Convention bump model
The convention-bump model, developed by Jim Campbell, predicts the two-party popular vote based on the state of the economy and the candidates’ standing in polls before and after the conventions.
Variable description
| Table 1: Overview of variables used in the convention-bump model | ||
| Variable | Description | Value |
|---|---|---|
PRE-CONVENTION POLL |
Incumbent party’s candidate two-party support in polls before the parties’ first convention | 52.2 |
NET CONVENTION BUMP |
Change in incumbent party’s candidate two-party support polls from before the first convention to after the second convention | 0.4 |
GDP |
Annualized real GDP growth rate in the second quarter (April–June) of 2016 as indicated by the Bureau of Economic Analysis’ second estimate released at the end of August | 1.1 |
A |
Constant | 51.2 |
| V | Incumbent share of the two-party presidential vote | |
Vote equation
The regression model’s vote equation reads as:
V = A + b1 PRE-CONVENTION POLL + b2 NET CONVENTION BUMP + b3 GDP
2016 forecast
Jim Campbell presented the 2016 forecast of the convention bump model at the APSA meeting in Philadelphia. The model predicts Clinton to gain 51.2% of the two-party vote, compared to 48.8 for Trump.
Past performance
On average across the three elections since 2004, the trial-heat model missed the final results by 2.7 percentage points.
References
Campbell, J. (2012). Forecasting the Presidential and Congressional Elections of 2012: The Trial-Heat and the Seats-in-Trouble Models. PS: Political Science & Politics, 45(4), 630-634.