I think the reason is obvious - why would Google want to treat blogs differently because of something so trivial as subdomain vs directory?

I'll buy that it may be a challenge to determine which subdomains should be treated as part of the main domain and which should be considered separate. BUT you can make exactly the same argument for directories (or pages for that matter). In the end, the selection is meaningless - either the site is meant to be an extension of the main domain or it's not. Figuring out which is which is an exercise in determining relevance. And given that Google has spent an obscene amount of effort on that very problem over the last couple years, I think it's reasonable to assume they've got this one licked.

But I really wish someone had some data. Or even an anecdote. Because I could be totally overestimating Google's sophistication.