Your Daily Polling Update for Tuesday, November 17, 2020
WHAT HAPPENED?
A hard look at state polls in the 2020 presidential election
Part 2 in a Series
by Ron Faucheux
Now that most votes are counted, we’re taking a hard look at the polls in 10 key states in the presidential election. This is what we found:
- Final poll averages indicated the right winner in seven of the 10 states analyzed: Nevada, Arizona, Pennsylvania, Michigan, Wisconsin, Iowa and Ohio. (Biden won five of these seven states and Trump won two.)
- Final poll averages missed the winner in three of the 10 states: Florida, North Carolina and Georgia. But––Georgia came very close in terms of actual points and North Carolina was not that far off. (Biden won one of these three states and Trump won two.)
- Final poll averages came within 1 point of the actual vote in two states: Nevada and Georgia. (Biden won both states).
- Final poll averages came within 1-2 points of the actual vote in two states: Arizona and Pennsylvania. (Biden won both states).
- Final poll averages came within 2-3 points of the actual vote in two states: North Carolina and Michigan. (Biden won one state and Trump won one).
- Final poll averages were 5-7 points off of the actual vote in four states: Wisconsin, Florida, Ohio and Iowa. (Biden won one state and Trump won three).
- Most embarrassing misses by pollsters:
- Trafalgar polls had Trump leading in five key states that Biden carried (Nevada, Georgia, Arizona, Pennsylvania and Michigan).
- PPP, Morning Consult, Emerson College, and Quinnipiac Univ. polls showed Biden ahead by 5- to 7-point margins in Florida, a state he lost by 3.4 points.
- Quinnipiac showed Biden ahead by a 4-point margin in Ohio, a state he lost by 8.2 points.
- PPP and Civiqs showed Biden ahead by a 1-point margin in Iowa, a state he lost by 8.2 points.
- Most overstated Biden poll margins in a state he carried:
- Eight of the 12 final polls in Wisconsin showed Biden with wide 8- to 13-point margins in a state he won by less than 1 point: Morning Consult (Biden +13), NYT/Siena (Biden +11), Survey Monkey (Biden +10), Reuters/Ipsos (Biden +10), CNN (Biden +8), Emerson College (Biden +8), Research Co. (Biden +8) and CNBC/Change Research (Biden +8)
- Polls that came closest to actual vote counts in each of the 10 key states:
- Trafalgar came closest in Wisconsin and Ohio; Washington Post/ABC and Trafalgar came closest in Florida; Emerson College came closest in Nevada; Data Orbital came closest in Arizona; Des Moines Register came closest in Iowa; PPP and Morning Consult came closest in Georgia; AYTM/Aspiration came closest in Pennsylvania; Insider Advantage came closest in Michigan; and Rasmussen came closest in North Carolina.
Polling averages vs. actual voting results in the 10 key states analyzed:
GEORGIA
- Poll average showed wrong winner, but was only five-tenths of a point off actual result
- Winner of actual vote: Biden +0.3
- Final poll average: Trump +0.2
- 4 polls had Biden ahead (by 2 to 4 points)
- 5 polls had Trump ahead (by 1 to 5 points)
- Polls showing right winner closest to actual result: PPP (Biden +2) and Morning Consult (Biden +2)
- Polls showing wrong winner by most overstated margins: Trafalgar (Trump +5) and WSB/Landmark (Trump +4)
NEVADA
- Poll average showed right winner and was off nine-tenths of a point from actual vote
- Winner of actual vote: Biden +2.4
- Final poll average: Biden +3.3
- 3 polls had Biden ahead (by 2 and 6 points)
- 1 poll had Trump ahead (by 1 point)
- Poll showing right winner closest to actual result: Emerson College (Biden +2)
- Poll showing wrong winner: Trafalgar (Trump +1)
ARIZONA
- Poll average showed right winner and was off only 1.5 points from actual vote
- Winner of actual vote: Biden +0.3
- Final poll average: Biden +1.8
- 9 polls had Biden ahead (by 1 to 6 points)
- 2 polls had Trump ahead (by 3 to 4 points)
- 1 poll had a tie
- Poll showing right winner closest to actual result: Data Orbital (Biden +1)
- Polls showing wrong winner: Rasmussen (Trump +4), Trafalgar (Trump +3)
PENNSYLVANIA
- Poll average showed right winner and was off 1.8 points from actual vote
- Winner of actual vote: Biden +1.1
- Final poll average: Biden +2.9
- 7 polls had Biden ahead (by 2 to 9 points)
- 3 polls had Trump ahead (by 1 to 2 points)
- Poll showing right winner closest to actual result: AYTM/Aspiration (Biden +2)
- Polls showing wrong winner: Trafalgar (Trump +2), Insider Advantage (Trump +2), Susquehanna (Trump +1)
- Poll showing right winner by most overstated margin: Morning Consult (Biden +9)
NORTH CAROLINA
- Poll average showed wrong winner and was off 2.4 points from actual vote
- Winner of actual vote: Trump +1.3
- Final poll average: Biden +1.1
- 10 polls had Biden ahead (by 1 to 2 points)
- 4 polls had Trump ahead (by 1 to 4 points)
- 1 poll had it a tie
- Poll showing right winner closest to actual result: Rasmussen (Trump +1)
- Polls showing wrong winner by most overstated margins: CNN (Biden +6), NBC/Marist (Biden +6)
MICHIGAN
- Poll average showed right winner, but was off 2.7 points from actual vote
- Winner of actual vote: Biden +2.6
- Final poll average: Biden +5.3
- 7 polls had Biden ahead (by 2 to 8 points)
- 1 poll had Trump ahead (by 2 points)
- Poll showing right winner closest to actual result: Insider Advantage (Biden +2)
- Polls showing right winner but with most exaggerated margins: Ipsos (Biden +8), Emerson College (Biden +7), Morning Consult (Biden +7), Mitchell Research (Biden +7) and CNBC/Change Res. (Biden +7)
- Poll showing wrong winner: Trafalgar (Trump +2)
FLORIDA
- Poll average showed wrong winner and was off 5.6 points from actual vote
- Winner of actual vote: Trump +3.4
- Final poll average: Biden +2.2
- 12 polls had Biden ahead (by 1 to 7 points)
- 4 polls had Trump ahead (by 1 to 2 points)
- 1 poll had it even
- Polls showing right winner closest to actual result: Washington Post/ABC (Trump +2), Trafalgar (Trump +2)
- Polls showing wrong winner by most overstated margins: PPP (Biden +7), Morning Consult (Biden +6), Emerson College (Biden +6), Quinnipiac (Biden +5)
OHIO
- Poll average showed right winner, but was off 6.6 points from actual vote
- Winner of actual vote: Trump +8.2
- Final poll average: Trump +1.6
- 6 polls had Trump ahead (by 1 to 5 points)
- 2 polls had Biden ahead (by 1 to 4 points)
- Poll showing right winner closest to actual result: Trafalgar (Trump +5)
- Poll showing wrong winner by most overstated margin: Quinnipiac (Biden +4)
IOWA
- Poll average showed right winner, but was off 6.7 points from actual vote
- Winner of actual vote: Trump +8.2
- Final poll average: Trump +1.5
- 4 polls had Trump ahead (by 1 to 5 points)
- 2 polls had Biden ahead (by 1 point)
- Poll showing right winner closest to actual result: Des Moines Register (Trump +7)
- Polls showing wrong winner: PPP, Civiqs (Biden +1)
WISCONSIN
- Poll average showed right winner, but was off 6.8 points from actual vote
- Winner of actual vote: Biden +0.6
- Final poll average: Biden +7.4
- All 12 polls had Biden ahead (by 1 to 13 points)
- Poll showing right winner closest to actual result: Trafalgar (Biden +1)
- Eight of 12 final polls showed the right winner but with substantially overstated margins: Morning Consult (Biden +13),NYT/Siena (Biden +11), Survey Monkey (Biden +10), Reuters/Ipsos (Biden +10), CNN (Biden +8), Emerson College (Biden +8), Research Co. (Biden +8) and CNBC/Change Research (Biden +8)
Even though many state polls came close to the actual vote, why were others so far off track?
Here are possible reasons:
- The normal “margin of error.” Possible error in most polls is usually plus-or-minus 3 points. While this may explain why polls missed the exact mark, it doesn’t explain why errors were often in one partisan direction––showing a bigger Democratic lead than actually materialized.
- Some pollsters screwed up. Administration of some polls may have been sloppy, questionnaires may have been poorly written and sample targets and weighting may have been biased. While this doesn’t apply to every––or even most––polls, it may explain why some of the results were off beam.
- Donald Trump tends to close strong. It’s highly possible that he made gains in the final hours of the campaign that were too late for polls to catch. Much of the interviewing for these state polls were conducted between Oct. 28-Nov. 1; if there was a late-breaking Trump vote, many of these polls would not have detected all or most of it. Polls released the day before the November 3rd election did, in fact, show that Biden’s margins in Pennsylvania, Michigan and Nevada were declining. Remember: polls are snapshots, not crystal balls.
- There is a “hidden” Trump vote. This is the easiest and maybe the best explanation, something we warned about in this newsletter before the election. Voters in some places may not have wanted to admit they were voting for Trump for fear of social ostracism or professional retribution, especially in such a bitterly polarized election…. Most pollsters didn’t find this hidden vote because they didn’t believe it existed; of those who thought it may have existed, many didn’t know how to measure it…. A hidden Trump vote of 2 to 4 percent would explain the gap between his polling strength and the actual vote in a number of states; but this clearly needs further study before conclusions are reached.
NOTE ON THE SERIES: The purpose of this series is neither to indict nor to vindicate the polling industry. We simply want to present the facts and explain, to the best of our ability with the data we have, what may have happened and why.
- The first part of this series (last week) focused on national polling.
- The second part of the series(today) focuses on state polls in the presidential election.
- The third part will focus on Senate and gubernatorial polls.
PERSONAL NOTE: Neither I nor my firm, Certus Insights, conducted any of the public polls from this year’s presidential, Senate or gubernatorial races that we are analyzing.
Subscription note: If you ever stop receiving Lunchtime Politics, check your spam or junk folders. Sometimes, for mysterious reasons, the newsletter may re-route to there. Please know that we will never take you off our subscriber list unless you ask us to do so.
Follow here to get the latest updates on the Lunchtime Politics.
Be the first to comment