Models -> Mixed

DeSart’s long-range model

  • Aims at forecasting U.S. presidential election outcomes more than a year ahead
  • Uses factors such as state electoral histories, national polling data, and contextual variables

The model


Jay DeSart’s long-range presidential election forecast model aims at predicting U.S. presidential election outcomes more than a year in advance. The model integrates four variables:

  • Previous result: State i’s result from the previous election.
  • Prior national polls: Average of all national head-to-head matchup polls taken in month X before the election.
  • Home state: 1 if state i is the home state of the Democratic candidate, -1 if it’s the home state of the Republican candidate, and 0 otherwise.
  • Regime age: The number of terms the current party in the White House has occupied.

The model generates forecasts of state-level outcomes using the following vote equation:

Vi = A + b1 Previous resulti + b2 Prior national polls + b3 Home statei + b4 Regime age

The forecast of the national popular vote is then calculated by weighting each state by its proportion of the total number of votes cast in the previous election.