The equivalent circuit method is used to simulate the lunar semidiurnal geomagnetic tides. Samples of observations of the lunar geomagnetic tide are also presented, derived from analyses of hourly values of the horizontal component of the magnetic field recorded at several observatories. These are compared to the results of the simulations at different seasons. The winds used in the simulation are derived from the Global Scale Wave Model (GSWM) and from those determined by Forbes and Gillette [1982]. The calculations are only carried out for a lunar age of zero (new moon), but this still enables a comparison to be made with the observed phases of the lunar geomagnetic tide. A mostly fair agreement is obtained both for the amplitude and phase of the lunar geomagnetic tide. The agreement is generally better when winds from the Global Scale Wave Model are used rather than those from the earlier Forbes and Gillette model. The July simulation gives poorer agreement: there is corroborating evidence that the phase of the winds near the equator in July should be about 3 hours earlier than that presently given by the GSWM. The situation in December is different from other seasons both in the simulation and the observations, but agreement is difficult to achieve on account of the year‐to‐year variability in this season.