As we know, Early Voting this year is running well ahead of recent years. In the first two days, 8295 in person ballots have been cast, compared to 6685 over the first two days in 2003. I’ve suggested that this may mean that turnout this year will be higher than previously thought. But that was based on an assumption that the early voting pattern this year will be similar to what it has been in previous Novembers of odd numbered years. Given the huge shift to early voting we saw in 2008, is that a reasonable premise?
Well, we do have some more recent data to help guide us. I’m speaking of the special election and runoff from this year in District H. Here’s how the vote went for the election in May and the runoff in June:
Month Mail Early Total Pct Early ====================================== May 630 1,251 4,141 45.42% June 788 1,453 4,680 47.86%
That’s a considerable change from the elections of 2003, 2005, and 2007, in which about 25% of ballots were cast early. If we assume that 45% of votes will be cast this week and next, then the projection of 110,513 early ballots I made before translates into a total turnout of 245,584 for Harris County, and 184,188 for the city of Houston, assuming Houston has a 75% share of the vote.
Now I think that projection is too low – I think 200,000 is the floor, and 250,000 is well within reach. But it’s possible, and just as there were some giddy predictions about turnout in Texas and Harris County last year that didn’t quite pan out based on the high volume of early voting, we may just not really understand yet what the new normal level is for odd numbered years. I’ll make a final guess later, but for now, I don’t think we really have a good idea yet how big the electorate will be this year.
By the way, the Day Three totals are now in the spreadsheet. They dropped a little from the first two days, but as you can see from the previous years, a dip on Wednesday isn’t that odd. If it doesn’t bounce back today, however, then we can say that things have slowed down.
I’d guess that H has a lower turnout than the City as a whole, so extrapolating to 184,000 would result in too low a number. Why don’t you compare H vs City turnout levels and adjust your 184k number up based on that?
JJMB – I’m just comparing early voting versus Election Day voting. It’s entirely possible that Dist H votes early at a different rate than the rest of the city, and it would be interesting to explore that if so, but unfortunately I don’t have that data available to me.