Just because it’s far too early to get bent out of shape about any single poll doesn’t mean the media and political analysts will be exercising restraint.
Last week, pundits had a field day with a CNN poll showing President Joe Biden’s approval ratings were 22 points underwater at 39% approve, 61% disapprove. Some Democratic analysts promptly jumped aboard Team Hysteria.
“[T]he reality is, if this were a referendum, he would be in deep, deep trouble,” David Axelrod, a former Obama White House aide and campaign strategist, told The Hill.
The freakout was reminiscent of the final months of last year’s midterms, when media organizations and election analysts, who were riding high on bogus polls and conventional wisdom inside the Beltway, whipped themselves into a red-wave frenzy.
And guess what? The CNN poll has proven to be an outlier, as Democratic strategist and Hopium Chronicles substacker Simon Rosenberg tweeted Sunday.
Among the September surveys conducted by reputable outlets featured on FiveThirtyEight, here’s the breakdown:
Biden’s Job Approvals in September surveys
Approve
Disapprove
Unsure
net approval
YOuGov/CBs newS
40%
60%
0
-20
YouGov/Economist
41%
53%
6%
-12
Morning Consult
40%
52%
7%
-12
Redfield/Wilton Strategies
38%
46%
16%
-8
TIPP Insights
41%
49%
10%
-8
The only survey with net-negative approval ratings approaching those of the CNN poll was YouGov/CBS News, at -20. But one other data point separates YouGov/CBS and CNN from the other outlets: The number of respondents who were unsure or expressed no opinion in either poll was 0%.
All the other polls featured a slice of respondents who declined to answer, while everyone had an opinion in the two outlier polls. If one looks purely at Biden’s approval rating, every poll has him hovering right around 40%. The same is true for Civiqs tracking of Biden’s approvals: 39% approve, 53% disapprove, 8% neither.
The real difference in the net results of these approval polls is whether respondents can or did choose to opt out of giving a response. And as we pointed out last week, Biden’s approvals in Civiqs today are exactly where they were on Election Day last year, when Democrats wildly outperformed expectations.
Once more, Biden’s favorable ratings completely dust those of Republican 2024 hopefuls.
Between now and next November, political observers are likely to see a steady string of freakouts in the runup to Election Day. Let’s all keep several things in mind:
Polls at this juncture (even good polls) are functionally meaningless as predictors of next year’s election outcome.
Last cycle, the Beltway media built an entire narrative around a series of GOP-leaning junk polls while cherry-picking high-quality surveys to buttress the conventional wisdom.
No single poll is the end all, be all; tracking trends and comparing polls to their peers are the only sane ways to analyze the political landscape.
Finally, the predictive quality of polls today has fundamentally changed in this era of hyperpolarization. Data points that historically proved useful to predict electoral outcomes—such as right track/wrong track numbers and feelings about the economy—no longer carry the weight they once did. That is exactly why nearly a century of data suggesting the party in control of the White House was practically destined for a midterm drubbing proved to be terribly misleading in the 2022 cycle.
Where polling is concerned today, there is no infallible North Star. In short, any analysis or pundit asserting certainty in knowing what’s going to happen next year should be immediately discounted. The only thing that’s certain about polling today is that there’s simply no certainty in any measure of the landscape.
