This study addresses the challenge of effectively evaluating the performance of automated trading systems based on Machine Learning (ML) techniques, particularly in light of the complexity of financial markets and their associated risks. The central issue identified is the lack of standardization in the selection of metrics to assess profitability and the robustness of strategies in the face of risk. To address this gap, a Systematic Literature Review (SLR) was conducted following Kitchenham’s guidelines to map the most commonly used ML techniques between 2019 and 2024, as well as the financial metrics employed in the evaluation of algorithmic trading strategies. As key findings, the study highlights the need for broader metrics that combine both profit and risk indicators to ensure more robust evaluations of automated trading strategies. Furthermore, it points to a research gap in studies applied to non-traditional markets. It proposes future investigations focused on analyzing the temporal evolution of Artificial Intelligence techniques and assessing the impact of regulation on the development of systems targeting the cryptocurrency market.