I have been trading the forex market for over 16 years and in all my years as a trader I have witnessed several arguments about the effectiveness of forex robots (expert advisors). Many quarters argue that manual trading is the best while the others believe that Expert advisors (robots) don't work.
My first concern as a forex trader was to learn how to trade consistently. I learnt by trying so many things to see the ones that had positive expectancy. It did not help that back in those days there were very few websites to learn from.
Many years passed and I have been able to learn how to make consistent profit. I have also learnt that "the fastest way to success is slowly". This is arguable but history is clear on this. In the future things may change but going by past events, the fastest way to success is slowly.
When I learnt how to trade the forex markets and started becoming consistently profitable, I became very lazy, more so as I was trading on the H4 timeframe. My profit margins were small but very consistent but as a result of boredom I decided to try automated trading.
I am still unable to properly express my trading strategy in codes but I have tried several times.
I have spent thousands of dollars buying several robots (expert advisors) that did not work. I created a particular robot but after testing the EA with the MT4 strategy tester, I did not have the motivation to try it in a live account.
Recently, I stumbled on an argument online. a group of seasoned forex traders with many years experience were debating the reasons why , several forex robots work well while backtesting but flop in real time trading.
Here are some of the points I picked from the discussion:
1. When backtesting, the backtester(MT4) does not consider floating spreads. It is possible to even choose a fixed spread for testing. During overnight swaps, the spreads widen in real time but goes unrecognized in backtesting. This happens during news events too. This makes the real time result very different from the back test result.
2. Asides the MT5 new python features that can enable machine learning (artificial intelligence ) models to be integrated to EA's, the conventional EA's do not adapt to new or changing market conditions. The regular EA's usually focus on technical setups and are helpless during drastic market changes.
3.The bulk of the EA's that do well in backtest but flop in real time are either scalping , martingale or grid based trading strategies. These trading methods are difficult to manage during spikes and when an EA is built based on any of these strategies , it will most likely not do well in real time.
I finally had the courage to test an EA I built but did not bother to test because I was convinced that it would flop. Unsurprisingly, after a few tweaks, the result in real time is interesting. I figured this is because it is not based on any of the dangerous trading strategies that dominate the MT4 forex EA space.
Test this DEMO and let me know what you think: https://www.mql5.com/en/market/product/77937
My first concern as a forex trader was to learn how to trade consistently. I learnt by trying so many things to see the ones that had positive expectancy. It did not help that back in those days there were very few websites to learn from.
Many years passed and I have been able to learn how to make consistent profit. I have also learnt that "the fastest way to success is slowly". This is arguable but history is clear on this. In the future things may change but going by past events, the fastest way to success is slowly.
When I learnt how to trade the forex markets and started becoming consistently profitable, I became very lazy, more so as I was trading on the H4 timeframe. My profit margins were small but very consistent but as a result of boredom I decided to try automated trading.
I am still unable to properly express my trading strategy in codes but I have tried several times.
I have spent thousands of dollars buying several robots (expert advisors) that did not work. I created a particular robot but after testing the EA with the MT4 strategy tester, I did not have the motivation to try it in a live account.
Recently, I stumbled on an argument online. a group of seasoned forex traders with many years experience were debating the reasons why , several forex robots work well while backtesting but flop in real time trading.
Here are some of the points I picked from the discussion:
1. When backtesting, the backtester(MT4) does not consider floating spreads. It is possible to even choose a fixed spread for testing. During overnight swaps, the spreads widen in real time but goes unrecognized in backtesting. This happens during news events too. This makes the real time result very different from the back test result.
2. Asides the MT5 new python features that can enable machine learning (artificial intelligence ) models to be integrated to EA's, the conventional EA's do not adapt to new or changing market conditions. The regular EA's usually focus on technical setups and are helpless during drastic market changes.
3.The bulk of the EA's that do well in backtest but flop in real time are either scalping , martingale or grid based trading strategies. These trading methods are difficult to manage during spikes and when an EA is built based on any of these strategies , it will most likely not do well in real time.
I finally had the courage to test an EA I built but did not bother to test because I was convinced that it would flop. Unsurprisingly, after a few tweaks, the result in real time is interesting. I figured this is because it is not based on any of the dangerous trading strategies that dominate the MT4 forex EA space.
Test this DEMO and let me know what you think: https://www.mql5.com/en/market/product/77937