2020 Poll Hype Thread

[redjohn]

New WSB-TV/Landmark polls of GA (presidential, Senate races) coming in about an hour:

https://twitter.com/Brosner85/status/1314670250315706370

I'll take it. We had a great amount of polls for the past few days but today's been lacking. At least we get NYT/Siena polls on a weekly basis from here on out.

[redjohn]

New WSB-TV/Landmark polls of GA (presidential, Senate races) coming in about an hour:

https://twitter.com/Brosner85/status/1314670250315706370

Trump +1

Close, Trump+2. Biden steady at 47% but Trump's up 4 points. Good news for him in GA.

[redjohn]

I'll take it, but I'd *really* like to have similarly devastating polling five days before the election. Hopefully this trend continues, because we need a complete and utter collapse of Trump's campaign to rack up the largest margins in the EC and Senate races as possible.

[redjohn]

NYT/Siena has announced more polls for this week:

- Michigan and Wisconsin tomorrow at 1 PM ET
- Alaska, North Carolina, and South Carolina later in the week

Previous NYT/Siena poll of MI (from June) had Biden 47%, Trump 36%. Their most recent WI poll (mid-September) had Biden 48%, Trump 43%. Hoping for far fewer undecideds in both.

[redjohn]

Polls are meaningless if they don’t translate into actual votes. Between the disaster of mail in ballots, voter suppression due to Covid, and the beyond drastic enthusiasm gap between Trump and Biden, these are all strong indicators for a Trump re-election.

The polls are misleading, and the talking heads of the mainstream media are wrong.

This is a very lazy take. Of course polls don't translate into actual votes...What they do is predict who is going to vote, and based on a representative sample size they predict the ballpark of the results. I'm not sure what "voter suppression due to Covid" means or how it hurts Democrats, and the enthusiasm gap is not real. You believing that the ~42% of likely voters who are currently planning on voting for Trump are more enthusiastic does not mean that pool of voters is going to make up pluralities or majorities in the battleground states Trump needs to win.

It's fine to base predictions on gut feelings, but this is discussion of polls and if you truly believe the polls are wrong then you need to quantify your reasoning and your reasoning is just random assumptions that won't actually change anything.

[redjohn]

Trump really only needs FL and PA, both of which are getting much closer. It’s your funeral, I suppose.

Is the margin in PA getting closer?

Recent PA poll, Biden +2. I’ll take that heading into the election should it hold.

Give it up, Biden leads by 7+ in every recent high-quality PA poll. Cherry-picking low-quality polls that *still show Biden with a lead* isn't going to help the incumbent President who's badly trailing in the state most critical to his re-election.

[redjohn]

The gig is up on trafalgar.  Totally figured out how they manipulate the numbers.   Look at the last page every report.  It gives you the relative % of the poll by congressional district.  In EVERY case the lower percentage districts are the democratic districts.   They over sample and then cut back the dem districts to give the narrative they want. Trump by 2 in Florida is amazing news bc that means Biden is truly ahead.

yup we all know he's up 5 points in florida 

I swear to god, if Biden wins 413 electoral votes you people will come on here a week later with "BuT BiDeN DiDn'T WiN SoUtH CaRoLinA". I don't think the cognitive dissonance will ever end.

[redjohn]

https://twitter.com/Nate_Cohn/status/1317813987782893578

Going to be interesting to get HQ polling of battleground states this week. I doubt the state of the race has shifted much, but any shifts from now until election day are going to become increasingly meaningful with each passing day.

[redjohn]

Kansas poll

Gonna need a little more, SD.

There's people on twitter saying we're getting NYT polls of Kansas and Iowa this week, but who knows.

[redjohn]

If the NYT national poll has more or the same % of third-party/undecided as last time (10%) it'll be a little ridiculous. They've received a lot of criticism over not pushing undecideds so hopefully they push a little more.

[redjohn]

National polls are getting ever closer.

Practically all swing states have Trump gaining ground, and are now within margin of error.

Trump is going to win this thing. My confidence is growing ever stronger.

Biden+9 gives you confidence?

Please. You can’t just look at numbers at face value. You need to know how to synthesize and use the data avaiable.

You mean read the numbers in any way that distorts the result to be positive for Trump... anywho. Trump is going to need drastically better numbers in the next two weeks to have a chance at winning.

[redjohn]

https://twitter.com/PpollingNumbers/status/1318651432598384643

RCP selectively picks pollsters that show better results for Trump and always picks the smaller margin. Their polling average this time around is not a real polling average.

[redjohn]

https://twitter.com/MULawPoll/status/1319020907780780037

Final Marquette poll of WI comes out in one week.

[redjohn]

https://twitter.com/PollsAndVotes/status/1319389408068587522

Not sure if the October poll he's referring to is the one currently being taken, or the one from Sept-Oct at the very beginning of the month. Still, these are obviously pretty bad trends for Trump in WI.

[redjohn]

https://twitter.com/PollsAndVotes/status/1319389408068587522

Not sure if the October poll he's referring to is the one currently being taken, or the one from Sept-Oct at the very beginning of the month. Still, these are obviously pretty bad trends for Trump in WI.

I don't think they are going into the field until after the debate.

Nope, seems they're currently in the field (surprisingly).

https://twitter.com/PollsAndVotes/status/1319083767533850627

Not sure why they're starting so early, but they normally finish up their polls several days before release.

[redjohn]

What's the point of being a gold standard poll if your last poll is done like a week and a half before election day??

Their last poll in 2016 also ended a week and a half before election day, and they ended up not at all catching Trump's final surge. I'm also surprised they're not taking the poll closer to election day.

[redjohn]

https://twitter.com/bluestein/status/1320470460975534080

I really needed a reason to wake up at 5 AM, this is great

[redjohn]

Tomorrow:
TX - NYT/Siena
GA - AJC

Wednesday
WI - Marquette Law School

Pretty remarkable that we're on the verge of getting our last quality snapshots of the race in these battleground states. We're so close to the finish line.

[redjohn]

Do we know what time Monmouth is releasing?

[redjohn]

Do we know what time Monmouth is releasing?

I think they usually release at 2pm Eastern time.

Great, so we'll have just enough time to react to Marquette once this hits.

[redjohn]

Way less polls than I would have expected.No fox polls last night.My theory is that a lot of these people don’t want to be wrong.

I feel like we've gotten a pretty big number of polls the past couple of days. Maybe Fox drops some final polls this weekend.

[redjohn]

Selzer is the last big poll I'm interested in. Probably the last data point that will shape many of our predictions.

[redjohn]

Has Nate Cohn said when the final four NYT/Siena polls will be released? I'm guessing they won't release them one-by-one with such little time left.

[redjohn]

The Selzer poll deserves all the hype (although wasn’t great in 2018).  Grab a beer or drink, we are tailgating for this 6pm reveal!

True. They were spot on with the Democratic caucuses earlier this year though, seems they haven't lost their touch yet.

[redjohn]

New Quinnipiac polls are literally disastrous for Republicans.