Measures of frequency selectivity from forward masking suggest sharper tuning than those from simultaneous masking. To account for this result, various interpretations involving additional tuning mechanisms have been proposed. In the present study, it is shown that a simple multiplicative relation between on‐frequency forward and off‐frequency simultaneous masking predicts this result quite well. The relation assumes that changes in masking produced by separating the masker from the signal in frequency, in time, and in both frequency and time are related to one another by Weber’s law. The accuracy of the predictions suggests that the limits of auditory frequency selectivity are already established in simultaneous masking and that special interpretations involving additional tuning mechanisms are not required to account for the difference between simultaneous and forward measures. Implications for a dB scale of masking are discussed.