which tick data service is best for 99% modeling when doing back test?

 
Hi, I am currently using Tick Data Suite to test a very simple EA. 

It is a martingale EA with no entry requirements. Basically it just opens trades when auto-trading is switched on and uses martingale to exit with a profit. 

So basically it will make money when there is volatility and retracements within the calculated equity range. 

However, as I test back further, the annual profit drops drastically. I am wondering if its because the quality of older data is lower and hence the difference in backtest

For example, for 2025, it might show a profit of 75K year to date and something at a similar rate for 2024. But for 2020, the entire year profit is like 20K.

I am new to back testing. Any guidance is greatly appreciated. 
 
  1. When asking a question, please explain in detail, and if relevant, also include screenshots, log output, example code, etc.
  2. Is this MT4 or MT5?
    1. If it is MT4, then you posted in the wrong section (see the MQL4 and MetaTrader 4 section).
    2. If it is MT5, why are you using external data when a broker already provides tick data for testing?
  3. Markets continually change. You can't expect it to be equally profitable every year.
  4. Martingale is not a strategy, it is a money management method, and a dangerous one at that. It can easily blow your account or cause very deep drawdowns from which it is especially difficult to recover.
  5. Also, discussions about 3rd party products and services are not allowed on the forum. Please do your own research with regard to the 3rd party product/service.