Experience. There is no ONE good indicator, and that is why so many lose at this game. But experience, backtesting EVERYTHING, one finds that excellent trading is constantly seeing relationships between Price, Action, Speed, Time, Indicators, S/R, Distance, Volume, Emotions, Reasons, and my favorite "Where to Screw the little guy the Worst". I have found by watching one indicator for years, you develop a sense of what it should do in relationship with Price and when it doesn't, NOW you are on to something. But too often people jump around and they never get good at one thing, by time you find one thing, they are out of money. Chart enclosed, normally EMA's track price pretty good, but have you ever noticed when it does something abnormal? Abnormalities make higher profitable percentage trades, cause no one is looking for them. But the chart pattern/indicator to get you in, seldom ever is good enough to get you out, and the money is in the getting out. Having one signal to get in and have 40 reasons to manage the trade.
There are many indicators available so practice with the indicator like bollinger bands and if u are able to get the positive results from it then follow it.
There is no one-size-fits-all technical indicator. What works in trending markets will not work in choppy markets and vice versa.
Just to be awkward.. The timing indicator I use (custom) works equally well in a range as it does in a trend. It also works on multiple timeframes, allowing identification of higher quality situations, and then the timing of entry for those. Its effectiveness was accidental - in trying to overcome a 'buy the high' problem, I thought it up - and it turned out much better than I imagined. And its very simple. For the rest I use classic chart stuff and simple ma's.
The rest? What does that mean? And why does your custom timing indicator need to be supplemented by primitive stuff like SMAs?
The rest? Its all layers of permission using an effective tool to accomplish that particular task. Re: supplemented by 'primitive' stuff.. First, let me address 'supplemented' The whole point of a 'timing' method, is that it tells me when to enter - and that's all it does. Its just one component. Used on its own in isolation, taking every signal on every market, it would lose money. Like taking a cog out of a machine, you wouldn't expect that cog to work on its own. But used in conjunction with a layered approach to the decision process, its about as good as it gets. MAs are simply one part of that layering. Second, ill address 'primitive': If a market is going up, its going up. Adding complexity to a tool that signals 'going up!', is not only unnecessary, but can be a hindrance if it switches state too often. In this respect, MAs have a place. Its the layering of *simple* decisions which makes it all effective - not the complexity or brilliance of any single component.