New Times,
New Thinking.

Should we worry about Donald Trump’s poll leads over Joe Biden?

US midterm surveys are historically unreliable – and today’s may prove to be no different.

By Ben Walker

To a number of commentators, particularly here in Britain, Joe Biden is something of a success. The US president’s administration has passed bills left, right and centre and he has performed well in response to the war in Ukraine (unlike some of his western counterparts). Partly as a result, his poll slump has been reduced. Last July about 57 per cent of Americans disapproved of Biden, now 51 per cent do. That might not sound like much of a shift but in an intensely tribal America it is worthy of note. Brand Biden still has steam.

And yet, if you believe some opinion polls, Biden is on course to lose it all. And not just to a generic Republican but to his opponent in the 2020 presidential election, Donald Trump. An Emerson poll conducted this month showed that 44 per cent of American voters would back Trump at the next presidential election, compared with 41 per cent who said they would vote for Biden.

Excluding the undecided, this represents a stark shift from 2020, when Biden won with 51 per cent of the popular vote. Today he’d lose with 43 per cent of the vote to Trump’s 47 per cent. And Trump, as his victory in 2016 showed, doesn’t need to win the popular vote to win the electoral college.

But don’t get excited by this Emerson survey. Of the ten most recent polls, four have put Trump ahead, five have put Biden ahead and one had them tied. On average, Biden beats Trump, albeit by a slim margin – a margin that, again, could still hand the electoral college to Trump. And not only that, but swap Trump for the Florida governor Ron De Santis, who is tipped as a potential Republican nominee, and Biden is more often than not trailing behind.

The truth, however, is that polls such as these, midway through a presidency, are not an accurate way of predicting the next election – sometimes even less so than similar polls in the UK. In Britain dissatisfaction with the incumbent party is often at its height in the middle of a parliament; in the US the same applies.

Select and enter your email address Your weekly guide to the best writing on ideas, politics, books and culture every Saturday. The best way to sign up for The Saturday Read is via saturdayread.substack.com The New Statesman's quick and essential guide to the news and politics of the day. The best way to sign up for Morning Call is via morningcall.substack.com
Visit our privacy Policy for more information about our services, how Progressive Media Investments may use, process and share your personal data, including information on your rights in respect of your personal data and how you can unsubscribe from future marketing communications.
THANK YOU

[See also: Where did America's recession go?]

Take these examples. In February 2019 polls understated Trump’s eventual vote share in 2020 by three to five percentage points and overstated Biden’s – who was then merely a hypothetical candidate, as Trump is today – by two to four points.

In early 2015 Hillary Clinton was polling between four and six points more than her eventual vote share in 2016. In 2007 Barack Obama was polling similar margins behind his eventual vote in 2008 and in some instances was even behind John McCain, the Republican candidate. Eventually Obama won by 53 per cent to 46 per cent.

One of the drivers of such inaccuracy in midterm presidential polls is that the party bases aren’t yet rallied. Or rather, are not yet rallied to the candidates who will be put before them. The “Trump leads” headlines are based on an electorate that was nonplussed about the November midterm elections (turnout was just 47 per cent). Democratic voter apathy, while smaller than in 2021, played its part in those results, gifting the Republicans control of the House of Representatives.

Look at this chart.

Here we have divided the US electorate into key demographics and examined them as a share of the overall voting public. This is for 2020 and 2022 and I’ve coloured them by the leading party/candidate. For instance, 18- to 24-year-olds, who reliably vote Democrat, made up 9 per cent of total voters at the 2020 presidential election but just 7 per cent at the 2022 midterms. Those aged over 50, meanwhile, who typically vote Republican, made up 59 per cent of the total voting electorate last November, up from 52 per cent in 2020.

To summarise these figures, what happened last November was a (disproportionately) lower turnout of poorer voters, younger voters and registered Democrats than in 2020. Middle-income earners were also less likely to turn out. And it is in this context that we’re seeing tighter than expected polls for the president’s party – a reflection of an unenthused Democratic base and a fired-up Republican one.

This is entirely in line with past midterm polling, so I wouldn’t panic (or celebrate, depending on persuasion) just yet. Trump may lead in some polls but there’s no guarantee he will when – or is it if? – Democrats come out to vote.

[See also: What has Kamala Harris been doing?]

Content from our partners
No health, no growth
Tackling cancer waiting times
Kickstarting growth: will complex health issues be ignored?

Topics in this article : ,