What the polls say after the first presidential debate

[TheProle]

Much of the initial post-debate polling has shown Trump in a stronger position relative to where he was in the last pre-debate surveys from the same pollsters. For instance, Democratic-aligned pollster Data for Progress released a national survey conducted the day after the debate that found Trump ahead of Biden 48 percent to 45 percent, whereas an early May poll from the same outfit had Biden ahead 47 percent to 46 percent. In its post-debate survey, Suffolk University/USA Today found Trump leading 41 percent to 38 percent (with Robert Kennedy Jr. at 8 percent), a shift from a tied race in late April. Leger, a Canadian pollster, found an especially dramatic shift in its polling collaboration with the conservative New York Post, going from Biden +2 just before the debate to Trump +7 right afterwards.

That's not to say every poll found Trump gaining. Some other pollsters, such as SurveyUSA, also released post-debate numbers that had Trump ahead nationally, but don't have a relatively recent point of comparison (SurveyUSA last released a nationwide poll in February, in which Trump actually had a slightly larger lead). Morning Consult's national polling found the race essentially unchanged, with a pre-debate tracking survey showing the two candidates tied at 44 percent and the comparable post-debate tracker showing Trump up 44 percent to 43 percent. CNN/SSRS's new poll found Trump leading 49 percent to 43 percent in a head-to-head matchup, but those are the same numbers as the pollster found in late April. And we're still waiting for some other high-profile pollsters to give us their first post-debate snapshot of the race. In line with G. Elliott Morris's analysis of historical polling last week, we usually like to see about two weeks of data after a high-profile event to gauge just how much a race has shifted.

Nonetheless, some post-debate surveys offered other negative data points for Biden's standing with voters after the debate — including those in his own party. In the 538/Ipsos post-debate survey, conducted using Ipsos's KnowledgePanel, just 20 percent of likely voters, down from 27 percent before the debate, rated Biden's mental fitness to be president as excellent or good — that includes a drop from 56 percent to 42 percent specifically among Democrats. And just 15 percent of voters after the debate said his physical fitness to be president was excellent or good, down from 21 percent before the event. Similarly, a YouGov/CBS News post-debate poll found that only 27 percent of registered voters thought Biden had the mental and cognitive health to serve as president, compared with 50 percent overall who felt Trump did. Among Democrats, only about three-fifths thought Biden had the mental wherewithal to do the job

.

https://abcnews.go.com/538/polls-after-presidential-debate/story?id=111610497

[Silent Type]

1. Thanks for the summary. Wonder how long before "polls are not predictive" posts show up?

[msongs]

2. president hillary probably has an opinion on this :-) nt

[TheProle]

3. He is probably copying and pasting his take as we speak.

[Deek1935]

4. Um yup. Because they aren't. Polls are momentary snapshots with large variances and inaccuracies. For example:

* In 1948 Tom Dewey was supposed to beat Harry Truman on election day according to the polls. Truman won.

* In 1988 Dukakis was ahead of Bush by 17 points in the spring. He got pounded on election day months later.

* In 2012 Romney was ahead of Obama in polls for much or most of the campaign. Obama won handily.

* In 2020 Clinton was ahead of Trump right into election day in polls. We all know how that turned out.

* At the Congressional level we were supposed to get pounded in 2022 according to most polls. We lost a few seats but won well
at the state level and only lost a small number of seats in the House. We actually gained a little in the Senate.

* This past winter Tom Suazzi in NY for Congress was neck and neck in the polls with the R candidate. He won by 8 points.

Polls this far out don't mean shit, and they are often very shaky even in the last days leading up to the election.

Polls aren't predictive. Key factors around governance in office is predictive, especially at the presidential level.

[Silent Type]

5. History ain't a good predictor when it's a totally new situation, Deek. Dukakis and Romney sunk

when they had that singular moment that we couldn’t erase.

[jimfields33]

6. We do need some swing state polls. That matters a bit more.

Hopefully that’s where Biden fares well and best.

[Fiendish Thingy]

7. Well, individual polls are not predictive

Trends can sometimes give a clue, however ,

The shift in the DFP poll is just 2-3 points, likely within the MOE.

Leger polls have a well known RW bias (NY Post!), and IIRC, they were part of the October 2022 flood of shitty GOP polls that manipulated 538’s average/forecast model and caused Nate Silver to lose his job due to his failure to exclude these flawed polls from the averages.

[Silent Type]

8. Like I said "pollS." The OP listed several polls AND trends.