Is back-testing any good?

The purpose of back-testing is to see how an EA is doing in the past and to optimize it so that it bring best performance. Remember that when we test it on a time range (say from Jan 1, 2022 to April 1, 2022), then optimize it and then save the input setting, we only make that EA to produce best result for that particular time range. There are many EAs for sale in different market (mql5.com is one example) that have a very impressive equity curve, an un-resistible return with small maximum drawdown, and the price is only some thousand dollars or even only a few hundred dollars. Something like 100% return in a month with only 10% drawdown-max, selling for 600USD. Un-experienced traders will buy it with high hope that it will make their account grow without them having to trade!

If the EA can do that kind of return, the seller wouldn't sell it with that price. He/she will just use it to double his account every month, he always "trade" other people's money or get a loan and getting very rich with no sweat and with no risk.

The reality is no one know for sure how the EA will perform in the future!