I encountered so many triangles that didn't work as expected. Is there anything to figure out when it's more likely to work or not?
Is there any evidence that 'triangles' should work at all? What academic studies can you cite to support the reality of 'triangles'?
Momentum, mean-reversion, and factors such as quality, value, and smallness have academic support.
Here's an article: Momentum and Value Everywhere
While I don't agree with the conclusions of this article, the author wrote what I was going to say in a more eloquent way:
So you read a book once that explained a bunch of different price patterns and you figured that looking for these would make you money in the market. But then time has passes and you only remember the patterns with the silly names: head and shoulders, cup and handle, etc. You’re constantly looking for these things in all of your charts, but since these are the only ones you know, you find yourself calling them out when they’re either not there or not yet confirmed. Then you start to get mad at technical analysis because your “patterns” aren’t making you money.
The market is here to make you look like a fool. That is the number one job of Mr. Market. Do you really think it’s that simple to do this correctly? You read something silly in some book and that’s it, you’ll make money forever?