The Other McCain

"One should either write ruthlessly what one believes to be the truth, or else shut up." — Arthur Koestler

‘Don’t Get Cocky’

Posted on | July 13, 2024 | Comments Off on ‘Don’t Get Cocky’

What’s the first thing you notice when you look at today’s Real Clear Politics average of presidential election polls? What stands out to your eye? For me, it’s that NPR/PBS/Marist poll which, alone among all polls in the last two weeks, shows Joe Biden in the lead. Any average is only as good as the components of the total dividend, and this one number — which is eight points off of four different polls showing Trump +6 — ought to attract careful scrutiny. Here’s their story:

Contest for President Still “Up for Grabs”
Biden +2 Percentage Points Against Trump, Despite Concerns about Biden’s Mental Fitness
With just days to go before the start of the Republican National Convention, President Joe Biden and former President Donald Trump continue to be closely matched among registered voters in both a head-to-head matchup and a multicandidate field. Biden’s support remains relatively unchanged from last month despite the view of many Americans that he lacks the mental fitness to serve as president. However, Biden outperforms Trump on whether either candidate has the character to be President of the United States, and by more than two to one, Americans are more concerned about a president who lies than they are about someone who is too old to serve. Both candidates remain flawed in the eyes of Americans, and majorities say neither should be on the ballot. Kamala Harris, Gavin Newsom, and Gretchen Whitmer do not improve the Democrats’ chances against Trump.
Among registered voters nationally including leaners, Biden receives 50% to 48% in the two-way presidential matchup. Two percent are undecided. Biden and Trump were tied (49% to 49%) in last month’s pre-debate NPR/PBS News/Marist survey.

Did you catch that? According to Marist, Biden has actually gained two points since their pre-debate poll. Are they trying to sell us the Brooklyn Bridge or what? It’s not just that their number is out of line with most other polls, but the direction of the trend is backwards. For example, the New York Times/Siena College poll has shown the race going from Trump +1 in April to Trump +4 in June (pre-debate) and now Trump +6 post-debate. Similarly, the Emerson College poll, which had Trump +1 in June (pre-debate) now has Trump +3. Stipulate that a shift in the range of 2 to 4 percentage points may just be margin-of-error statistical “noise,” but that Marist poll still sticks out like a sore thumb.

Here, perhaps, it’s worthwhile to mention that I’ve been reading polls for a long, long time. When I was at the Washington Times, one of my jobs was to update the weekly roundup of polls during election years. This was years before Real Clear Politics had begun its handy aggregation and, indeed, before Google, back in the day when you had to use search engines like Yahoo or Dogpile to find anything on the Internet. So for four consecutive election cycles — 2000, 2002, 2004, 2006 — I was the guy grabbing up all these poll numbers from wherever I could find them and compiling them for a weekly graphic, trying to give readers an idea of where things stood. The longer you do that kind of work, the more you notice certain patterns. For example, as September gives way to October, the media will look to the polls and declare that the race is “tightening.”

What they really mean — the underlying truth — is that most polling operations have a built-in slant toward Democrats. During most of the year leading up to the November election, whether it’s a Senate contest or a presidential race, most polls will show Democrat leading by a larger margin than the final election result. The “tightening” of poll numbers in October always involves the Democratic candidate losing ground, as these liberal pollster/propagandists adjust their numbers so that their final poll isn’t preposterously out-of-whack with the official vote totals.

All of which is to say that Democrats have mastered the art of using polls to influence public opinion, rather than to reflect public opinion, which is why you will rarely see polls overestimate the likelihood of Republican victory. (The 2022 midterms were an exception in this regard.) Democrats know that many people see polls as justifying either hope or gloom, and therefore they publish polls that reliably inflate hopes for Democratic victories, trying to create self-fulfilling prophecies.

This brings us back to my suspicious side-eye glance at that Marist poll number. The only way you can judge the value of a pollster in terms of predictive accuracy is to look at their past results. In October 2020, Marist released a poll showing Biden leading by 11 points. The final official popular vote had Biden winning by 4.5 points, which meant that Marist’s poll overstated Biden’s support by 6.5 points. Apply that error rate to the current Marist numbers? Biden’s down by at least 4 points.

Not that I’m blowing sunshine up your skirts, you understand. We have no idea what will occur between now and November, whether Biden will be forced off the ticket, or whether Democrats will be able to manufacture enough fake “votes” to elect their candidate (who may or may not be Biden). Overall, however, I’d rather be playing Trump’s hand than to be playing Biden’s hand right now, and I’m not going to let some dubious polls convince me otherwise. Remain calm. All is well.



 

Shop Electronics at Amazon

Save on Groceries and Everyday Essentials

Shop Amazon Basics

Office & School Supplies

Category: Donald Trump, Election, Joe Biden, Polls

Comments

Comments are closed.