Khalil Abokwaik ATR reversal EA

It is as I explained, because the robot performs the H1 setting and does not pay attention to the tester setting. This is also confirmed by the fact that you have fewer trades, which is normal because the H1 capabilities are always less, because the robot works with an input on a completed candle. The coder is good and therefore provides various options that protect the trader from unnecessary errors. It also provides more opportunities for testing. For example, the function "Exit on opposite Signal" when it is = true, and TProfit = a very large number, then the robot will work on the principle of input signal to signal, capturing the entire price movements. These are already considerations on the operation of the robot, although the ATP indicator is not for such functions, because it is a technical analysis indicator that measures market volatility by calculating the average price range of a given asset for a certain period / in this case based on the number of candles in the relevant time period /. It does not show the direction of the price, but rather the degree / pressure / of its movement.

I tested on demo account with 24 points spread, where backtest results are Superman in demo test its totally broken with large SL hit frequently.
Another fake EA, backtest superman!

That's to be expected due to how the EA works. The EA uses the determined ATR value to calculate signal and trigger levels:

For example, if the ATR is $2.00 and the Signal ATR% is set to 150%, then it starts looking for signals whenever the current bar exceeds a range of $3.00 ($2.00 x 1.5). Now, ATR(N) is "Average True Range" of a single bar over N bars. The ATR(21) on M1 is the average size of an M1 bar. The ATR(21) on H1 is the average size of an H1 bar (a full hour). Of course the average size of a full hour bar is much larger then that of an M1 bar. It will be 10-20 times more. Now, that value is the base for signal calculation. With ATR(21) on H1 you will get almost no signals because an M1 bar will very very rarely excceed the average H1 bar by 1.5 times. Maybe that happens at NFP, once a month. I hope this extreme example (M1 vs H1) makes it clear, why you get less trades if you switch the ATR timeframe to a different then the current one.

That option to change the ATR timeframe is kind of wrong. It should not exist at all in the EA inputs.

Another interesting point is the ATR period. Let's say ATR(21) vs ATR(100). Do you think it makes a difference?

No. Why is that? Because the average size of an M1 or M5 or H1 bar will not change siginificantly if the period value gets larger. From a certain period number on the average size will roughly stay the same. ATR(21) and ATR(100) values will differ only slightly. With few periods, e.g. ATR(3) and ATR(21), they can differ significantly. But 21 is already too long and any volatility spikes are already leveled out, compared to ATR(100), ATR(200) or ATR(500). There will be no visible difference.

$0.24 spread is way too much. Ticksize of a standard contract is $0.10. Some brokers with their own price aggregation go even below that value. That's possible because at many times liquidity is so good, that you can trade without a spread. So, even if the broker offers $0.05 he can still make money.

You should change the broker.

i apologise. i do not want to spam this thread. i asked my AI to optimize the testing part of Heiken Ashi Multi-Timeframe EA (fixed).mq4 so it is is a lot quicker now.
it finished in seconds.

The test was done from november 2025 - today on M5. Spread was 6-8 points + 0.04$tax per trade. not sure if tester applies tax when doing simulations


AI wrote this part in a non tech way.
What I fixed in the EA:

1. Made it run way faster in the Strategy Tester
   1. Before: The bot was checking the H4, H1, M30, and M15 charts on every single tick (multiple times a second). It’s like asking four people "Is the trend still up?" every split second. This made backtesting super slow.
   2. Now: It only checks the higher charts once every 5 minutes (when a new M5 candle starts). This makes backtesting run in seconds rather than minutes, while the trade trailing still runs in real-time.

   
   
   

2. Stopped the log files from spamming "OrderModify error 1"
   1. Before: Because of tiny computer rounding math errors, the bot was constantly telling MT4 to change the stop loss to the exact same price it already had. MT4 got annoyed and spammed the logs with Error 1 ("No changes made").
   2. Now: We tweaked the math so the bot only modifies the stop loss when the price has actually moved a noticeable amount.

   
   
   

3. Cleaned up compiler warnings
   1. Added code safety checks to make sure the program compiles cleanly in MetaEditor without any warnings.

   
   
   

woow thats good. But, as far I have seen on Tickmill is offer lowest like 24 points fixed, while XM offer 30-32 and 5ers Prop Firm offer 55-60 points!!!
Anyway, just tell me which broker offer 10-5 points spread, I have no idea. I will try.

Just found Tickmill Raw offer 8 poins fixed spread, will try it now.

IC Markets Raw spread account, or for EU residents IC Trading.

Never do that. Those are not prop firms. Those are bucket shops in disguise.