From my experience with MACD specifically, it basically came down to the fact that with longer periods (or less sensitive thresholds), you lose out on most of a price swing by the time your trigger fires, and with shorter periods (or more sensitive thresholds), you end up making a lot of trades that don't move the needle (can also rack up significant fees if you aren't careful). You can find something reasonable in the middle, but by doing so, you're decreasing both risk and potential returns.
There are at least two other important considerations apart from the indicator itself: how to actually execute effectively with limited funds (when your trigger fires, do you trade everything you can, only a certain percentage, or some variable amount based on technical factors? - that's an entirely separate algorithmic rabbit hole); and even if you do find a good algorithm and it performs great in backtesting historical data, it's almost always harder to achieve the same results in real market conditions (it's difficult to simulate the spread and dynamics of limited market depth).
Would you care to elaborate on which indicators you tried, and what happened?