Fidel Castro was a major Cuban politician of the twentieth century who influenced different left-wing regimes and political movements in Latin America and around the world. His ability to seduce the masses and captivate his audience relied to a large extent on his rhetorical abilities. Therefore, studying Castro's political speeches is a crucial step towards understanding his political success. Some previous studies have addressed this issue, mostly using small discursive samples and only a few speeches. However, using a small and possibly non-representative sample is likely to lead to biased results. To overcome this problem, this work was carried out on a large corpus of 1,018 speeches and 7,548,480 words, combining state-of-the-art data mining tools and the linguistic discourse analysis methodology proposed by Patrick Charaudeau. Combining both techniques, we provide here a more representative characterization of Cas-tro's main discursive strategies.