TPM Reader GH makes a pitch for Arizona rather than Wisconsin as the real tipping point state in 2020. For me I’m still inclined to see Wisconsin as the critical state, not because I disagree with any of the particular arguments, simply we’ve seen Dems win it before in recent history. There’s a proof of concept for lack of a better phrase. But these are still very good points worth considering …
I love the work you do. I was so happy to read about your union, and I’m even more supportive of TPM because of it.
On the criticality of Wisconsin, I do agree that it’s looking more like the critical state in comparison to Michigan and Pennsylvania. But Arizona really looks like it’s ready for purple Prime Time. The two states (WI/AZ) are worth nearly the same amount of electoral votes, making them interchangeable for 2020 purposes. And in the 2018 races, the two big headliners in each state wound up as very close Dem victories. Sinema actually beat McSally by more than Evers beat Walker. (2 points in AZ, 1 point in WI)
The polls were a bit skewed in 2016. Nate Cohn, while he was working to run all those Siena polls in fall 2018, noted something that has stuck with me. The polling error in 2016 might not have been a failure of the polls so much as a rapid late shift that didn’t get picked up by polls that had basically wrapped up by E-Day. Cohn noted that the late undecideds in the Siena polls were disproportionately non-college. He found that when those voters did decide, they tended to gravitate toward where you’d expect them to. White non-college to the GOP. Non-white non-college to the Democrats.
In the industrial Midwest, those late-breakers are mostly white non-college. Very typically Trump-backing and Republican. This is why polls showing Clinton ahead in ’16 were off. And it’s why some of the Dem victories in 2018 in that region were tighter than polls predicted. In an extreme example, the Ohio Gov race turned into a solid GOP win when polls showed a modest Cordray win. But this effect occurred in Iowa and Michigan as well.
But the opposite happens in states with high numbers of non-white non-college voters. This is especially important in Nevada, Arizona, and Texas. In all three big headline races in 2018, the Democrats outperformed polls by decent margins. All of those late breaking California Congressional races should be noted here, too. Especially CA-21, which is possibly the most non-white non-college district in the entire country.
All of this is a long-winded way of saying: I think Arizona may be on par with Wisconsin because of these polling biases. Wisconsin is redder than polling makes it appear, and Arizona is bluer than it appears. If Democrats win MI/PA, they only need one of AZ/WI, and I think it’s a close call as to which will be the tipping point state in 2020.
And not to get too far ahead of myself, but if that same effect continues into 2020, it spells very bad things for Republicans in Texas, where polls already show a close race.