How close did the polls get?
Compare what the polls measured with how people voted. Explore the size of the miss, its direction, and the history behind some of the most consequential polling surprises.
The same question. Six different records.
Choose a category to explore its trend, directional error, sample coverage, and distance from Election Day.
President · national
2000–2020 · National popular vote
President · state
2000–2020 · Only polled states
U.S. Senate
1998–2022 · Changing states and race coverage
House · individual districts
1998–2022 · Individual races, not all districts
House · national generic ballot
1998–2022 · National vote, not seat counts
Governor
1998–2022 · Includes odd-year elections
2024 improved on 2020. Error remained.
A separate, published AAPOR review supplies a common evaluation window across offices. Its values are not appended to the archive lines.
2024: the same final-fieldwork window
AAPOR’s weighted analysis · October 23–November 5 · raw D–R margin error
House district polls cover only 20 contests; this is not an estimate for all 435 districts. Presidential congressional-district polls concern electoral votes and are distinct from House races.
Landmark upsets, examined closely.
Some polls picked the wrong leader. Others picked the winner but missed the scale of the result. Both matter.
1936: millions of returns, the wrong electorate
A huge voluntary return cannot guarantee a representative sample.
Read the evidence and analysis →1948: the Dewey lead that disappeared
The sampling and timing of a poll matter more than confidence in its winner call.
Read the evidence and analysis →2016: small state margins, consequential polling misses
A reasonably close national popular-vote estimate can coexist with consequential errors in the states that decide the presidency.
Read the evidence and analysis →Iowa 2024: a striking miss, not a settled explanation
One prominent outlier can dominate the election story. Evaluate it against the result and other polls without pretending that its cause has been established.
Read the evidence and analysis →Explore every case study and the archive’s largest measured misses →
A close race magnifies a small miss.
A national poll measures popular support, not Electoral College votes. A House generic ballot measures national party preference, not a particular district or a seat forecast. A small numerical miss can change the apparent leader in a tight race.
Accuracy also means more than “called the winner.” A poll can identify the winner and still badly miss the margin. Conversely, a close wrong-leader estimate can be numerically closer than a comfortable correct call.
Hypothetical result = polled margin − signed error.