How accurate the pollsters were
For every pollster: the average difference between its final poll and the result, how many elections that rests on, and the period those polls span.
Generic Congressional Ballot
| Pollster | Average difference | Elections | Period |
|---|---|---|---|
| AtlasIntel | 0.2 percentage points | 1 too few elections | 2022-11-06 – 2022-11-06 |
| Benenson Strategy Group | 0.2 percentage points | 1 too few elections | 2022-10-29 – 2022-10-29 |
| Fabrizio Lee/Impact Research | 0.3 percentage points | 1 too few elections | 2022-10-24 – 2022-10-24 |
| WPA Intelligence | 0.3 percentage points | 1 too few elections | 2022-10-26 – 2022-10-26 |
| J.L. Partners | 0.3 percentage points | 1 too few elections | 2024-10-29 – 2024-10-29 |
| RKM Research and Communications | 0.4 percentage points | 1 too few elections | 2016-11-03 – 2016-11-03 |
| Data for Progress | 0.6 percentage points | 1 too few elections | 2022-11-05 – 2022-11-05 |
| NewsNation | 0.7 percentage points | 1 too few elections | 2022-10-30 – 2022-10-30 |
| Yankelovich Partners | 0.7 percentage points | 2 too few elections | 1998-10-14 – 2000-10-25 |
| Cygnal | 0.8 percentage points | 2 too few elections | 2022-10-29 – 2024-10-26 |
| Kaplan Strategies | 0.8 percentage points | 1 too few elections | 2024-10-31 – 2024-10-31 |
| Siena College | 0.8 percentage points | 2 too few elections | 2022-10-11 – 2024-10-23 |
| Quinnipiac University | 0.8 percentage points | 1 too few elections | 2022-10-28 – 2022-10-28 |
| Fabrizio Lee/GBAO | 0.8 percentage points | 1 too few elections | 2024-10-22 – 2024-10-22 |
| co/efficient | 0.8 percentage points | 1 too few elections | 2024-10-08 – 2024-10-08 |
| Noble Predictive Insights | 0.8 percentage points | 1 too few elections | 2024-10-27 – 2024-10-27 |
| Pioneer Polling/Tarrance Group | 0.8 percentage points | 1 too few elections | 2022-10-09 – 2022-10-09 |
| Change Research | 0.9 percentage points | 1 too few elections | 2022-11-03 – 2022-11-03 |
| American Viewpoint | 1.1 percentage points | 2 too few elections | 1998-10-11 – 2000-10-20 |
| ABC News/Washington Post Poll | 1.1 percentage points | 8 | 1998-10-31 – 2022-11-01 |
| InsiderAdvantage | 1.2 percentage points | 1 too few elections | 2022-10-25 – 2022-10-25 |
| John Zogby Strategies | 1.2 percentage points | 2 too few elections | 2020-11-01 – 2024-11-03 |
| Echelon Insights | 1.3 percentage points | 3 | 2020-10-18 – 2024-10-31 |
| Florida Atlantic University | 1.3 percentage points | 1 too few elections | 2024-10-27 – 2024-10-27 |
| HarrisX | 1.3 percentage points | 1 too few elections | 2024-11-02 – 2024-11-02 |
| Lucid | 1.4 percentage points | 1 too few elections | 2018-10-24 – 2018-10-24 |
| Selzer & Co. | 1.5 percentage points | 2 too few elections | 2010-10-25 – 2016-11-05 |
| Marist College | 1.5 percentage points | 6 | 2002-10-09 – 2024-11-02 |
| Polling Company/Global Strategy Grp | 1.6 percentage points | 1 too few elections | 2000-11-01 – 2000-11-01 |
| Lake Research/Tarrance Group | 1.6 percentage points | 6 | 2000-11-05 – 2016-10-10 |
| Schoen Cooperman Research | 1.7 percentage points | 1 too few elections | 2010-10-22 – 2010-10-22 |
| Emerson College | 1.7 percentage points | 2 too few elections | 2022-10-18 – 2024-11-02 |
| Fox News Poll | 1.8 percentage points | 4 | 2014-10-26 – 2022-10-11 |
| Suffolk University | 1.8 percentage points | 4 | 2018-10-20 – 2024-10-18 |
| Hart Research/Public Opinion Strategies | 1.9 percentage points | 13 | 1998-10-26 – 2024-11-02 |
| Tufts University | 1.9 percentage points | 1 too few elections | 2022-10-26 – 2022-10-26 |
| Paradigm | 1.9 percentage points | 1 too few elections | 2024-10-22 – 2024-10-22 |
| Los Angeles Times Poll | 2.0 percentage points | 1 too few elections | 2008-10-12 – 2008-10-12 |
| GfK Group | 2.0 percentage points | 4 | 2008-10-18 – 2016-10-22 |
| YouGov | 2.0 percentage points | 9 | 2004-10-19 – 2024-10-31 |
| Gallup | 2.1 percentage points | 8 | 1998-11-02 – 2018-10-21 |
| Rasmussen Reports | 2.1 percentage points | 8 | 2000-11-05 – 2022-11-01 |
| Monmouth University | 2.1 percentage points | 2 too few elections | 2022-10-15 – 2024-10-21 |
| CBS News/New York Times Poll | 2.1 percentage points | 9 | 1998-10-27 – 2016-10-30 |
| SSRS | 2.3 percentage points | 5 | 1998-10-23 – 2024-10-23 |
| Marquette University Law School Poll | 2.3 percentage points | 1 too few elections | 2024-10-10 – 2024-10-10 |
| CBS News Poll | 2.4 percentage points | 7 | 1998-10-31 – 2018-10-16 |
| Pew Research Center | 2.4 percentage points | 7 | 1998-10-30 – 2022-10-13 |
| Hill Research Consultants | 2.5 percentage points | 1 too few elections | 1998-10-15 – 1998-10-15 |
| Morning Consult | 2.5 percentage points | 4 | 2016-10-29 – 2022-11-04 |
| Public Opinion Strategies | 2.5 percentage points | 1 too few elections | 2022-10-09 – 2022-10-09 |
| Premise | 2.6 percentage points | 1 too few elections | 2022-11-01 – 2022-11-01 |
| McLaughlin & Associates | 2.6 percentage points | 2 too few elections | 2018-10-19 – 2022-10-14 |
| Harris Insights | 2.7 percentage points | 5 | 2002-10-23 – 2022-10-12 |
| Big Village | 2.7 percentage points | 7 | 2006-11-04 – 2022-11-03 |
| Wirthlin Worldwide | 2.7 percentage points | 2 too few elections | 1998-10-18 – 2002-10-06 |
| Navigator Research | 2.8 percentage points | 2 too few elections | 2020-10-17 – 2022-10-22 |
| Ipsos | 2.9 percentage points | 8 | 2002-11-02 – 2022-11-06 |
| Zogby Analytics | 2.9 percentage points | 5 | 1998-10-13 – 2010-10-31 |
| USC Dornsife | 2.9 percentage points | 2 too few elections | 2018-10-31 – 2020-10-26 |
| AP-NORC | 3.0 percentage points | 1 too few elections | 2018-10-13 – 2018-10-13 |
| Targoz Market Research | 3.0 percentage points | 1 too few elections | 2024-11-01 – 2024-11-01 |
| Clarity Campaign Labs | 3.1 percentage points | 1 too few elections | 2024-10-23 – 2024-10-23 |
| Opinion Dynamics | 3.2 percentage points | 6 | 1998-10-28 – 2010-10-27 |
| SurveyMonkey | 3.3 percentage points | 1 too few elections | 2016-11-03 – 2016-11-03 |
| Princeton Survey Research Assoc. | 3.3 percentage points | 8 | 1998-10-08 – 2014-10-25 |
| TIPP Insights | 3.6 percentage points | 2 too few elections | 2002-10-10 – 2018-10-29 |
| Abt Associates | 3.8 percentage points | 1 too few elections | 2006-11-02 – 2006-11-02 |
| Public Religion Research Institute | 3.9 percentage points | 2 too few elections | 2016-10-14 – 2018-10-18 |
| American Pulse Research & Polling | 3.9 percentage points | 2 too few elections | 2018-10-08 – 2024-10-30 |
| FTI Consulting | 4.4 percentage points | 2 too few elections | 2006-10-21 – 2008-11-01 |
| Opinium | 4.6 percentage points | 2 too few elections | 2020-10-28 – 2022-10-13 |
| Knowledge Networks | 4.7 percentage points | 2 too few elections | 2004-10-15 – 2006-10-10 |
| GQR/Public Opinion Strategies | 4.8 percentage points | 1 too few elections | 2002-10-16 – 2002-10-16 |
| GBAO Strategies | 5.1 percentage points | 1 too few elections | 2020-10-17 – 2020-10-17 |
| RT Strategies | 9.4 percentage points | 1 too few elections | 2006-10-28 – 2006-10-28 |
President
| Pollster | Average difference | Elections | Period |
|---|---|---|---|
| Swayable | 0.8 percentage points | 1 too few elections | 2020-11-01 – 2020-11-01 |
| TIPP Insights | 1.0 percentage points | 1 too few elections | 2020-11-02 – 2020-11-02 |
| RMG Research | 1.1 percentage points | 1 too few elections | 2020-10-31 – 2020-10-31 |
| Echelon Insights | 1.5 percentage points | 1 too few elections | 2020-10-20 – 2020-10-20 |
| Suffolk University | 1.8 percentage points | 1 too few elections | 2020-10-27 – 2020-10-27 |
| Data for Progress | 1.9 percentage points | 1 too few elections | 2020-10-29 – 2020-10-29 |
| Change Research | 2.1 percentage points | 1 too few elections | 2020-11-01 – 2020-11-01 |
| Research Co. | 2.1 percentage points | 1 too few elections | 2020-11-02 – 2020-11-02 |
| Leger | 2.3 percentage points | 1 too few elections | 2020-11-01 – 2020-11-01 |
| Qriously | 2.5 percentage points | 1 too few elections | 2020-11-01 – 2020-11-01 |
| Redfield & Wilton Strategies | 2.6 percentage points | 1 too few elections | 2020-11-01 – 2020-11-01 |
| Siena College | 2.7 percentage points | 1 too few elections | 2020-10-18 – 2020-10-18 |
| ABC News Poll | 2.8 percentage points | 1 too few elections | 2020-10-09 – 2020-10-09 |
| Pew Research Center | 2.8 percentage points | 1 too few elections | 2020-10-05 – 2020-10-05 |
| John Zogby Strategies | 3.3 percentage points | 1 too few elections | 2020-11-01 – 2020-11-01 |
A pollster with fewer than 3 scored elections is listed here with its figures and marked as such. Its average is shown for completeness, not as a verdict: an average over one or two elections says more about those elections than about the pollster.
Every election we hold a result for
- House election 2024 2024-11-05
- Presidential election 2024 2024-11-05
- House election 2022 2022-11-08
- House election 2020 2020-11-03
- Presidential election 2020 2020-11-03
- House election 2018 2018-11-06
- House election 2016 2016-11-08
- Presidential election 2016 2016-11-08
- House election 2014 2014-11-04
- House election 2012 2012-11-06
- Presidential election 2012 2012-11-06
- House election 2010 2010-11-02
- House election 2008 2008-11-04
- Presidential election 2008 2008-11-04
- House election 2006 2006-11-07
- House election 2004 2004-11-02
- House election 2002 2002-11-05
- House election 2000 2000-11-07
- House election 1998 1998-11-03
How this is measured: for each past election we take each pollster’s last poll in the 30 days before polling day and compare it, party by party, with the result. The figure is the mean absolute difference across the parties that appear in both. Only parties the pollster actually reported are counted. A late shift in opinion is not the same thing as a polling error. For the generic congressional ballot, polls and results are both compared as two-party shares: each party’s share of the combined Democratic and Republican total. Undecided and third-party respondents therefore do not count as polling error, and the result figures in these comparisons are two-party shares too, not the raw popular vote.