Consider this a close of a loop I started two weeks ago. When considering voter models in elections, one has to understand the shape of an electorate in each cycle. One of the key measures is party identification, and Gallup has been the gold standard on that metric. In fact, they’re about the only pollster that measures this on a consistent basis.
Two weeks ago, the split had Republicans up one at 28/27, and independents at 42%, with those splitting in slight favor of Democrats at +3. The week after that, Democrats had bounced back to lead 31/29 and a +8 lead among the 38% identifying as independent. Today, however, Republicans have ht their highest mark since February, tied with Democrats at 31%, while indies hit their lowest level since April:
The electorate has been trending — ever so slowly — in the GOP’s favor all summer and fall. And among the decreasing portion of non-affiliated voters, Democrats only have a +4 advantage, 49/45.
Consider what this means in terms of the historical model for Gallup’s final reading in the previous six elections. I laid this out in the previous VIP post, but this time I’ll just chart the differences to make the point more plain:
- 2018: D+6, indie split 42/53
- 2016: D+4, I split 43/46
- 2014: D+6, I split 41/46
- 2012: D+5, I split 42/50
- 2010: D+3, I split 43/44
- 2008: D+5, I split 40/51
As I pointed out two weeks ago, Democrats won when they had a +5 or better advantage over Republicans in the overall partisan split and at least an eight-point advantage in the indie split. Otherwise, Republicans tend to win — and win big in 2010, for instance, while still down three points on party identification.
Why does this matter? Democratic party registration tends to be more concentrated in high-population blue states, such as California and New York. That excess cuts against national polling, not just in presidential contests but also in the generic congressional ballot, where anything better than a D+5 for Republicans usually signals a decent night. This is the first time in a generation that Gallup shows parity between the two parties on the eve of an election.
That might — might, mind you — signal that a high-turnout election will actually be good news for Republicans, especially in more competitive states. It’s something to watch, and if you’re looking for reasons for optimism, it’s at least an indirect, correlative indicator of potential good news.
