Search This Blog

Wednesday, November 2, 2016

What The Early Voting Tells Us

I just finished reading/watching what seems like the 99th report on what the early voting tells us about the likely outcome of the presidential election.  The accurate answer is that it tells us very very little.  The media, however, has the early voting telling us how the candidates are doing in various states and perhaps what the weather will be on election day.  The reports are a joke.

Here's the one thing you need to know about all these early voting reports:  no one knows what candidate the early voters chose.  That's right; there are no public results for early voting.  Indeed, none of the states actually count the early votes until election day.  One thousand early voters may have split 99% for Trump and 1% for Hillary or just the opposite.

So how do all these reports come out telling us that this candidate or that one is ahead in the early voting?  The answer is that in some, but not all, of the early voting states, the officials in charge of early voting will tell us how many people have voted and will also break down the results by party registration.  We may know that 55% of the early voters were registered as Democrats in one state and that 62% of early voters were registered Republicans in another.  But that really tells us nothing.  First, only about half of the early voting states give out this sort of information.  Second, among the states that do put out statistics, many do not have registration by party.  That really limits the number of states for which we can get the breakdown by party.  Third, even for these states that do give detailed results by party, there are large numbers of voters registered as independents or not members of either party.  Indeed, in many of these states, the bloc of unaffiliated voters are larger than either the Democrats or the Republicans.  There's no way to tell how the independents voted.  Fourth, for some reason, we know that historically in these states, a higher percentage of Democrats vote early while a higher percentage of GOP voters go to the polls on election day.  Put all this together and the early voting numbers tell us almost nothing.

There is some small bit of information that we can get from early voting results.  In some states, we can see where the early votes are being cast.  In Florida, for example, we can tell what percentage of the early votes are coming from precincts that are predominantly African American.  In 2012, about 25% of all early votes had come from those precincts at this point in the process.  This year, the early vote is about 16% from those same precincts.  This gives us a rough idea that black voter turnout is down so far.  Maybe these voters will come out on election day.  Maybe they will vote early but in the next few days.  Maybe they won't vote at all -- which would be a disaster for the Clinton campaign.  In other words, it's something of an indicator, but not really a very good one.

Remember, the articles and broadcasts about what the early voting is showing are really just nonsense.  Don't believe them, whatever they show.

No comments: