,
4 tweets,
2 min read
Read on Twitter
1. This is an interesting claim from Justice Roberts. He—correctly!—notes that gerrymandering is not perfectly predictive. But let’s zero-in on the question of split-ticket voting: 
2. Here (thank you, @pbump) is a chart that shows voters splitting their votes for Senate and president, over time. There’s a trend here:
@pbump 3. But what if presidential preferences are driving this? Here’s another chart (thank you, @geoffreyvs) showing ticket splitting when the top of the ballot is a gubernatorial bout. And, yeah, there’s a trend here, too.
4. Maybe 10 percent of the electorate is still “many voters”! But the shift here—away from ticket splitting—is both real and large. Voters are growing more predictable, which increases the returns on partisan gerrymandering. That seems worth noting.
