“Why are the EA backtest results different from the real (live) market results?”

New comment
 
“I am using the same EA, the same settings, and the same period in both the backtest and the live account. Why can the results be different, even though I am already using real ticks? On the live account my EA is profitable, but in the backtest it ends up with a margin call, even though the period is the same.”
 
Muhammad Nurrochim:
“I am using the same EA, the same settings, and the same period in both the backtest and the live account. Why can the results be different, even though I am already using real ticks? On the live account my EA is profitable, but in the backtest it ends up with a margin call, even though the period is the same.”
Different execution models. Same conclusion: improve the strategy.
 
Muhammad Nurrochim:
“I am using the same EA, the same settings, and the same period in both the backtest and the live account. Why can the results be different, even though I am already using real ticks? On the live account my EA is profitable, but in the backtest it ends up with a margin call, even though the period is the same.”
many many possible reasons, however, most agreed upon reason is the coding of the ea. This topic has been discussed many many times thru the years. Next time do a search before posting.
 

It depends on the strategy.

Backtest is like demo trading, execution is ideal.
Live trading has real delay and slippage, and that depends on the broker liquidity provider.

That s why people can get very different results across brokers.

 
Muhammad Nurrochim:
“I am using the same EA, the same settings, and the same period in both the backtest and the live account. Why can the results be different, even though I am already using real ticks? On the live account my EA is profitable, but in the backtest it ends up with a margin call, even though the period is the same.”

This difference between backtest results and live trading results is normal and can be caused by several factors, assuming the EA is properly programmed, developed in a serious and professional manner, and without any intention to manipulate results or generate misleading data for commercial purposes, even when real ticks are used.

Slippage, spread variation, execution delays, margin calculation, liquidity, and the way the strategy tester models tick sequences can all lead to different outcomes. In live trading, orders are executed under real market conditions, while the strategy tester still operates in a simulated environment.

I have written a detailed explanation on this topic based on my experience, which may help clarify these differences:   https://www.mql5.com/es/blogs/post/765944

The False Religion of "Real Ticks": The Silent Theft of the Backtest
The False Religion of "Real Ticks": The Silent Theft of the Backtest
  • www.mql5.com
The False Religion of "Real Ticks": The Silent Theft of the Backtest For years, across forums, marketplaces, and private groups, one idea is repeated as if it were an absolute truth: "If you use Every
 
Andres Felipe Carvajal Rodriguez #:

This difference between backtest results and live trading results is normal and can be caused by several factors, assuming the EA is properly programmed, developed in a serious and professional manner, and without any intention to manipulate results or generate misleading data for commercial purposes, even when real ticks are used.

Slippage, spread variation, execution delays, margin calculation, liquidity, and the way the strategy tester models tick sequences can all lead to different outcomes. In live trading, orders are executed under real market conditions, while the strategy tester still operates in a simulated environment.

I have written a detailed explanation on this topic based on my experience, which may help clarify these differences:   https://www.mql5.com/es/blogs/post/765944

Great article ! Thanks
 
Andres Felipe Carvajal Rodriguez #:

This difference between backtest results and live trading results is normal and can be caused by several factors, assuming the EA is properly programmed, developed in a serious and professional manner, and without any intention to manipulate results or generate misleading data for commercial purposes, even when real ticks are used.

Slippage, spread variation, execution delays, margin calculation, liquidity, and the way the strategy tester models tick sequences can all lead to different outcomes. In live trading, orders are executed under real market conditions, while the strategy tester still operates in a simulated environment.

I have written a detailed explanation on this topic based on my experience, which may help clarify these differences:   https://www.mql5.com/es/blogs/post/765944

“The article you wrote is truly exceptional—very enlightening. Thank you very much.”...
 
Muhammad Nurrochim:
“I am using the same EA, the same settings, and the same period in both the backtest and the live account. Why can the results be different, even though I am already using real ticks? On the live account my EA is profitable, but in the backtest it ends up with a margin call, even though the period is the same.”
It can be profitable on the live account, but for how long? You need to make it as safe as possible and protected from drawdown...some brokers close losing positions automatically if the equity drawdown reaches 50%
 
The only thing a back test is viable in is the drawdown , has it steady DD is it crazy swings etc . Does it look like it could survive . Thinking taking an EA especially all these martingale time bombs  and running a back test for a year  24/5 data and thinking wow this looks good , then you are delusional . An EA does basically one thing , works in Trends , works , in choppy markets etc . Anybody thinking a trend EA in a backtest  profit higher and higher  is a clown.  Factor in Broker spreads and all that entails . Live testing is the only way . 
 
Victor Paul Hamilton #:
The only thing a back test is viable in is the drawdown , has it steady DD is it crazy swings etc . Does it look like it could survive . Thinking taking an EA especially all these martingale time bombs  and running a back test for a year  24/5 data and thinking wow this looks good , then you are delusional . An EA does basically one thing , works in Trends , works , in choppy markets etc . Anybody thinking a trend EA in a backtest  profit higher and higher  is a clown.  Factor in Broker spreads and all that entails . Live testing is the only way . 
“Yes, that’s correct—live testing is indeed necessary. I tested my EA in backtesting and it got margin called, yet on a real account it’s profitable.”
 
Conor Mcnamara #:
It can be profitable on the live account, but for how long? You need to make it as safe as possible and protected from drawdown...some brokers close losing positions automatically if the equity drawdown reaches 50%
“My EA has been running continuously for the last 3 months and stays on even during high-impact news.”
12
New comment