To the extent we're able to tell (and we probably can't) it sure looks like the *debate* shift in the polls was very modest but the post-debate Dem and media freak-out shift in the polls may be somewhat larger. Still small enough that all of it could easily dissipate quickly. Or not! At any rate...
As I said, hard to separate. But the one-day polls seem (*seem*) to show less of an effect than those taken later -- which are capturing both the debate and the freak-out effect.
But absolutely it's hard to know.