This study aims to assess the quality of two sources of surface winds, i.e. the ECMWF (European Center for Medium Range Weather Forecasts) modeled and the measured data, for wave modeling in the Persian Gulf. A third generation model, i.e. SWAN was employed for wave simulation and the results were compared with the recorded wave data. It was found that ECMWF underestimates the wind magnitude and the results of the wave modeling need to be modified. In addition, it was revealed that the adaptation of the model parameters cannot lead to a comprehensive improvement of the model's results. The calibration of the wave model for high waves led to the overestimation of low waves. On the other hand, the employed measured wind data was found to be a relatively good source for wave hindcasting at the studied location. Extreme value analysis was also conducted based on the measured and modeled wave data to investigate the influence of wave simulation on the estimation of design wave height. It was found that the Weibull distribution is better fitted to the measured and modeled wave data. Modeled wave heights forced by the ECMWF wind showed a different behavior compared with the measured and modeled wave heights forced by the measured wind from the viewpoint of the exceedance probability. A marginal difference was found between extreme wave heights obtained from measured and modeled data.