My issue is that I'm still not confident that the underlying conditions that created the correlations over the last 50+ years of analysis still hold.
As I was saying all thorugh the primary season, there is lots of reason to believe that the underlying factors that have made politics more predictable over the years are fraying. Huge population shifts, dramatic changes in how these populations abosrb news and views, the fact that the system is terribly complex (by which I mean lots of variables that are mutually dependent) all push us into an increasingly unknown country.
So sure, getting a post convention bounce is better than not getting a post-convention bounce. And we will see what happens when we have the debates and have the two candidates side-by-side. The impact of eliminating the voter i.d. rules in the three most restrictive states may also have significant impact (even in TX).
All this is to say that we will actually have an election, and it is the election that will decide who gets to be President, who are the members of Congress, and who are the members of your state legislature. And whatever ballot issues you care about.
What you chose to do with that information is entirely up to you.