Avar has a good point in general, although I'm not sure it's relevant to the recent San Andreas earthquake history. Over geological time, the actual probability

Avar has a good point in general, although I'm not sure it's relevant to the recent San Andreas earthquake history. Over geological time, the actual probability distributions are unstable, because the boundary conditions causing the events change as plate configuration, climate, etc. changes.