The Presidential election forecasting models I’ve been following this election cycle are all pointing toward a Clinton victory. Now we have to wait and see.
Election Analytics @ Illinois
Princeton Election Consortium (Sam Wang)
FiveThirtyEight (Nate Silver)
New York Times Upshot forecast
Daily Kos (Drew Linzer)
David Rothschild’s prediction market forecasting model
Huffington Post Election Forecast
Sabato’s Crystal Ball
13 Keys to the White House
- Trump to win the popular vote, no electoral college prediction
- This model does not use state-level information or the electoral college (see my blog post here). It appears to be a better mid-term or long-term forecasting model.
All models are wrong, some are useful. "13 Keys to the White House" seems useful months (not days) before an election #ElectionFinalThoughts—
Laura Albert McLay (@lauramclay) November 07, 2016
Why don’t all of these models agree? A few articles I’ve read lately about forecasting models and polling:
- Andrew Gelman: Different election forecasts not so different.
- Don’t be fooled by Trump’s bounce. “What seems like a big change in public opinion turns out to be little more than changes in people’s inclination to respond to polls“
- FiveThirtyEight: Why Our Model Is More Bullish Than Others On Trump. Covariances are crucial.
- Is 99% a reasonable win probability? Sam Wang: Yes.
- Me: “Your model is only as good as the inputs you give it”
- Nate Silver is unskewing polls — all of them — in Trump’s direction
- Which forecasts are true? Are win probabilities even defensible?