Of course we have a common ancestor- not necessarily H. sapiens- but he's certainly wrong about 7000 years ago. That's so recent as to be virtually historic, and would mean that, for example, Australian and American aboriginals populated those continents since then. There's a huge mass of evidence that American aboriginals were in place 25000 years ago, and 20-40000 years is probable for Australia.

Mitochondrial comparisons place our most recent bottleneck at about 150000 years ago. Of course, this depends on certain assumptions about rates of mutation, but is likely to be the right sort of order of magnitude. These guys have just got their sums wrong, their model is too simple. When you come up with a result that defies common sense and existing evidence, it's your job to check against reality.

I think we evolutionary materialists should celebrate Mitochondrial Eve just as Christians celebrate Christmas. Problem is, I don't know when Mitochondrial Day is.