I will try to do it for mt4. Thanks and have a nice day
Do NOT use it on a live account. In fact those are not good test results. Look at the details. cfudge's test has a win rate of 100%. It looks good because he didn't have a single loser. Of course that's fully unrealistic for live trading. You simply cannot reproduce such theoretical results in practice.

The key to long-term profitability is not "maximizing the win rate". Long-term profitability requires an edge. An edge is defined as "higher probability that Scenario A will occur than that Scenario B will occur". The outcome of the next trade cannot be predicted, since individual trade results are random and do not correlate with previous trade results. Therefore, to make money, you need an edge and a sufficient number of trades so that the overall result of all trades can become positive.

In cfudge's test the losing trades are completely ignored (there are none). Such a test is not representative and cannot work.

The problem with this EA is that the profit potential of a single trade is too small (reduced by the Trailing-Stop). In other words, the system attempts to make money without losers (Scenario B). Based on the idea that if we have no losers, we will still make money even with minimal targets. With minimal targets we don't need any profit potential - or in other words - we don't need an edge.

Of course, this kind of logic doesn't work in real life.

Increase input "Trailing ATR %" to at least half of "TP ATR %", and tests will become more realistic and start producing more losers. You need to get away from profit targets below 50-60 cents. The average target must become at least 4 times the average spread. Then you can see from the results whether the entry condition (the ATR reversal logic) has any edge. It hasn't.

Quoting arteks
...if anyone knows the logic of actions, please explain it...
Read the thread. All important parts have been explained before.
It takes blood, sweat and enough tears to wash away the blood.
  • #51
  • Jul 12, 2026 5:22am Jul 12, 2026 5:22am
Quoting emmzett
The key to long-term profitability is not "maximizing the win rate". Long-term profitability requires an edge. An edge is defined as "higher probability that Scenario A will occur than that Scenario B will occur". The outcome of the next trade cannot be predicted, since individual trade results are random and do not correlate with previous trade results. Therefore, to make money, you need an edge and a sufficient number of trades so that the overall result of all trades can become positive.
Yes, that’s exactly the point.
Traders think they have a strategy, but they don’t. Given that the outcome of the next trade cannot be predicted with certainty, the only real advantage is a statistical advantage. And this is where the problem begins—one that few people consider: in order to develop a real strategy that produces a statistical edge, you need sufficiently long-term and reliable data.
In light of this, it may now become clear why you can only download short-term data (i.e., a small number of candles for a given time frame) from your broker. Quite simply, they do not provide you with sufficient and reliable data to begin working with. Obtaining such data is cumbersome, and most people don’t bother with it. Without it, however, trading is nothing more than gambling.
  • #52
  • Jul 12, 10:53pm (34 hr ago) Jul 12, 10:53pm (34 hr ago)
[quote=WM4X;15800845]{quote} Yes, that’s exactly the point. Traders think they have a strategy, but they don’t. Given that the outcome of the next trade cannot be predicted with certainty, the only real advantage is a statistical advantage. And this is where the problem begins—one that few people consider: in order to develop a real strategy that produces a statistical edge, you need sufficiently long-term and reliable data.

Precisely and clearly written, and the confirmation today from 03 to 05 am is a positive test result +112 and negative in real time - 72. Although the time is short, it was enough for me to see differences in the results, and different ones. The EA does not take into account the costs of commission and spread, which further reduces the specified profit. Too bad for the good idea and execution of the coder. So be careful / as emmzett said / with this bot and do not get carried away and live with the hope of making money quickly. Still, I have one hope that I did not target the exact settings and timeframe. Good luck
  • #53
  • Jul 13, 12:49am (32 hr ago) Jul 13, 12:49am (32 hr ago)
the code listed in the first post of this thread has a variable "atr period".. this tells me that this is moving average based...

if you change this setting, you will get entirely different results in your backtest...

now here is the catch... nobody knows what that setting should be when trading forward in real time.

moving averages or anything that rhymes with moving averages will lose money.

best wishes.
  • #54
  • Jul 13, 3:47am (29 hr ago) Jul 13, 3:47am (29 hr ago)
Quoting JaniceKelly
the code listed in the first post of this thread has a variable "atr period".. this tells me that this is moving average based...
You'd better drive the car and let someone else steer.
It takes blood, sweat and enough tears to wash away the blood.
  • #55
  • Edited 7:14pm Jul 13, 9:02am (24 hr ago) | Edited 7:14pm
Here we go again...lol

Look Ma...no losses........................

I am still intrigued with this and to that end I have started again on the 1M gold with the settings posted here. I wanted to be on the 5M chart but ended up on 1M by mistake...I will rectify that today at close. Note the ATR is from the 1H

Could all still go to shit...

Attached Image (click to enlarge)
Click to Enlarge Name: newatr.jpg Size: 42 KB

Attached Image (click to enlarge)
Click to Enlarge Name: newatrset.jpg Size: 25 KB
  • #56
  • Jul 13, 9:52am (23 hr ago) Jul 13, 9:52am (23 hr ago)
Recent testing has shown surprisingly high profits. I'm now running it on a live demo account. I'm aware that EAs can sometimes "look ahead" during backtesting—a common coding trap I've encountered before. I’m eager to verify if these results are genuine or just a backtesting artifact.
  • #57
  • Jul 13, 12:29pm (21 hr ago) Jul 13, 12:29pm (21 hr ago)
Quoting cfudge
Here we go again...lol Look Ma...no losses........................ I am still intrigued with this and to that end I have started again on the 1M gold with the settings posted here. I wanted to be on the 5M chart but ended up on 1M by mistake...I will rectify that today at close. Could all still go to shit... {image} {image}
in the settings it is H1 and not M1, which is correct?
because, on both frames, the results today are negative
  • #58
  • Jul 13, 7:12pm (14 hr ago) Jul 13, 7:12pm (14 hr ago)
Quoting arteks
{quote} in the settings it is H1 and not M1, which is correct? because, on both frames, the results today are negative
I set it up on the 1M chart (supposed to be 5M) but the ATR TF is 1H

I am assuming it will take trades on the lower TF but take the ATR readings from the 1H

Can I assume I assume correctly?????
  • #59
  • Jul 13, 7:25pm (14 hr ago) Jul 13, 7:25pm (14 hr ago)
Quoting cfudge
{quote} I set it up on the 1M chart (supposed to be 5M) but the ATR TF is 1H I am assuming it will take trades on the lower TF but take the ATR readings from the 1H Can I assume I assume correctly?????

no. I have two options for applying the time graph. 1. On a test - when it is set to current in the settings, you adjust the time graph from the tester and 2. in real time, the setting will tell it where to get information from regardless of what time graph it is installed on and vice versa, if it is current, it will take information from the time graph it is installed on. At least that's how I would do it, to eliminate a possible error in case of unintentional change of the time range of the graph on which the robot is installed
  • #60
  • Jul 13, 8:22pm (13 hr ago) Jul 13, 8:22pm (13 hr ago)
Quoting arteks
{quote} no. I have two options for applying the time graph. 1. On a test - when it is set to current in the settings, you adjust the time graph from the tester and 2. in real time, the setting will tell it where to get information from regardless of what time graph it is installed on and vice versa, if it is current, it will take information from the time graph it is installed on. At least that's how I would do it, to eliminate a possible error in case of unintentional change of the time range of the graph on which the robot is installed
On the tester, when I was using the 5M chart I got "X" amount of trades and when I changed the ATR TF to 1 hour and stayed on the 5M chart I got a lot less trades.
5 traders viewing now