War market regime

New comment
 

My EA on backtest is running very well, until it comes to the US-IRAN war April 2026 it keeps hovering around the same performance, yes, it is not losing, but also not showing the same performance as before the war

Have anyone face the same issue?

Files:
mt5.png  17 kb
mt5-2.png  17 kb
 
sam-86:

My EA on backtest is running very well, until it comes to the US-IRAN war April 2026 it keeps hovering around the same performance, yes, it is not losing, but also not showing the same performance as before the war

Have anyone face the same issue?

Try to use "Every tick based on real tick" modelling and you are likely to make a discovery.
 
sam-86:

Have anyone face the same issue?

Forum on trading, automated trading systems and testing trading strategies

Machine learning in trading: theory, models, practice and algo-trading

fxsaber, 2025.12.08 08:09

Throwing this clock out of the trade?


This failure is observed on almost all symbols. It was a tweak to Trump. From my point of view, this piece should be thrown out completely when training. Right now it is in OOS, but next year it will definitely fall into the learning interval, and then you have to throw it out. But even when assessing robustness at OOS, it is worth ignoring the performance at this point (the Spring 2025 slice).


I assume that there have been such intervals before.

 
sam-86:

My EA on backtest is running very well, until it comes to the US-IRAN war April 2026 it keeps hovering around the same performance, yes, it is not losing, but also not showing the same performance as before the war

Have anyone face the same issue?

Your EA probably works well on the slower movements and is not optimized for larger and more sudden movements in the market.
 
Oleksandr Medviediev #:
Try to use "Every tick based on real tick" modelling and you are likely to make a discovery.
the backtest already with "Every tick based on real tick"
 
fxsaber #:
@fxsaber I agree that these intervals usually happen, but this take longer period, check the photo (10 years backtest and this period is not short on the curve)
Files:
mt5-2.png  17 kb
 
classicaltrading #:
Your EA probably works well on the slower movements and is not optimized for larger and more sudden movements in the market.
the EA already designed to benefit from those moves and working well along the full backtest period except this time from end of March 2026 until now
 
sam-86 #:
@fxsaber I agree that these intervals usually happen, but this take longer period, check the photo (10 years backtest and this period is not short on the curve)
The answer is in the topic title. Global crisis.
 
sam-86:

My EA on backtest is running very well, until it comes to the US-IRAN war April 2026 it keeps hovering around the same performance, yes, it is not losing, but also not showing the same performance as before the war

Have anyone face the same issue?

For you, the real ticks start from the point where the EA’s profitability begins to stagnate. Make sure you have 100% historical data quality, not 99%.

 
Ghenadie Tumco #:

For you, the real ticks start from the point where the EA’s profitability begins to stagnate. Make sure you have 100% historical data quality, not 99%.

 the real ticks start from the point where the EA’s profitability begins to stagnate, so you think MT5 ticks all trash? only starting from April 2026 are real ticks?
 
sam-86 #:

 the real ticks start from the point where the EA’s profitability begins to stagnate, so you think MT5 ticks all trash? only starting from April 2026 are real ticks?
Not all brokers offer a long history based on real tick data. Check the quality of the history used in your backtest. Is it 100%?
New comment