Useful in trend following, in which case lag is irrelevant, to confirm trend and as a rough indication of momentum if say you want to compare two trends.
There are better tools than MA for trend regime but you have to use what you got i suppose , still plenty of lag in MA as trend filter imo but its better than nothing
Just to paraphrase Jack Schwager, successful ones have something that suite their temperament and give them an edge. All methods will fail sometimes and knowing that it has failed is an edge itself. If a method that makes you money, that's a holy grail for you. Be patient and don't put oneself into a corner. Jack Schwager's Zen and the art of Trading, any force will fail. In chapter 12 of Reminiscences of a Stock Operator, everyone was going broke while trying to get the stock profit to pay for that fur coat.
I have found that a short term EMA work very well on tick charts. However I use the median price rather than the closing price of the bar. I use it as an indicator of trend direction and then use a short term adx for the strength. I just look for a retracement against the trend and then jump on board for the continuation. Works very well for CL, GC, NQ, and ES. Other than those I don't use any other indicators.
The key to avoiding the above-mentioned fallacy is in how one defines "rise to its normal level." In my case, I am not simply looking for the moving average to revert to the mean, but rather, I want to see it convert from a down trend to an up trend (or vice versa, depending on the situation). If you define an up trend as higher highs and higher lows, then you will not have a problem with price dropping further as it "rises" to its "normal level." (I have my own unique approach to defining trend direction and do not use this classic description.) If this key concept was completely ignored by my statements back in March 2018—my bad! I suppose it just seemed kind of obvious to me that one should not enter a long position unless price is actually rising, whatever method one uses to verify that this is happening.
%% Exactly; 200 dma helps, plenty of lag + that's good. Some[IBD, investors.com......] use a 40 week moving average; i still like 50 week moving average,also, much slower, less commissions than 40 week ma /stocks/ETFs.If one has to close by 3rd friday could use something faster/less lag.
I use 21-pd ema on all TFs except 13-pd ema on Monthly. Works for me, but most of the time I don't know what I'm doing.