✕

PUBLISHED Mar. 25, 2021, at 10:00 AM

FiveThirtyEight’s Pollster Ratings

Based on the historical accuracy and methodology of each polling organization’s polls.

Read more Download the data See the latest polls

CallFire

Pollsters with a relatively small number of polls now receive a provisional rating rather than a precise letter grade.

B/C

Polls analyzed	4
Races called correctly	75%
Most recent cycle	2012
Affiliated with professional polling organization (NCPP/AAPOR/Roper)	No

How accurate are the polls from CallFire?

Polls conducted within 21 days of an election (based on the poll’s median date) are used to determine the pollster’s grade.

2012

Key

A = Adults

RV = Registered voters

V = Voters

LV = Likely voters

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 29-31	891LV	President (IA) General election	IVR	Poll Obama (D) +3.0	Actual Obama (D) +5.8	Difference 2.8
Oct 29-31	839LV	President (CO) General election	IVR	Poll Obama (D) +1.0	Actual Obama (D) +5.4	Difference 4.4
Oct 23-25	909LV	President (NV) General election	IVR	Poll Obama (D) +4.0	Actual Obama (D) +6.7	Difference 2.7
Oct 17-18	624LV	President (VA) General election	IVR	Poll Romney (R) +2.0	Actual Obama (D) +3.9	Difference 5.9

How this works: FiveThirtyEight’s pollster ratings are calculated by analyzing the historical accuracy of each polling organization’s polls along with its methodology. Accuracy scores are adjusted for the type of election polled, the poll’s sample size, the performance of other polls surveying the same race and other factors. We also calculate measures of statistical bias in the polls. When a pollster publishes multiple versions of the same survey (for example, versions with and without third-party candidates included), FiveThirtyEight uses an average of the different versions to calculate the pollster’s rating. However, all versions of these polls are listed here. Read more »

Download this data as an Excel spreadsheet or get it on GitHub.

Design and development by Elena Mejía and Aaron Bycoffe. Research by Mary Radcliffe and Dhrumil Mehta. Statistical model by Nate Silver.

Send us feedback.

Comments