1) The housing price does not revert to the mean so much as it reverts through the mean. It’s interesting that residential real estate has, for the two previous booms, had a somewhat sinusoidal characteristic about the 120% level. There is no appreciable period of time where housing prices settle at the historic mean. Indeed, if the change in housing prices slows or appears flat, that seems entirely indicative of a shift from boom to bust or bust to boom, based on the previous two cycles.
2) Over the first two cycles, the peak above the mean on one cycle was the same distance from the mean as the valley below the mean on the next cycle.
Will that pattern carry through this cycle? If so, that graph implies far more than a 50% drop from peak to valley.