With Texas holding its primary elections on Tuesday, we have moved solidly into what does it mean territory for the November midterm elections. We’ve been watching various signs and indicators for months, of course, curious what the available data says about what’s shaping up to be an unusual political year. So we see things like a big surge in Democratic early voting in Texas and we say, aha! Democrats are enthusiastic about voting! That portends a favorable November.
Sure. Maybe. But the direct lesson from Texas, we can say, is simply this: The relationship between the number of votes cast by Democrats in the primaries and the votes cast by Republicans suggests that the Democrats may see gains in the number of House seats they hold in the state.
Not particularly satisfying. Let’s walk through it nonetheless.
The Texas Secretary of State website has data on primary vote tallies by party back to 1970. Like the rest of the South, Texas was solidly Democratic until it flipped and became solidly Republican. The number of votes cast in the Democratic primary on Tuesday night (according to preliminary data from the Associated Press looking at the race with the most votes cast) was about the same as the number of votes cast in the Democratic primary in 1970, 1 million or so. During that same period, the number of votes cast in the Republican primary … what’s the word for something becoming 14 times as much? Decaquadrupled? Republican primary voting decaquadrupled.
But notice the bar graph at the bottom of that chart. The difference between the Democratic and Republican turnout was lower on Tuesday than at any point since 2008, when the Democratic presidential primary drove a lot of interest.
We can look at this as a ratio: The number of Republican votes cast for every one Democratic vote cast. If the ratio is 1, that means that there was 1 Republican vote for every Democratic one. If the ratio is 2, there were 2 Republican votes for every Democratic one. A ratio below 2 means that more Democratic votes were cast.
Using that metric, we can see that the ratio between Republican and Democratic votes is also much narrower than at any point since 2008.
Why’s that ratio helpful? Because it allows us to compare primary voting to the actual election results in November.
So, for example, we can use data from the secretary of state to see the ratio of Republican votes cast in House races in the general to the votes cast statewide for Democratic House candidates. As the ratio of Republican-to-Democratic primary votes has increased, so has the ratio of votes in November. Unsurprisingly.
Here’s what’s more interesting, though: The ratio of House seats won by Republicans vs. Democrats tracks more tightly with the ratio of Republican-to-Democratic primary votes. In other words, the wider the gap between the Republican primary vote and the Democratic one, the more seats Republicans have won relative to the Democrats in November.
If we take out 2008 — given the unique circumstances of that year — the correlation between the ratio of seats won and the primary votes cast becomes much stronger.
Here’s what that looks like with the ratios compared directly.
Of the 13 years for which we have data, there were four years in which the Democrats got more primary votes and won more House seats. In five years, Republicans led in both, including four of those years in which they doubled up the Democrats on both primary votes and House seats.
In four years, there was a split. In 1996 and 2000, the Democrats won more House seats despite the Republicans casting more primary votes. In 2004 and 2008, the opposite was true. It’s worth noting that each of those was a primary election year, with Republicans more motivated in 1996 and 2000 and Democrats more motivated in 2004 and 2008.
The data suggest, then, that the Democrats are likely to narrow the Republicans’ lead in the number of House seats held after November, but that they won’t assume a majority of the Texas delegation. At no point in a nonpresidential election year has one party cast more primary votes and the other party won more House seats.
One of the problems with our prognostications over the past year or so has been that many of the highest-profile contests that have been contested have been ones in which the Republicans have an advantage. President Trump elevated various people to his administration and his party was left defending various House and Senate seats. The Democrats made gains, but only occasionally scored an upset, which allowed Trump and others to claim that the Democratic wave was overstated.
The results from Tuesday are in the same vein. Those looking for evidence that the Democrats will rout Republicans in even solidly red states didn’t receive it. But those wondering if the Democrats will see incremental gains in a variety of places — enough to take control of the House and, as more of a stretch, the Senate?
Tuesday certainly didn’t undercut that perception very much.