|
 |
 |
|
MINNESOTA |
| |
|
|
 |
POLLSTERS: Biden +3.9 - Biden 46.7, Trump 42.8
REF'S AVG: Biden +3.9, Biden 46.7, Trump 42.8 |
| RETURN TO ALL BATTLEGROUND STATES |
| 2016 Exit Poll - 37D/28I/35R without leaners (PDF) |
| 2018 Exit Poll - 46D/14I/40R with leaners (PDF) |
| Pollsters' Averages Chart |
Political Ref's Averages Chart |
|
|
|
| POLL RESULT SUMMARY (SEE FULL RESULTS & EXPLANATIONS BELOW) |
| Poll |
Date |
Sample + Party Distribution |
Result |
Pollster Avg |
Ref's Avg |
| *Polls well outside of the consensus margin of error, such as a poll showing a 16-point lead for either candidate when the average of all others show a 4-point lead, will be excluded to preserve the legitimacy of the average. |
| Survey USA |
10/23-10/27 |
649LV, Not provided |
Biden 47, Trump 42 |
x |
|
| Political Ref's Take on the above poll |
From poll: Not Insurmountable Leads; 59 and Sunny Forecast for Election Day, Which GOP Will Need if Trump and Lewis Want Drama When Gopher State Votes Are Counted: |
Included |
|
x |
| Democracy Institute |
9/30-10/2 |
450LV, 46D/15I/39R |
Biden 44, Trump 46 |
x |
|
| Political Ref's Take on the above poll |
The partisan distribution oversamples Dems somewhat, but is reasonable. No adjustment necessary. |
Included |
|
x |
| Mason-Dixon |
9/21-9/23 |
800LV, 38D/28I/34R |
Biden 48, Trump 42 |
x |
|
| Political Ref's Take on the above poll |
The partisan distribution makes sense. No adjustment necessary. |
Included |
|
x |
| Suffolk Univ |
9/20-9/24 |
500LV, 40D/21I/36R |
Biden 46.6, Trump 40.2 |
x |
|
| Political Ref's Take on the above poll |
The partisan distribution makes sense. No adjustment necessary. |
Included |
|
x |
| Morning Consult |
9/4-9/13 |
643LV, Party distribution not provided |
Biden 48, Trump 44 |
x |
|
| Political Ref's Take on the above poll |
Biden won D's 93-5, I's split 40-39 and Trump won R's 90-7. Applying these voter ratios to the exit poll from 2016, it seems Morning Consult used the 2016 voter turnout. No adjustment necessary. |
Included |
|
x |
| FULL POLL RESULTS AND EXPLANATIONS OF ADJUSTMENTS |
| Poll |
Date |
Sample + Party Distribution |
Result |
Pollster Avg |
Ref's Avg |
| Survey USA |
10/23-10/27 |
649LV, Not provided |
Biden 47, Trump 42 |
x |
|
| Political Ref's Take on the above poll |
From poll: Not Insurmountable Leads; 59 and Sunny Forecast for Election Day, Which GOP Will Need if Trump and Lewis Want Drama When Gopher State Votes Are Counted: |
Included |
|
x |
| Survey USA |
10/1-10/6 |
929LV, Party distribution not provided |
Biden 47, Trump 40 |
x |
|
| Political Ref's Take on the above poll |
Neither party distribution or voter ratios are provided. |
Excluded |
|
|
| Democracy Institute |
9/30-10/2 |
450LV, 46D/15I/39R |
Biden 44, Trump 46 |
x |
|
| Political Ref's Take on the above poll |
The partisan distribution oversamples Dems somewhat, but is reasonable. No adjustment necessary. |
Included |
|
x |
| Morning Consult |
9/4-9/13 |
643LV, Party distribution not provided |
Biden 48, Trump 44 |
x |
|
| Political Ref's Take on the above poll |
Biden won D's 93-5, I's split 40-39 and Trump won R's 90-7. Applying these voter ratios to the exit poll from 2016, it seems Morning Consult used the 2016 voter turnout. No adjustment necessary. |
Included |
|
x |
| YouGov |
9/9-9/11 |
1087LV, 36D/28I/34R |
Biden 50, Trump |
x |
|
| Political Ref's Take on the above poll |
Party distribution seems reasonable. No adjustment necessary. |
Included |
|
x |
| New York Times/Sienna |
9/8-9/10 |
814LV, 33D/33I/30R |
Biden 50, Trump 41 |
x |
|
| Political Ref's Take on the above poll |
D's and R's were undersampled and I's oversampled. Will adjust to the 37D/28I/35R. According to the poll data, Biden won D's 93-4, I's 52-32 and Trump won R's 90-5. Applying the voter ratios to the more likely turnout, Biden receives 50.7 to Trump 41.9. |
Biden 50.7, Trump 41.9 |
|
x |
| SurveyUSA |
9/4-9/7 |
553LV, Party distribution not provided |
Biden 49, Trump 40 |
x |
|
| Political Ref's Take on the above poll |
The party distribution is not provided but voter ratios are. Biden won D's 94-2, I's 47-30 and Trump won R's 90-5. After applying the voter ratios, it is evident the party distribution was reasonable. No adjustment necessary. |
Included |
|
x |
| Morning Consult |
8/29-9/7 |
649LV, Party distribution not provided |
Biden 49, Trump 44 |
x |
|
| Political Ref's Take on the above poll |
Neither party distribution or voter ratios are provided. |
Excluded |
|
|
| PPP |
9/3-9/4 |
877RV, 39D/29I/32R |
Biden 52, Trump 44 |
Registered voter polls excluded after conventions |
| Political Ref's Take on the above poll |
The D+7 sample considerably oversampled Democrats. Adjusting to the much more likely D+2, or 37D/28I/35R. I will also apply a registered voter filter to the I column. Voter ratios are not provided but can be inferred. Biden had a +7 partisan sample and an 8-point lead, suggesting he won all registered Independents by a point, or 3 percentage points. Assuming Biden won I's in the registered voter sample 48-45. Adjusted the I column shows Biden at 31 and Trump at 32.3. Assuming each candidate won their parties 90-5, Biden receives 43.7 to Trump 42.4. |
Biden 43.7, Trump 42.4 |
|
x |
| Harper Polling |
8/30-9/1 |
501LV, Party distribution not provided |
Biden 48, Trump 45 |
x |
|
| Political Ref's Take on the above poll |
Party distribution is not provided but voter ratios within those parties can be inferred. Independents in the state split 40-40, which is the trickiest number to estimate. Assuming each candidate won his party 90-5, and voter turnout is 37D/28I/35R, Biden receives 46.3 to Trump 44.6. |
Biden 46.3, Trump 44.6 |
|
x |
| Morning Consult |
8/21-8/30 |
647"LV," Party distribution not provided |
Biden 50, Trump 43 |
Repeat |
|
| Political Ref's Take on the above poll |
Neither party distribution or voter ratios within those parties was provided. |
Excluded |
|
|
| Democracy Institute |
8/26-8/28 |
450LV, 46D/15I/39R |
Biden 45, Trump 48 |
x |
|
| Political Ref's Take on the above poll |
The voter turnout assumption almost mirrors the 2018 turnout exactly. It also seems to favor Dems more than the 2016 turnout. No adjustment necessary. |
Included |
|
x |
| Emerson College |
8/8-8/10 |
733LV, 38D/27I/36R |
Biden 48, Trump 45 |
x |
|
| Political Ref's Take on the above poll |
Party distribution looks reasonable. No adjustment necessary. |
Included |
|
x |
| PPP |
7/22-7/23 |
1218RV, 36D/34I/30R |
Biden 52, Trump 42 |
x |
|
| Political Ref's Take on the above poll |
The D+6 sample oversamples Democrats. Will adjust to the more likely D+2 from 2016, 37D/28I/35R. Voter ratios are not provided by can be inferred. Biden had a 6-point turnout advantage and a 10-point lead. That suggests he did better with Independents and Republicans than Trump did with Independents and Democrats. For the sake of simplification, I will assume the difference came from Independents, or that Biden won Independents 17-13 with 4% choosing other candidates, or 50%-38%. Will assume that both candidates won their parties 90-5. Applying these ratios to the more likely voter turnout, Biden receives 49.1 to Trump 44.0. |
Biden 49.1, Trump 44.0 |
|
x |
| Morning Consult |
7/17-7/26 |
662LV, Party distribution not provided |
Biden 47, Trump 44 |
x |
|
| Political Ref's Take on the above poll |
Neither party distribution or voter ratios within those parties are provided. |
Excluded |
|
|
| Trafalgar Group |
7/23-7/25 |
1129LV |
Biden 49, Trump 44 |
x |
|
| Political Ref's Take on the above poll |
Neither party distribution or voter ratios within those parties was provided. |
Excluded |
|
|
| Fox News |
7/18-7/20 |
776RV, 47D/14I/39R |
Biden 51, Trump 38 |
>2weeks |
|
| Political Ref's Take on the above poll |
This is a voter turnout that pushed leaners to pick a party. We can see from the 2016 and 2018 exit polls that when leaners are pushed, the state looks more Democratic. The 2018 showed D+6 but it was a non-presidential year when fewer Democrats tend to vote. For that reason a D+8 seems reasonable. No adjustment necessary. It is important to note, however, that this is a registered voter poll which almost always pad the Democrat's numbers somewhat. The margin is likely closer than this poll reflects. |
Included |
|
>2weeks |
| Gravis Marketing |
6/19 |
600RV, 38D/31I/31R |
Biden 54, Trump 37 |
>2weeks |
|
| Political Ref's Take on the above poll |
This poll oversamples Democrats at D+7. Will adjust to the much more likely 37D/28I/35R voter turnout model from 2016. The Biden/Trump vote ratios are not provided but can be inferred. Biden had a 7 point advantage in the voter turnout model (D+7) but had a 17 point lead. This means he won Independents and did well with Republicans. For purposes of analysis, I will assume each candidate won their party 90-5, although I know it's not true. I will give Biden a bigger win with Independents than he likely had to adjust for this, but in the end the inference will be sound for adjusting the result. Assuming Biden won Independents by 20.5-10.5. or 66% to 34%, and Biden and Trump each won their parties 90-5, the adjusted result is Biden 53.5 and Trump 42.3. It is important to remember that this is a registered voter poll. Democrat leaning registered votes fail to show up at a higher rate than Republican leaning registered voters. Because this is true, the margin is likely closer than even this adjusted result. |
Biden 53.5, Trump 42.3 |
|
>2weeks |
| Mason-Dixon |
5/18-5/20 |
800RV, 38D/29I/33R |
Biden 49, Trump 44 |
>2weeks |
|
| Political Ref's Take on the above poll |
The D+5 voter turnout model is too high. Adjusting to the more likely 2016 exit polls where party distribution was 37D/28I/35R. According to the poll data, Biden won D's 95-2, I's were split at 41 Biden to 42 Trump, and Trump won R's 95-2. Applying these voter ratios to the more likely 2016 voter turnout, Biden receives 47.3 to Trump 45.8. It is important to remember that this is a registered voter poll. Democrat leaning registered votes fail to show up at a higher rate than Republican leaning registered voters. Because this is true, the margin is likely closer than even this adjusted result. |
Biden 47.3, Trump 45.8 |
|
>2weeks |
