Turnout projections On the Monday before the election, Derek Ryan tweeted out that projections showed 8.2 million people were likely to vote. On Tuesday afternoon, before polls closed, he revised the number to 8,166,750. The final total number of voters ended up being 8,334,221 (based on the number of votes cast in the senate race). Our projection was only off by 2%.
Who were the early voters with no primary election history? Of voters with no primary election history who voted early in the top 15 largest counties, 60% lived in precincts that Hillary Clinton carried in 2016. This accounted for a +367,848 advantage of voters who lived in Clinton precincts versus those who lived in Trump precincts. Assuming that a large portion of these people were Democrat voters, this advantage immediately negated the advantage that Republican Primary voters had over Democrat Primary voters (1.2M Republican Primary voters voted early compared to 1.0M Democratic Primary voters who voted early). In counties where Republicans saw significant losses, the percentages were even greater. For example, in Dallas County, 82% of these voters lived in precincts carried by Clinton. Below is a breakdown for the largest counties:
Who were the early voters with no previous primary or general election history between 2010 and 2018? All data used in this presentation takes into consideration a voter's previous election history for the previous two presidential and two midterm elections. Of the people who voted early who had not participated in a previous general or primary election from 2010 through the 2018 primary:
-62.4% were under the age of 40 (262,381 voters).
-52.7% were women.
-45.6% had an effective voter registration date of 1/1/2018 or later; 24.1% registered after the 2016 General Election through 12/31/2017; and 30.3% had a voter registration date prior to the 2016 General Election. State House races Perhaps one of the larger victories for the Democrats on Election Night was being able to win a dozen state house seats. I had projected that the Democrats would pick up at least nine seats. As mentioned in the section above, I allocated people with no previous primary election history based on how their precinct voted in the 2016 presidential election. Based on this, I came up with the "Trump model." I used this model to simply see how early voting numbers would look if voters broke the same way they did in 2016. As you will see below, the districts significantly under-performed this model. Is Texas turning purple? I have seen some people I know and respect say that Texas should now be considered a purple state while others say that it is not. I am taking a "wait and see" approach. While Senator Ted Cruz won with only 50.9% of the vote, Governor Greg Abbott was able to win with 55.2%. In the top 15 counties I was tracking closely, Lupe Valdez outperformed Abbott 50.4% to 47.7%. However, in the remaining 239 counties, Abbott won 71.8% to 26.8%. Assuming that rural Texas remains heavily Republican, Democrats will need larger vote margins in the largest counties to counter the rural Republican vote. While this may happen in the future, it may take a few election cycles for it to happen. Having said that, the results from last week do mean that a lot of new areas of the state will likely be in play in 2020:
-Republicans won nine congressional races with less than 55% of the vote.
-Republicans won seven state house races with less than 52% of the vote and an additional ten races with 52-55% of the vote.
-Several Republican state senators will be on the ballot in 2020 in areas which saw large growth in Democrat voters. 2019 and 2020 are going to be an interesting few years here!