An empirical model, to be used as a tool to aid in the definition of the optimal values of the main design parameters of oscillating water column wave energy converter devices, is proposed. An extensive dataset of capture width ratio of the device, obtained from both experimental tests and Computational Fluid Dynamics simulations, is used to formulate the model. The model has been developed by applying the dimensional analysis to select the non-dimensional independent variables of the functional form. A multiple non-linear regression method is used to compute the model power coefficients and empirical constants. The model can predict the capture width ratio of the oscillating water column device given the wave conditions, the water depth, the geometrical parameters of the device and the turbine damping as input variables. It can be used in the preliminary stage of the device design, allowing to comparatively test a considerable number of design alternatives with reduced computational efforts. Though based on regression analysis, the model implicitly includes all the non-linear effects observed experimentally and numerically. The relevance of the proposed model is demonstrated by an example application to a selected installation site in the Mediterranean Sea.