I stopped using ma's on smaller timeframes awhile back, however I still use the 50/100/200 sma on the larger timeframes (day/weekly/monthly).
When Al Brooks was asked why he only uses the 20ma, he simply replied that he's been using it forever, and that's just the one he kept using.
I remember when I actively traded SNDK, it would obey the 20 like clockwork, until one day it didn't. I agree with the op, you'll need to keep experimenting with technical strategies until you find one that works more consistently than another.
For larger timeframes, the ma's have greater value, especially when looking at the index. The SPY gapped down to the 100sma when the ES opened "limit down" the other day. There's no way to use fundamental analysis for this opening move. It would be erroneous to argue that the collective 500 stocks on the index had aggregate fundamental value to coincide with the NAV at that particular moment in time. It DID, however, coincide with the 100sma, and that's why SPY bounced. It was just a technical movement.
Large cap stocks that are part of the index also tend to obey the larger ma's when looking at long term charts. When oil crashed to the mid 20's, CVX was down 7% that day, magically trading down to the RISING 200sma on the monthly chart, which at that time was a 7 year low for the stock. This was also the day XLE bottomed at $50, which also happened to be a 10 year support line.
Some bozo from Citibank said oil was then going to $10, but oil staged a comeback and XLE rallied almost 40%. Again, NOTHING had changed in the fundamentals to warrant Chevron to magically bounce at that level, except for the fact that the 200sma was hit, and the buyers stepped up in droves.