2018 Gubernatorial Rankings

[Kodak]

For fun, I'm going to take the Atlas 3-poll average method with every available poll and see if it comes closer to predicting the outcomes than any of the other election models. Here is my prediction 100 days from election day.

[Kodak]

My poll-based prediction 90 days out from election day. For Kansas I am still assuming Colyer wins the primary after provisionals are counted; Kelly would lead Kobach by 1 if he pulls through.

And yes, Maine will have an exact tie.

[Kodak]

80 days out from election day, the infallible polls now decree that Kansas shall flip.

[Kodak]

70 days from election day, the polls are generally shifting towards the Republicans, except in Michigan, where Whitmer is now leading by double digits.

[Kodak]

Here's what the polls are saying 60 days from election day. Note the abundance of ties.

[Kodak]

Updated polls.

[Kodak]

Polls as of yesterday.

GOP is falling behind in the Rust Belt.

[Kodak]

Updated polls. New Hampshire and Arizona are slowly swapping places.

[Kodak]

Updated polls. Aside from Nevada, I think the final results will look something like this.

[Kodak]

Updated my polls.



Like the Senate map, I think some of these margins are missing the mark by a lot, but aside from the six states that are within 1%, I'm confident the polls are calling everything correctly.

[Kodak]

Final average of all non-internals conducted after Labor Day. Alaska is limited to 2-way polls or ones conducted after Walker dropped out.





My predictions:

1. Newsom will win by 5-10 points more.
2. Sisolak will win.
3. At least one of Kansas and South Dakota will hold for the GOP.
4. Idaho, Nebraska and South Carolina will hold for the GOP.
5. Everything else should match the map.