For months now, Tweety McTreason has trailed Joe Biden in the polls. First, it was only a 5- or 6-percentage-point gap, but since the middle of June, that margin has widened to anywhere from 8 to 9 points, according to FiveThirtyEight’s national polling average.
But until very recently, voters didn’t seem all that convinced that Biden could win. In poll after poll, comparatively more voters said they thought Trump would win reelection in November. Now, though, that view may be shifting.
Over the past two and a half months, the share of voters who said they expect Trump to win has fallen from about 45 percent to around 40 percent in polling by The Economist/YouGov, as the chart below shows, while Biden’s share has slowly ticked up to where Trump’s numbers are. (Roughly a fifth of respondents still say they’re “not sure.”)
Trump’s decline may not seem that dramatic — and it’s not; it’s only a few points lower — but it’s notable because prior to June, he had trailed on this question only once since The Economist/YouGov first asked it in December.In a series of polls starting on December 14, 2019, The Economist/YouGov asked U.S. citizens, “Who do you think will win the 2020 presidential election?” Respondents were asked to choose between Trump and “the Democratic nominee” until April 12, 2020, when the wording for the Democratic choice was changed to name Biden specifically. On May 3, the poll started including a “not sure” option.
” data-footnote-id=”1″ href=”#fn-1″>1
But it’s not just the Economist/YouGov polling that supports this finding. USA Today/Suffolk University found a more substantial drop in Trump’s numbers. In late June, 41 percent of voters said they expected Trump to win, whereas 50 percent said the same in the pollster’s late October 2019 survey. Conversely, 45 percent said Biden would win in June, an improvement from the 40 percent who picked the Democratic nominee in October. Republican pollster Echelon Insights has also observed a downward trend in Trump’s numbers: In a survey completed last week, 33 percent of likely voters said they expected Trump to win, which was down from 39 percent in the pollster’s June survey. Meanwhile, the share who thought Biden would win ticked up to 43 percent in July from 40 percent in June.
On the whole, it seems voters are now less confident in Trump’s reelection chances, and the main driver of that shift may be independent voters. In USA Today/Suffolk’s June survey, 47 percent of independents picked Biden versus 35 percent who chose Trump, a reversal from the October 2019 poll, when 54 percent of independents expected Trump to win compared with 30 percent who said the Democratic nominee would win. And looking across the Economist/YouGov data since early May, the share of independents who expect Trump to win has slid as well, from the low 40s to the mid-to-high 30s.
As for Democrats and Republicans, they mostly say their respective nominee will win, although that wasn’t always the case in 2016, as many Republicans thought Hillary Clinton would win. Nonetheless, that doesn’t seem to be happening in 2020. The Economist/YouGov and USA Today/Suffolk surveys found that Democratic voters are largely confident in Biden’s chances, while most Republicans believe that Trump will win. However, since May, the Economist/YouGov polls show an increase in Democrats’ belief in Biden’s chances and a slight downtick in Republicans’ faith in Trump’s.
Betting markets also point to diminished confidence in Trump’s reelection chances. From mid-March to late May, the president usually led Biden in RealClearPolitics’ average of betting odds: Trump’s chances hovered mostly around 50 percent, while Biden’s stood in the low 40s. But in early June, Biden’s odds surged and outstripped Trump’s; now the markets give Biden about a 60 percent chance of victory, while Trump’s chances have fallen into the mid-30s.
This change isn’t necessarily surprising, as betting markets mostly follow the polling averages. But it’s also not difficult to intuit why more Americans might think Trump will lose the election now than before. The president has consistently received poor marks for his handling of the coronavirus pandemic as well as for his handling of nationwide protests precipitated by the police killing of George Floyd in late May. And Trump’s overall job approval rating has now dipped to around 40 percent in FiveThirtyEight’s tracker. Simply put, past incumbent presidents with those sorts of marks have failed to win reelection.
This is coupled with the fact that Biden’s national lead has grown, and his margin over Trump is now larger than Clinton’s edge at any point during the 2016 cycle. Looking at the Electoral College, Biden also holds sizable leads in key battleground states, which could make it difficult for Trump to win despite those states’ tending to lean more Republican than the country as a whole.
If anything, Trump’s surprise victory in 2016 is likely the biggest reason why more people don’t take a dimmer view of his reelection odds. After all, he was behind in the polls four years ago and yet went on to win, so it’s understandable that even though the margins are larger now, some Americans might be taking an attitude of “fool me once, shame on you; fool me twice, shame on me.” Tellingly, a Monmouth University survey of Pennsylvania voters earlier this month found that about a quarter of respondents believe there’s a “secret” Trump vote, although there’s little evidence to support the idea that “shy” Trump voters exist.
All in all, though, voter expectations and election betting markets suggest that Americans increasingly view Biden as at least an even bet to win in November. None of this means Biden will actually defeat Trump, but these shifts do suggest that the conventional wisdom is catching up to what the state and national election polls have been telling us about the race. The electoral environment could very well change in the next three months, but these indicators are all starting to coalesce around the idea that Trump is a real underdog to win reelection.