Here are today’s numbers, and here are the daily totals from previous years:
And here’s a select comparison:
Year Early Mail Total Mailed
=======================================
2017 11,953 7,513 19,466 19,581
2015 36,322 19,789 56,111 42,520
2011 10,818 3,823 14,641 13,697
2007 8,080 3,126 11,206 12,775
So 2011 appears to be the closest comparison so far. That might imply a much higher level of turnout than what I’ve been suggesting, but I’m not prepared to believe that yet. The main reason for this is that less than 40% of the vote was cast early in 2011, and I seriously doubt that’s what we’re going to get this time. Odd year elections skew more towards Election Day and less towards early voting than even year elections – in 2015, just over half of the vote was cast early – but I think this year we will see a higher percentage of the vote cast early. The message from the County Clerk is to take advantage of the early voting period because a number of polling sites are unavailable thanks to Harvey, and I think people will heed that. We’ll take our guesses about that later in the EV period, but for now just keep that in mind. 2017 may be a bit ahead of 2011 in early voting, but I suspect that’s because more people will be voting early than usual.
It should also be noted that these reports encompass all of Harris County, so some of those numbers above are not for Houston or HISD. I’ve gone through this exercise before, but let’s review the percentage of county turnout that was in Houston in these elections:
Year Harris Houston Share
==============================
2015 421,460 268,872 63.8%
2013 260,437 174,620 67.0%
2011 164,971 121,468 73.6%
2009 257,312 178,777 69.5%
2007 193,945 123,413 63.6%
2005 332,154 189,046 56.9%
2003 374,459 298,110 79.6%
“Share” is just simply the percentage of the county vote that came from Houston. There’s a big span here, but that comes with an asterisk, because the conditions were not the same each year. For example, in 2015 and 2007, Harris County had bond elections in addition to the state constitutional amendments. In 2005, the notorious state anti-gay marriage referendum was on the ballot, which coupled with a non-competitive Mayoral election meant a much larger county share. Finally, in 2003 there was the Metro referendum, which covered all of the county. There were also no state constitutional amendments on the ballot, as those had been voted on in September, to enhance the odds of the tort “reform” amendment passing.
Bottom line, with boring constitutional amendments on the ballot, I’d suggest that county/city ratio will be like the other years, which is to say between 67 and 73 percent. Let’s say 70%, just to split the difference. That’s another thing we’ll have to take into account when we do our projections later on.