Can there be any doubt that Florida’s race for Governor is coming in for a photo finish?
It’s been a while since we have seen a valid poll showing this race as being anything but a squeaker… and the latest submission from Quinnipiac is no exception. So how does it square with the saltshaker test?
For starters, Quinnipiac did a good job of making sure cell phones were included and ensuring a +3 GOP universe. It is important to note that party registration was not selected from the voter file but self-declared. This means we really don’t know the party balance, especially when “Independent” plus “Other” is over 35%. We must presume then that these are what voters would like to be, but not as they actually are.
But here is what is really odd about the poll. Quinnipiac says that Crist is beating Scott by 4 points among those who have already voted, yet Secretary of State data shows that GOP voters – those who have already voted – are outpacing Democrats by around 12 percentage points and at a rate that is larger than either 2010 or 2012. This would mean that in a universe heavily skewed toward the Republicans, Crist is winning. Does that make sense? Something tells me the sample was among self-identified people who claim to have “already voted” and is not based on the actual voter file.
For that reason, Quinnipiac is probably correct on the overalls, but the internals have some odd inconsistencies and should be taken with a grain of salt.