That 10 point spread for Independents feels right. Trump has done nothing to attract voters who aren't already a part of his base.

+11, +8, +7, +7, and +10 in 5 big polls.

This was the first LV model Quinnipac has done for this election.

Even Joke Rasmussen has Joe +4.

Average is back up to Joe +7.4.

If these polls keep up

Joe was excellent today. It was a breath of fresh air to see the difference between he and Trump holding a press conference. Calm and measured words that are rational and coherent.

It was excellent!

Just give press conferences, keep airing commercials in the right places, going all out, and let people see the difference between a calm, measured, kind and competent person vs the insaneasylum escapee.

Swing independents will decide this election just like they decide every election. That independent category is all I care about. Trump won independents by 4 points in 2016 but they quickly shifted away to him in early 2017 and have remained at basically this same 10 point level toward Democrats subsequently. The House national vote in 2018 was independents 54-42 toward Democrats.

This poll has the sensible partisan numbers of Biden 93-6 among Democrats and Trump 90-8 among Republicans. As I've emphasized, don't believe any crap poll that has Trump receiving 20% of blacks or Biden receiving 20% of Republicans. It is not happening. The crossover anecdotes are merely that, trivial nonsense. It happens every cycle yet somehow the gullible suckers always find some rationale that it's going to be valid this time.

The 90/10 categories are difficult for polling to capture especially African-Americans at 12% of the population. Small number of outliers really skew the sample. Likewise there is always absurd hype every cycle that the Republican will win 15-25% of the black vote. Just ignore it and assign 90/10.

Do they use 2016 voting patterns? Because our turnout sucked. If that is the case I would guess there is another point or 3 for Biden since I fully expect us to have 2008 turnout levels.

Thanks in advance for the help.

will give the survey-taker a turnout score based on how many of the last n elections one voted. Others that use random-digit-dial might ask them directly how likely they are to vote and apply a threshold for likely voters.

That is the major change from 2016. The likely voter models in state polls underestimated the percentage of voters who never attended college. Those voters showed up in atypical percentage and voted for Trump.

National polls are better funded and more sophisticated and devote more time than state polls. That's why state polls are more prone to be all over the place and have strange errors.

Democratic turnout was not bad everywhere in 2016. For example, Hillary received almost 300,000 more votes in Florida than Obama did while winning Florida in 2012. But Trump managed nearly 500,000 more votes in Florida than Romney 2012. That's why the Florida election expert Steve Schale was so stunned on election night 2016. He was expecting Trump to receive 2012-like numbers then it was well beyond that in 47 different counties.

We can expect the same trump turn out this year.

I still think we win Florida.

That was the first indication of what was unfolding.

I think we win Florida this time because we have the correct candidate to pick off at the margins everywhere.

Graham would have won decisively.

Like in 08 and 12. We get out the vote we win.