The problem with much of technical analysis is that most financial time series (such as stock prices) are pretty close to being a "random walk". The "random walk hypothesis" states that stock market prices evolve according to a random walk and thus cannot be predicted. (This is consistent with the "efficient market hypothesis.") However, stock prices are not *exactly* a random walk. For one thing, there is generally a small deterministic (predictable) component. The implication is that much of technical is an effort to detect a small deterministic trend within a larger random process, and so it often doesn't work very well.
That's why there's been increasing interest in recent decades in analyzing stocks, ETFs, etc., for *cointegration*. It is often possible to find a linear combination of 2 or more stocks, stock market indices, ETFs, etc., which exhibits a large deterministic component. (Much of the "random walk" component is cancelled out.) This makes such a portfolio more predictable. Most algorithmic traders are familiar with "pair trading", in which one trades the spread between two cointegrated stocks, bonds, ETFs, etc. When two such price series diverge, or "get out of whack", one bets on their convergence. This can be a profitable trading strategy. However, pair trading has become so popular that it is now difficult to make money trading stock pairs. (When too many traders are using the same approach, it becomes less profitable.) Traders who use cointegration-based trading strategies (such as myself) are increasingly looking at ETFs, as well as cointegrated portfolios having 3 or more ETFs. In my experience, such approaches can be quite profitable.