Here’s an interesting summary from Pollster.com’s Steve Lombardo about the tendency in recent election cycles for Republican presidential candidates to trail in the summer and then make up lots of ground in the fall. The pattern didn’t occur in 2000, and there’s room for debate about its past causes and future applicability:

[The trend] could tell us that Republican candidates tend to do better once the electorate is more focused on the issues and the candidates (similar to what we see in registered voter/likely voter screens, where likely voters–those paying more attention–tend to be slightly more inclined to vote for the Republican candidate), or it could simply be a coincidence based on a variety of external factors related to those particular races and polls.

I’d throw in another, possibly gratuitous, dig at the Democratic political pros: they’ve run some pretty sorry fall campaigns. Their candidates have won anyway when the fix was pretty much in, like in 1996. Republican consultants seem to add a lot more value to their candidate’s final performance. I wonder if their higher productivity means they command better average pay? Some young economist aiming for tenure ought to do a paper.