How Far Do The Swing States Swing?

Thursday: The Paper Bullets Guide To Following Election Day (And The Three Weeks After)

Friday: The Paper Bullets Guide To The 2020 Senate Races

Saturday: Can Joe Biden Win Texas?

Yesterday, we did a deep dive to determine that, yes, Joe Biden can win Texas. Using a similar methodology, I have created a likely range of outcomes in most of the major swing states. Arizona? Just watch Maricopa County. Iowa? If we care about Iowa on election night, we have far greater problems to worry about. Nevada? Jon Ralston already called it. Minnesota? Come on. The very minimum requirement to be a swing state is that some hack somewhere can run a poll claiming their candidate is ahead there.

A few caveats before presenting the results:

  1. I could not construct a reasonable model where Trump wins Wisconsin, Michigan, or Pennsylvania. The one exception to this, of course, is if the election is re-run with depressed Democratic turnout, as it was in 2016. And, in fact, Wisconsin, Michigan, and Pennsylvania had some of the least robust early voter programs in the country, which means those states have generally banked votes at a lower rate than the other swing states. While it defies nearly everything we know about 2020 Democratic voters to expect depressed turnout when Donald Trump is involved, it still remains true that already-cast votes are better than intended-to-be cast votes. So turnout remains an important factor, and if Democrats don’t turn out in urban counties on Election Day in these states, a similar outcome is–well, still extremely improbable, frankly–but not impossible.
  2. What I am trying to do has nothing to do with trying to guess whether early votes are being cast for Republicans or Democrats. The only thing that early vote is truly predictive of is turnout. But turnout is not irrelevant, particularly in states like Texas and Georgia where the electorate has changed since 2016. In order to confirm the change, the new electorate has to show up. Safe to say in the high early-voter turnout states–Texas, Georgia, North Carolina, Arizona–they have, which is what makes the below range of outcomes possible in the first place.
  3. Still not a prediction, but a useful exercise to be prepared for the actual vote count. And for those not as familiar with the process of vote counting, useful for overcoming the Trump Administration’s gaslighting about how votes are counted and what the numbers mean.

Here we go…

These all seem like fairly reasonable outcomes. I’ve never really believed Trump could both (1) go full fascist and (2) maintain the low-turnout environment that he needed to win in 2016. As far as the Great Lakes states go, I consistently overweighted likely outcomes in red counties across the board as a rough proxy for a 2016-like turnout scenario, since those states have not yet put up the votes that most of the other states have. As a result, the most likely outcome outside these ranges is that Biden wins any or all of these states by more than the top end of the range, because Democratic voters do end up turning out in rates similar to other states.

As you go into the Sunbelt states, Biden’s margins improve as you adopt certain assumptions. For example, if Republican votes do collapse in the Atlanta suburbs–a reasonable and maybe even likely outcome–Georgia gets away from Republicans very quickly. The same is true for Texas and the Dallas suburbs. There, a 4% Biden win may not seem like much, but consider how far the Democratic Party in Texas would have come to achieve it. I expect a far closer outcome. This is why the top end of the ranges in those states tends to create a wider set of potential outcomes.

Florida has the narrowest range of outcomes, though I think that may underweigh the likelihood that registered Republicans–particularly seniors–are splitting their tickets and voting for Biden followed by Republicans down the ticket.

Interesting that Biden has added a Cleveland, Ohio stop to his schedule tomorrow. Ohio more than maybe any state–even Texas–really depends on Biden being able to chip away at the suburban/rural vote (which went to Trump 2 to 1 in 2016). But regardless of what happens on Tuesday, it is probably a decent sign that Biden feels he has the luxury to go after what is probably the hardest swing state still left on the list for him.

For some reason, Trump’s last rally on Monday is in Grand Rapids, Michigan. Hot take: He’s going to lose Kent County.

Final prediction map tomorrow, then live blogging starts on Tuesday around 4 pm, just as polls in Florida start closing. We’ll get through this together.

Leave a Reply