In this chapter we overviewed not only technical analysis but also some corresponding strategies, like neural networks and log-optimal portfolios. These methods are similar in the sense that when applying them, we implicitly suppose that past situations may reappear in the future; therefore we took the courage to challenge the concept of market efficiency and to build up an active trading strategy. In this setting, we discussed the problems of forecasting the price of a single asset (bitcoin), optimizing the timing of our trading, and also optimizing our portfolio of several risky assets (NYSE stocks) in a dynamic manner. We demonstrated that some simple algorithms based on the toolkit available in R can produce significant extra profit relative to the passive strategy of buying-and-holding. We also note however, that a comprehensive performance analysis focuses not only on the average returns, but also on the corresponding risks. Therefore, we suggest that when optimizing your strategy take care of the downturns, the volatility and other risk measures as well. And, of course, you must be aware of the limitations of the presented methods: you cannot be sure to know the return generating process; if you trade frequently, you have to pay a lot of transaction costs; and the more you get rich, the more you suffer from the adverse price impact and so on. However, we do hope you got new inspirations and useful hints to develop your own sophisticated trading strategy.