Numerical simulations are employed to investigate possible biases in Medium Frequency (MF) wind estimates. To explain apparent biases inferred from Arecibo Initiative in Dynamics of the Atmosphere (AIDA) campaign measurements, Hines et al. (1993) have suggested that MF radar estimates of horizontal winds should be biased toward the phase speed of perturbing gravity waves if the MF scattering process is modulated by the passage of such waves through the scattering volume. One of the proposed scenarios involves modulation of the tilts of anisotropic irregularities by the perturbation winds associated with gravity waves. We use a two-dimensional propagation model in which anisotropic scattering centers are represented by conducting line segments located at random positions inside the radar scattering volume. The scatterers are advected and tilted by the superposition of a mean wind and the perturbation winds associated with a gravity wave. Simulated complex time series, which would be recorded by a two-receiver antenna array with a typical MF radar system parameters, are analyzed to obtain spaced antenna (SA) wind estimates for comparison with the specified wind speed at the center of the scattering volume. Numerical simulations are carried out for many different sets of gravity wave parameters. Each simulation is performed twice, once with essentially isotropic scatterers (line segments of length λ/2) and again with anisotropic (“specular”) scatterers of length 10 λ. Numerical results indicate that individual SA wind estimates are accurately predicted by the wave-induced error model of Kudeki et al. (1993). This is true for both isotropic and specular scatterers as long as the gravity wave has a horizontal wavelength larger than the horizontal dimension of the radar range cell. The results also indicate that estimated winds averaged over one period of the perturbing gravity wave exhibit statistically significant biases when the scattering centers are highly anisotropic, but the bias is not necessarily toward the phase speed of the perturbing gravity wave and, more importantly, the magnitude of the bias exhibited in our simulations was always less than 15 m/s. Implications of the numerical results are discussed in light of observed discrepancies as large as 40 m/s in recent AIDA campaign comparisons between MF and incoherent scatter radar winds. In particular, it is argued that the simulation results support the wave-induced fluctuations model put forth by Kudeki et al. (1993) as an explanation for the AIDA results.