I don't base my investing decisions on TA but TA can be useful for timing to enter or exit a position. It's obviously not 100% accurate but definitely better than a coin flip...IMO.
I've heard that explanation many, many times, but I've never found it persuasive.
There are some interesting papers about the fact that after there are papers published on some market inefficiency, the inefficiency disappears. Of course it is virtually impossible to know whether the 'disappearing' is because it never really existed (i.e. the paper was a p-hacking exercise) or because publishing caused over-exploitation. I suspect some of both - but of course I doubt it will every be conclusively known.
Similarly, I'd be surprised if there weren't people who really can do TA-type investment successfully - but they use niche strategies, don't 'publish' them, and have a self control most TAers do not know or do not have. But I am equally confident that the vast majority of those who claim to be successful TAers are incorrect (just as the vast majority of biotech investors clearly have no idea what they are doing).
That said, my biggest beef with TA is that, by definition, it is a much higher frequency event. A TA investor has to look at stock price, and comment on it, 100x more frequently than a Long Term investor - thus a board with 50% TA and 50% LT investors will have 99% of the posts being about TA. A useless, unbelievably large, flood of noise to the LT Investors trying to use the same board.
Market Data 
Markets 
AI
