The 2016 Pollyvote Popular Vote Forecast: A Preliminary Analysis

Forthcoming in PS: Political Science and Politics

6 Pages Posted: 14 Dec 2016  

Andreas Graefe

Ludwig Maximilians University of Munich - Department of Communication Science and Media Research

J. Scott Armstrong

University of Pennsylvania - Marketing Department

Randall J. Jones

University of Central Oklahoma

Alfred G. Cuzan

University of West Florida

Date Written: December 13, 2016

Abstract

We assess how the PollyVote and its components performed in this election compared to the previous six (1992 to 2012). While always predicting that Hillary Clinton would win the popular vote, across the 100 days leading to the election on average the PollyVote overshot the mark by 1.9 percentage points, almost twice the MAE incurred in the previous six elections. This was because this year there was very little bracketing among the components. Citizen forecasts and econometric models performed best this year, while the Iowa Electronic Markets came in last. Across all elections from 1992 to 2016, the PollyVote error is only a little over one percentage point.

Keywords: PollyVote, 2016 Presidential Election, Forecast

JEL Classification: Z39

Suggested Citation

Graefe, Andreas and Armstrong, J. Scott and Jones, Randall J. and Cuzan, Alfred G., The 2016 Pollyvote Popular Vote Forecast: A Preliminary Analysis (December 13, 2016). Forthcoming in PS: Political Science and Politics. Available at SSRN: https://ssrn.com/abstract=2884855 or http://dx.doi.org/10.2139/ssrn.2884855

Andreas Graefe

Ludwig Maximilians University of Munich - Department of Communication Science and Media Research ( email )

Geschwister-Scholl-Platz 1
Munich, Bavaria 80539
Germany

HOME PAGE: http://www.andreas-graefe.org

J. Scott Armstrong

University of Pennsylvania - Marketing Department ( email )

700 Jon M. Huntsman Hall
3730 Walnut Street
Philadelphia, PA 19104-6340
United States
215-898-5087 (Phone)
215-898-2534 (Fax)

HOME PAGE: http://marketing.wharton.upenn.edu/people/faculty/armstrong.cfm

Randall J. Jones

University of Central Oklahoma ( email )

100 North University Drive
Edmond, OK 73034
United States

Alfred G. Cuzan (Contact Author)

University of West Florida ( email )

11000 University Parkway
Pensacola, FL 32514-5750
United States

Paper statistics

Downloads
43
Abstract Views
220