I found holy grail, or did I?

Maybe you know this situation, you are browsing mql5 market and found perfect EA (perfect equity curve, and you just need 100$ to turn them into 10000$ in one month), at least by it's description. Then you decide to test it in strategy tester, and see almost the same results as in product description. Great, everything looks good so you decide to buy this perfect piece of software. You can't wait to run it on your live account. After few days this EA makes some profit, and some losses, but no miracle like it was promoted in product desription. If you are lucky, you made little profit, but in most cases you will lost almost your whole account. Well I was there too, and I made this mistake more times, that I would like to admit. 

This is the reason why I decided to learn mql4 coding by myself and make my own EAs, so I can be sure there are no fake results. During my coding journey I realize, there are some pretty shady techniques to make backtest results looks incredibly beautiful, and maybe some of them were used in many EAs that almost wiped my account.

I don't want to make tutorial how to make this kind of EA, so I'm not going to explain how this works, or post source code of this EA.

I tried to code simple EA. It literraly tooks me only few minutes to write this kind of EA, and results was pretty amazing.

I know, equity curve isn't perfect, but for few minutes of work I think it's pretty good. If I were one of the bad sellers that makes this kind of EAs, I will invest some time to make equity curve looks much prettier, put this "Holy grail" on the market and wait for my "deserved" earnings. If you download this kind of EA, backtest it on tick data, you will get same amazing results. But as soon as you run this EA on live chart, you will see, that reality is not that great. I will attach compiled code of this EA to this post, so you can try it by yourself. It works only on GBPUSD H1 timeframe and only in tester.

You should get same results as shown below, if you are using external tick data or even tick data from your broker.

Well as you can see, make "profitable" EA is quite easy, and it can be matter of few hours. So few tips from me, that I learned hard way:

1. invest in Tickstory or TDS, so you can make backtests on long term tick data (10 years or more). You will be surprised how different results you can get. But as you can see from my example, results from long term backest with tick data, doesn't mean EA will be profitable on live account.
   
   
2. always run EA on demo account for few months, to see how it perform in real trading enviroment. If seller doesn't provide rent option or demo version of his EA, it should be warning sign.
   
   
3. don't believe in fairytales. After few bad experiences, personally I don't believe that any EA can make 10000$ profit from 100$ starting balance. It could be possible in short term, but in long term, you will lose all your money. So if you see, that EA has great results, but only in period of few months, that should be another warning sign. If you will test this kind of EA on longer period, you can get very different results.
   
   
4. be cautious if you see EA, that has only few inputs parameters, and you can't fully understand how it works.
   
   
5. test EA on diferent period of year. If you see in product description, that backtest was made, for example from 1.1.2016 to 31.12.2018, try to make test from 1.1.2019 to current date, or even better, try to make at least 10 year backtest.