How to Read a Poll and the Race for the White House

By now, polls should have mostly absorbed any potential impact of the conventions and it makes sense to look at them again. Joe Biden’s lead in the national polling average on realclearpolitics is still substantial (6.9%) and double the lead Hillary Clinton had on Donald Trump. I can’t remember an incumbent president running for reelection who was trailing so far behind as Trump does at this point in time of the campaign and who ended up winning. Am I saying that Joe Biden will win? No, because surveys are never a prediction. But I would much rather be in team Biden than in team Trump at the moment.

Let’s look at the underlying dynamics of the race.

Conventional wisdom has it that an election with an incumbent is foremost a referendum on the incumbent. When voting for a challenger, voters merely express a wish. When voting for an incumbent, they render a verdict. In that respect, the job approval rating of the incumbent is crucially important (and was sometimes a better prediction of the vote than the match-up question). At the time of writing, 44% approve of the job Trump is doing while 56% disapprove. This has actually been remarkably stable throughout his term. Trump’s personal favorability ratings are basically the same: 42% have a favorable opinion about him while 56% have an unfavorable one. So it’s not that voters disapprove of the job he is doing but would like him as a person (or vice versa). In addition to job approval and personal favorability, another important indicator is how voters think about the state of the country. At the moment, 27% say that the country goes in the right direction while 67% say it’s off the wrong track. This number has slightly improved but is still dangerously low for an incumbent.

Now, these are all nationwide numbers and one might argue that it’s not really a nationwide, but a state-by-state election. Let us also remember that the state polls were pretty skewed four years ago. But the race does seem to have tightened in the battleground states. Biden is still ahead in the realclearpolitics polling average in most swing states, but the lead has become smaller: Florida (+1.8%), Pennsylvania (+4.2%), Michigan (+2.6%) and Wisconsin (4.4%). In other words: screw the popular vote. This campaign is now all about the electoral college.

What would I advise team Trump? The most viable option is probably what I call the ruthless counteroffensive. The main point of this strategy is to try and change the dynamics from a referendum to a binary choice and to define Biden negatively. I am aware that Joe Biden is much less polarizing than Hillary Clinton. He is polling well among crucially important suburban women. Senior citizens, another key target group, are voting for Biden in numbers that no Democrat has reached since Al Gore in 2000. At the moment, Biden’s overall favorability rating is 48% favorable vs. 47% unfavorable. A good attack campaign could tip the balance. It’s the only option I see for Trump and one that actually fits him in the sense that it consists of what Trump usually does: double down. I would also try to keep Biden from establishing economic competence and credibility, which is actually still a relative strength for Trump. His job approval on the economy specifically is still at almost 50%. With the number of new coronavirus cases dropping, this might well be the crucial issue when voters go to the polls on the first Tuesday in November.

Dr. Louis Perron is a political scientist, consultant and TEDx speaker based in Switzerland. During the past years, he has helped two dozen candidates and parties win election and referendum campaigns. He is the author of the book "How to Overcome the Power of Incumbency in Election Campaigns."