Quantitative predictions of propagation, emission, and absorption of electron cyclotron (EC) waves rely on a solid theoretical background and are routinely employed in the analysis and preparation of present and future fusion experiments. Nonetheless, open problems still exist, and improvements are possible also within well-established models to make them faster, more accurate, or more general. This paper presents some of the recent advances in the theoretical investigation of EC waves. Particular emphasis is put on the "standard" approach based on the short-wavelength approximation and linear or quasi-linear computation of wave absorption, on the theory of electron Bernstein waves, and on the applications of cyclotron heating and current drive for ITER.