Princeton Election Consortium

Innovations in democracy since 2004

Sep 22: Biden 350 EV (D+5.1% from toss-up), Senate 52 D, 48 R (D+3.7%), House control D+3.5%
Moneyball states: President NV AZ PA, Senate MT KS AK, Legislatures KS TX NC

U.S. House 2020 – November forecast

July 19th, 2020, 11:34am by Sam Wang


We’ve combined the generic-Congressional ballot with special-election data to generate a November forecast.

The black trace shows the day-to-day snapshot using generic-Congressional surveys. Data come from FiveThirtyEight filtered using our own median rule. We generated a prior by using special elections since 2018. We combined the two by assuming (a) random drift from the snapshot, and (b) convolving it with the prior. The result is a “hurricane strike zone” with a red zone (one-sigma, about two-thirds of outcomes) and a yellow zone (two sigma, about 95% of outcomes).

Because this calculation doesn’t use specific district ratings, those ratings (for example, see The Center for Politics) provide a quasi-independent approach (though of course prognosticators do use national opinion as an input). In the case of the House Meta-Margin, the conversion factor is typically about 6 seats/%, i.e. D+7% would map approximately to a 42-seat margin, or 239-197.

For a deeper dive, see thisĀ explainer page. If you can read Matlab, see this script. Thanks to Mike Hallee for assistance!

Tags: 2020 Election · House

No Comments so far ↓

There are no comments yet...Kick things off by filling out the form below.

Leave a Comment