Quote:
Originally Posted by Former DJ
Here's a really deep dive into the 2018 Senate election courtesy of Nate Silver's Five-Thirty-Eight web site.
https://fivethirtyeight.com/features...a-better-time/
Democrats are going to need more than a "wave" election to take control of the Senate - they'll need more like a tsunami.
I don't like the methodology they used to conclude Democrats need a tsunami. The results since 1992 aren't totally relevant to what will happen in 2018. The candidates are different and the political climate is obviously much different. Whatever correlation does exist isn't going to be very predictive due to the small sample size. It's important to know that it's mostly Democrat seats at risk this election and many of those seats are in red states, but we knew that already.
It's not necessary to bring that into prediction models though. The races in each state are set enough to where you can assign reasonable odds to each and figure out the odds of flipping the Senate from there. I haven't done anything like that yet but considering most races the Democrats need to win are tossups at worst, and a wave that drives out an extra 10% of Democrats in each state would make those tossups near locks, you don't need a tsunami to flip, just a decent strength wave that touches every state. The wave we've seen in special elections is more than strong enough, but still unknown is whether it will hold another 6 months, whether it will apply on a more common election day, and whether it will affect every state.