Flavonoids are phenolic compounds, secondary metabolites of plants that cause several benefits to our health, including helping the treatment against cancer. These pharmacological properties are associated with the ability of flavonoids in attenuating the generation of reactive oxygen species, acting as chelate compounds or affecting the oxi-redox cycle. In spite of the large number of information in term of SAR and QSAR, no recent review has tabulated and discussed in detail these data. In view of this, we bring here a detailed discussion of the structure-activity relationships (SAR) and quantitative structure-activity relationships (QSAR) models. We have also analyzed the correlation between the chemical structure of flavonoids and analogues to their anticancer activities. A large number of methodologies have been used to identify the characteristics of these compounds with their potential anticancer: multiple linear regression, principal components analysis, comparative molecular field analysis, comparative molecular similarity indices analysis, partial least squares, neural networks, configuration of classification and regression trees, Free-Wilson, docking; using topological, structural and enthalpies' descriptors. We also discussed the use of docking models, together with QSAR models, for the virtual screening of anticancer flavonoids. The importance of docking models to the medicinal chemistry of anticancer flavonoids has increased in the last decade, especially to help in identifying the structural determinants responsible for the activity. We tabulated here the most important examples of virtual screening determined for anticancer flavonoids and we highlighted the structural determinants. The mode of action, the most potent anticancer flavonoids and hints for the structural design of anticancer flavonoids are revised in details and provided here.