President Barack Obama and Republican challenger Mitt Romney are neck and neck in opinion polls, but there is one area in which the incumbent appears to have a big advantage: those who have already cast their ballots.
Obama leads Romney by 59 percent to 31 percent among early voters, according to Reuters/Ipsos polling data compiled in recent weeks.
The sample size of early voters is relatively small, but the Democrat’s margin is still well above the poll’s credibility interval – a measurement of polls’ accuracy – of 10 percentage points.
Relatively small is one way of putting it: It’s only 361 voters. And Romney campaign political director Rich Beeson is out with a memo today calling the poll’s findings “flawed and untrue”:
Yesterday morning, a poll was released that claimed President Obama has a “big advantage” over Governor Romney among those who have already voted. This claim can easily be proved flawed and untrue.
To begin, it is important to outline a key distinction between how the campaigns are approaching the early vote. This can be determined not by asking “how many ballots have been cast;” instead, by asking “who has cast their ballot”? Many of the Democratic ballots are from high propensity voters who would almost certainly be voting on Election Day – meaning that President Obama is cannibalizing his turnout on November 6th. Governor Romney’s early voting effort has been, and will continue to be, focused on low propensity voters, which means his Election Day turnout will not be negatively impacted by the early vote program.
First and foremost, only 5% of early voting has completed. In 2008, nearly 16,000,000 votes were cast before Election Day in the swing states. So far this cycle, just over 950,000 votes have been cast – or 6.01%. Therefore, 94% of all early votes have not even been cast yet. To claim a “big advantage” based off of a phase that’s just 6% complete is almost as absurd as predicting the outcome of a baseball game after the second out.
Setting timing aside, this was a national poll conducted online with a margin of error around 10%. National polls do not capture the actual state of the race (especially as it relates to swing states) and aside from the usual hesitations with online polls, the margin of error should raise serious questions as to the poll’s accuracy.
Further examination shows just 361 people account for the entire nation’s sample of early voters. Even worse, only 115 of the people (or 31.8%) make up the sample in the swing states. So not only is the sample extremely small – but it’s meant to represent the votes of over 950,000 people who have already cast a ballot.
Perhaps most troubling is that, on average, less than 10 people per swing state made up the sample. This is hardly an accurate representation of those who have already cast a ballot – and it certainly isn’t an adequate indicator of an “advantage” for one candidate over another. The largest sample for a swing state was Florida at just 22 people. And in the states of New Hampshire and New Mexico, just two people were surveyed. In Colorado, a state where approximately 79% of every vote cast in 2008 came before Election Day, the poll only surveyed four people – but what should baffle observers here is that early voting has not even begun for the general population yet (only military).
Despite what this poll has claimed, it is clear that Governor Romney is poised to over-perform among those who will vote before Election Day. In swing states that have party registration, ballots cast belonging to registered Democrats only hold a 6% margin over those cast by registered Republicans. Furthermore, of the approximate 6.7 million voters in those states who have requested ballots, but have yet to return them, Republicans narrowly edge Democrats (GOP: 2.12 million, DEM: 2.02).
Although still early in the process, our campaign continues to be encouraged by the initial requests and returns we have seen. That, coupled with the Governor’s momentum and his advantage among enthusiastic voters, undeniably puts President Obama in an extremely compromising situation with just over three weeks remaining in the election.