They are inside the margin of error in all the battleground states...and they were horribly wrong last time.
They knew 20 years ago going in that polling had to be done differently but they don't really have another way of doing it.
I thought the polls were fairly accurate in 2016. It was what the pundits on tv made of them that misled people. The pundits on tv focused on the national polls, and those don't matter that much.
It is a constant struggle for pollsters to adapt to changes in communication. Land line versus cell phones caused problems.
Apparently republicans don't answer the phone and answer polls.
So the Republicans are saying. I remember when PPP was messing things up big time. They were estimating how many Democrats and how many Republicans should be answering their phones. They based their estimate on voter registration, I think. If they got a different number of Democrats and Republicans, they tinkered with the figures. For example, if they thought they got 10% fewer Democrats, they'd multiply those results by 1.1 trying to correct. If they thought too many Republicans were answering, they'd multiply the results by 0.9.
At any rate, that year their polls were horribly different from other polling companies; and they consistently had the Democrats doing better than the other polls. Along came the election, and PPP was crushed. It was connected with the Democratic Party; but at that point, even the Democrats wanted nothing to do with it. Why pay them money to get such wrong results?
When they revised how they did things, their results got much better. They became one of the more reliable polling companies.
Rasmussen has consistently over the years tilted their polls to Republicans. It's not as bad as PPP's tilting to Democrats; but it's still there.
The markup on every poll was getting more and more every year for registered conservatives.
And in 2016 they practically stopped answering the phone altogether.
That's why the poll margins have increased so much...the more guestimates the higher the error factor.
From 3% to 4% and now 5%.
I saw someone from Trafalgar was trying to explain their methodology. They start off with an assumption of a "silent majority" which doesn't want to talk about how they vote. They aren't really polling people -- they're assuming the people are out there and that they can predict how they'll vote. Their results this year are very different from the other polls.
I think they got lucky in 2016 when they were using this same method of assuming there was a silent majority. I say that because Trafalgar accurately predicted Trump would win in Pennsylvania and Florida. I don't see it. I give credit to Kellyanne Conway who headed up the Trump campaign for making sure his voters were excited by the election and went out to vote. The Clinton campaign was taking things for granted; they didn't work too much on getting out the votes. Clinton herself didn't bother to campaign that much in states she should have paid more attention to. I think the Democrats also took the black vote for granted. Many blacks had registered so they could vote for Obama; and turn out among blacks was high because they were excited to be voting for a black candidate. A lot of blacks weren't that enthusiastic in 2016; and they figured Clinton would win without their vote anyway.
I doubt many polling companies will go back to going door to door or even to having real people on the phones. They resort to robocalls. I tend to hang up on robocalls. I don't know if older or younger people are more apt to hang up on robocalls.
I think Trafalgar will wind up with egg on its face since I think voter enthusiasm is the more important factor. Biden may be ahead in the polls, and more Democrats have already voted by mail; but the Biden campaign knows they have to work to get out the vote. His supporters are not as wound up as Trump's. That could affect the outcome in a few states; but it won't be enough to get Trump to 270 electoral votes.
Notice how Republicans are trying to make it harder to vote in some places. That could backfire since Democrats can say Republicans are trying to suppress the vote, making Democrats want to vote even more. Republicans can talk about their having a silent majority; but if they did, why would they be trying to make it harder for some people to vote?
But we will see.