The National Center for Science Education critique of Explore Evolution argues that molecular data confirm the theory of common ancestry. This might be the case, if it were possible to show how molecular data could also disconfirm the theory of common ancestry.
But here, as elsewhere in their rebuttal, the NCSE presupposes common ancestry as a first principle. Molecular data should fit to a monophyletic tree. In this respect, of course, the NCSE rebuttal reflects widespread reasoning, and actual practice, within evolutionary biology and systematics. Start with the geometry (or topology) of a tree, and locate the data on that geometry. As Sober and Steel (2002, 395) note,
This proposition [of common ancestry] is central because it is presupposed so widely in evolutionary research. When biologists attempt to reconstruct the phylogenetic relationships that link a set of species, they usually assume that the taxa under study are genealogically related. Whether one uses cladistic parsimony, distance measures, or maximum likelihood methods, the typical question is which tree is the best one, not whether there is a tree in the first place.
But what if common ancestry isn't the case? Could the molecular data send that signal back to the investigator? Probably, or indeed almost certainly, not: if one presupposes a tree in the data, then it is a tree that one will find. As this flowchart demonstrates, methods in molecular systematics effectively preclude the discovery of relationships other than monophyletic trees.
Sober, Elliott and Michael Steele. 2002. "Testing the Hypothesis of Common Ancestry." Journal of Theoretical Biology, 218: 395-408.