Post

Conversation

Nate Silver
@NateSilver538
FWIW, we looked at this since AK's system is complicated, and there's a lot of new code in the model to handle it. But it's basically the model performing as intended given what we believe are empirically well-grounded priors about ranked-choice voting. 🧵
Quote
Basil🧡
@LinkofSunshine
Nate Silver gives Begich a 72% of winning AK-Gov, compared to betting odds which have him at 27%
Image
44.6K
Views
David Watson 🥑
Post your reply

Nate Silver
@NateSilver538
We have the 1st round projected as something like: Begich [D] 41 Wilson [R] 28 Bronson [R] 23 Kreiss-Tomkins [D] 8 So people might look at that and say, well, under RCV, Republicans will eventually get 51% and Democrats (Begich) 49%; therefore, Republicans win.
5.6K
Nate Silver
@NateSilver538
However, that assumes 100% of the same-party vote transfers under RCV (e.g., Wilson gets Brunson's entire 23%) when in reality it's typically more like an 85/15 or 80/20 split, and also a significant number of ballots are exhausted because some voters only rank one candidate.
4K
Nate Silver
@NateSilver538
Long story short: it's a big advantage to have a big lead on the 1st round in RCV as 2nd choices aren't quite as predictable as people seem to assume and a lot of ballots wind up being wasted.
13K
Nate Silver
@NateSilver538
I will say, though: it's a bit hard to know how to show RCV races on the model landing page. Should we be showing 1st-choice votes or votes after reallocation? So there may be a presentation issue here that we'll take a look at. But we don't believe there's a model issue.
13K
Evy Jo
@Evy_Josephine
Ha, when has Nate ever been correct? I followed his NFL predictions on FiveThirtyEight for years, they were garbage.
622
godstears
@Godstears3
I buy the first-round-lead argument. But how sensitive is the 72% to the transfer/exhaustion assumptions? E.g. if same-party transfers are 90% rather than 80%, holding exhaustion constant, how much does the 72% move? Would love to see a sensitivity table my guess is transfers are
331