Greg's Opinion Greg's big blog of whatnot


4-wk sprint: Early Voting Starts (Plus,Mail Ballot Voters through 10/16)

Today is the first day of Early Voting. Starting tomorrow, the metrics get a bit more interesting since we'll have the Early Voters added to the mix. As for Mail Ballots, here is the lay of the (Harris County) land, so far. I've included the count of ballots returned as of the date in each column.

                DEM Support
District | 10/13/2014 |  10/16/2014
 COUNTY  |   54.65%   |    52.97% 
  HD126  |   29.11%   |    28.86% 
  HD127  |   31.40%   |    31.19% 
  HD128  |   38.04%   |    36.47% 
  HD129  |   38.64%   |    37.64% 
  HD130  |   29.66%   |    28.70% 
  HD131  |   82.35%   |    80.79% 
  HD132  |   34.09%   |    34.97% 
  HD133  |   27.66%   |    26.50% 
  HD134  |   44.92%   |    43.21% 
  HD135  |   40.41%   |    39.41% 
  HD137  |   53.76%   |    52.12% 
  HD138  |   36.82%   |    34.96% 
  HD139  |   80.24%   |    77.84% 
  HD140  |   78.04%   |    76.30% 
  HD141  |   86.96%   |    86.79% 
  HD142  |   86.90%   |    86.55% 
  HD143  |   76.72%   |    76.43% 
  HD144  |   62.14%   |    59.63% 
  HD145  |   66.75%   |    65.18% 
  HD146  |   79.75%   |    78.33% 
  HD147  |   85.73%   |    84.86% 
  HD148  |   69.01%   |    67.68% 
  HD149  |   55.05%   |    53.16% 
  HD150  |   26.55%   |    26.14% 
 Ballots |  26,111    |   32,128

For the sake of context, there were 55,460 Mail Ballots cast in 2010. And Campos notes that 40,566 were returned as cast ballots on the first day of Early Voting in 2012. I'm hoping someone saves me the effort of having to go dig up old PDF files to see how all prior elections look compared to this one.

Theoretically, if the 2014 count of mail ballots is the same as that of 2010, then what we're seeing is a significant shift in who is voting by mail. Ultimately, though, you'd expect to see an increase in the number of mail ballots over 2010. That we may see more than 2012 is pretty big news. And it's not like Dems even need to win this category. Winning just 45% among Mail Ballots would represent improvement.

The update for tomorrow should also include the first batch of Early Voters. Regardless of how many voters have shifted from voting Early or on Election Day to casting a ballot by mail in this election, I think there still has to be some factoring in for where the GOP votes are, so far. The first week of Early Voting should be a place to see the GOP catch up.

I don't have good metrics on this from 2010, but the 2012 experience basically saw the following: GOP wins the first week of EV; Dems win the weekend; and the second week was a draw. And since the number of votes is heavily weighted toward the Saturday and final two days of Early Voting, the net result was a showing close to 50-50. Overall, it may seem like an awful lot of number-crunching to end up with a shoulder shrug that passes for a guesstimate on how the election turns out. But if that's where we end up in 2014 in Harris County, that would be an improvement over 2010.

Anyways, the level of reporting gets a lot more complicated with Early Voting. So feel free to occupy a small sliver of my free time by going to cast your vote.

Related Posts:

Comments (4) Trackbacks (0)
  1. What do the percentages in the DEM Support columns mean? Is that Democratic support from the people actually mailing in ballots that were received? If so, is that based on a comparison with primary party comparisons?

  2. Technically, it’s just the average of the Clarity Partisan scores assigned to individual voters. It takes in a lot of factors in addition to Primary history and is aimed at giving more insight into partisan likelihood for the large number of voters who don’t vote in primaries.

    So it should be seen as a baseline for Democratic support. Individual candidate support will typically fluctuate a few points north or south of that.

  3. What causes the numbers to change day to day? Are they looking at propensities of individual voters, i.e., are they tracking from whom the mail ballots are received?

    Thanks for doing all of this!

  4. As new individual voters are added to the voter rolls reported, the mix of scores change, giving a different average.

    There’s another way to calculate that I’ll try to start this week, that assumes a precinct Dem % is fixed (either for a particular candidate or average) and that a Dem % can be calculated by putting the vote counts by precinct into a spreadsheet to calculate a result. They typically end up being very close. But I like to use both to end up with a range of support level.

Leave a comment

Trackbacks are disabled.

Fatal error: Call to undefined function xwp_footerx() in /nfs/c01/h12/mnt/4748/domains/ on line 18