UPDATED Mar. 13, 2023, at 3:53 PM

FiveThirtyEight’s Pollster Ratings

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

B/C

Impact Research

Also known as Anzalone Liszt Research, Anzalone Liszt Grove or ALG Research

Polls analyzed	21
Races called correctly	60%
Most recent cycle	2022
Affiliated with AAPOR/Roper	No
Other names	Anzalone Liszt Research Anzalone Liszt Grove ALG Research

What are pollster ratings?

How accurate are the polls from Impact Research?

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

2022

Key

A = Adults

RV = Registered voters

V = Voters

LV = Likely voters

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 21-28	800LV	House (National) General election	Online	Poll Republicans +2.0	Actual Republicans +2.7	Difference 0.7

2020

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 16-21	500LV	House (AR-2) General election	Live Phone	Poll Even	Actual Hill +10.7	Difference 10.7

2018

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 25-28	500LV	House (FL-27) General election	Live Phone	Poll Shalala +10.0	Actual Shalala +6.0	Difference 4.0
Oct 15-21	906LV	Governor (MI) General election	Live Phone	Poll Whitmer +10.0	Actual Whitmer +9.6	Difference 0.4
Oct 15-21	906LV	Governor (MI) General election	Live Phone	Poll Whitmer +11.0	Actual Whitmer +9.6	Difference 1.4

2012

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 24-25	446LV	Senate (IN) General election	Live Phone	Poll Donnelly +7.0	Actual Donnelly +5.8	Difference 1.2
Oct 18-21	400LV	House (IL-13) General election	Live Phone	Poll Gill +9.0	Actual Davis +0.3	Difference 9.3
Oct 18-21	501LV	House (FL-22) General election	Live Phone	Poll Frankel +10.0	Actual Frankel +9.3	Difference 0.7
Oct 17-21	800LV	Senate (AZ) General election	Live Phone	Poll Carmona +4.0	Actual Flake +3.0	Difference 7.0
Oct 16-18	400LV	House (IL-8) General election	Live Phone	Poll Duckworth +15.0	Actual Duckworth +9.5	Difference 5.5

2010

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 27-28	400LV	Senate (IL) General election	Live Phone	Poll Giannoulias +2.0	Actual Kirk +1.6	Difference 3.6
Oct 20-24	600LV	Senate (IL) General election	Live Phone	Poll Giannoulias +2.0	Actual Kirk +1.6	Difference 3.6
Oct 20-21	400LV	House (LA-2) Jungle primary	Live Phone	Poll Richmond +17.0	Actual Richmond +31.1	Difference 14.1
Oct 17-19	600LV	Senate (LA) Jungle primary	Live Phone	Poll Vitter +3.0	Actual Vitter +18.9	Difference 15.9
Oct 13-17	601LV	Senate (IL) General election	Live Phone	Poll Giannoulias +5.0	Actual Kirk +1.6	Difference 6.6
Oct 11-13	400LV	House (AR-1) General election	Live Phone	Poll Causey +2.0	Actual Crawford +8.3	Difference 10.3
Apr 27-29	406LV	House (PA-12) Special election	Live Phone	Poll Critz +2.0	Actual Critz +7.7	Difference 5.7

2008

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 13-15	400LV	President (NE-2) General election	Live Phone	Poll McCain +4.0	Actual Obama +1.2	Difference 5.2
Oct 13-15	400LV	House (NE-2) General election	Live Phone	Poll Terry +1.0	Actual Terry +3.9	Difference 2.9
Apr 3-7	500LV	House (MS-1) Special jungle primary	Live Phone	Poll Davis +2.0	Actual Childers +3.2	Difference 5.2

2006

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 20-23	500LV	House (FL-22) General election	Live Phone	Poll Klein +6.0	Actual Klein +3.8	Difference 2.2

2004

Dates	Sample	Race	Methodology	Poll margin	Election margin	Difference
Oct 20-21	400LV	House (UT-2) General election	Live Phone	Poll Matheson +23.0	Actual Matheson +11.5	Difference 11.5

Show all

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 about our methodology»

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, Cooper Burton and Dhrumil Mehta. Copy editing by Andrew Mangan. Visual editing by Christopher Groskopf. Statistical model by Nate Silver.

Send us feedback.