This research investigates Bitcoin price prediction by reviewing the current state of the art, comparing Time Series and Deep Learning models, evaluating their performance on a range of metrics, and assessing the selected model’s real-world applicability. It reviewed existing studies on Deep Learning, Time Series, and Bitcoin, and utilised 44,419 hourly Bitcoin data points for data visualization and mining. The findings showed that ensemble models, particularly stacked ensemble, outperformed other models in predictive accuracy, with the lowest error metrics and highest R2 value. Deep learning models also performed well, but with slightly higher errors. Time Series models were inadequate for Bitcoin price prediction, as evidenced by their negative R2 values. These results contribute to our understanding of effective modelling approaches for Bitcoin price prediction. The study also suggests promising avenues for future research, such as fine-tuning ensemble models, incorporating advanced feature engineering techniques, and exploring volatility forecasting. Thus, the study offers a valuable contribution to the field of cryptocurrency research, advancing knowledge on Bitcoin price prediction and fostering a deeper comprehension of the intricacies underlying cryptocurrency markets.