American Research Group, Inc.

 

2016 US Presidential Election Pollster Ratings

National Popular Vote

 

The following table lists the accuracy of pollsters in the November 8, 2016 US Presidential election using the measure of polling accuracy proposed by Martin, Traugott, and Kennedy.

An accuracy measure of 0.00 indicates that the odds ratio of the survey results matches the odds ratio of the actual results (the closer to 0.00, the better).

 

2016 Presidential Election - National Popular Vote
Clinton Trump Odds
Ratio
Accuracy
McClatchy/Marist 46% 44% 1.00 0.00 
Politico/Morning Consult 51% 48% 1.02 0.02 
ABC/Post 49% 46% 1.02 0.02 
UPI/CVOTER 49% 46% 1.02 0.02 
IBD/TIPP 43% 42% 0.98 -0.02 
Bloomberg/Selzer 46% 43% 1.02 0.02 
YouGov 49% 45% 1.04 0.04 
Fox 48% 44% 1.04 0.04 
CBS/Times 47% 43% 1.05 0.04 
Gravis/OANN 50% 50% 0.96 -0.04 
Ipsos/Reuters 44% 39% 1.08 0.08 
Monmouth 50% 44% 1.09 0.08 
NBC/SurveyMonkey 51% 44% 1.11 0.10*
USC/LA Times 43.6% 46.8% 0.89 -0.12*
Actual 48.2% 46.1%

 

*The accuracy measures for the NBC/SurveyMonkey poll (sample size of 70,194) and the USC/LA Times poll (sample size of 2,972) fall outside the 95% confidence intervals.

The accuracy measures for the other 12 polls fall within their respective 95% confidence intervals.

The absolute average accuracy for all 14 polls in 2016 is 0.046.

The absolute average accuracy for the 12 polls with accuracy measures that fall within the 95% confidence intervals is 0.036.

The absolute average accuracy for all polls in the 2012 presidential election was 0.055. The average absolute accuracy for the 22 polls in 2012 falling within their respective 95% confidence intervals was 0.052.

The absolute average accuracy for all polls in the 2008 presidential election was 0.037. The accuracy measures for all polls in 2008 fell within their respective 95% confidence intervals.

ARG Home