…Candidates polling ahead after the conventions have won the popular vote in 16 straight elections.

 

No candidate in Donald Trump’s position at this stage of the campaign has gone on to win the popular vote in November in the modern polling era.
That’s the sobering news confronting the Trump campaign as it seeks to rebound from his recent slump.

Those who have studied presidential campaigns since 1952 — the advent of modern polling and TV — aren’t writing off Trump yet. But they say that time is running short for Trump to reverse the trajectory of the race before voters’ preferences become locked in.
There are only a handful of opportunities left for Trump to impact the race, including the three general-election debates, in which Trump has indicated he intends to participate, though he wants to negotiate some of the details. But some pollsters caution that there is a significant risk that, even by the first debate in late September, the race will be locked in against Trump, who would be on an inexorable path to defeat regardless of how he performs.
“When you come out of the conventions, the leader in the last 16 elections has not lost the popular vote,” said University of Texas professor Christopher Wlezian, who co-authored the book “The Timeline of Presidential Elections: How Campaigns Do (and Do Not) Matter.”
There’s another complication for Trump that candidates trailing historically didn’t face: the proliferation of early and no-excuse absentee voting over the past decade or so. According to the National Conference of State Legislatures, 37 states allow voters to cast their ballots before Election Day if they wish. And, in some states, the no-excuse absentee ballots are mailed to voters as early as next month.
The expiring clock is one reason why dozens of Republicans are appealing to Reince Priebus, the chairman of the Republican National Committee, to shift resources away from Trump now and apply those funds and staff to downballot races. (The RNC said this weekend that such a decision won’t be made “until late September or October.”)
But Trump likely needs to reverse his deficit sooner than that. Wlezian predicted if Clinton holds onto her current margin over Trump after just one more week of polling — making it three weeks since the Democratic convention — she would have a nearly 90 percent chance of winning the election, based on past campaigns.

And there won’t be much Trump can do about it, he said, because the time period around and immediately following the conventions is when voters make up their minds and rarely waver even as the campaign progresses.
“It’s not magical,” Wlezian said. “[The electorate is] substantially baked. Not everybody’s locked in, but there’s a lot of voters locked in now.”
Still, there are reasons to suggest that history may not be the best predictor: 2016 has been an outlier election in many respects, and the final 12 weeks of the campaign won’t be conventional. Trump is a unique candidate, and while he is easily the most unpopular presidential nominee in modern American history, Clinton isn’t well-liked either.
“It is not that her voters are in love with Secretary Clinton – they just dislike her less than they disdain Trump,” Quinnipiac University pollster Peter Brown said last week. Quinnipiac found 45 percent of Clinton backers in three states — Florida, Ohio and Pennsylvania — said their support for Clinton was mostly motivated by opposition to Trump. But 56 percent of Trump supporters are mostly driven by scorn for Clinton.
Then there is the relatively large bloc of undecided voters, a trove of support that could, theoretically, boost Trump’s chances. In the RealClearPolitics average, the combined vote shares for Clinton and Trump in a two-way matchup equal about 89 percent. That’s consistent with the combined vote share for Gary Johnson and Jill Stein: more than 11 percent in the four-way RealClearPolitics average.
That’s a much higher share of uncommitted voters than other recent elections. In 2012, President Barack Obama and Mitt Romney combined for about 94 percent in polls about two weeks following the conventions.
Among those voters who’ve committed, though, Clinton has a clear lead: Her margin in most national polls is in the high-single-digits, according to most polling averages. And she leads in all 11 swing states that comprise POLITICO’s Battleground States project — including 5 states where Clinton’s lead is 7 percentage points are greater. Trump would have to sweep the remaining six states — where Clinton leads, but by smaller margins — just to force an Electoral College tie.
Some of these Clinton advantages already appear virtually impenetrable. A CBS News/YouGov poll last week in New Hampshire gave Clinton a 9-point lead, 45 percent to 36 percent.
But as CBS News Elections Director Anthony Salvanto explained on “Face the Nation” on Sunday, many of the voters Trump would need to win over in order to come back aren’t open to voting for him.
“So, we asked people who aren’t voting for Donald Trump, would you consider voting for him? And among women, with whom he is down almost 20 points anyway, women who are not voting for him, the number who say, yes, they would consider it is zero. And the number who say maybe is 9 percent,” Salvanto said. “So, if you’re at zero in the number of people who will consider you going forward, that just emphasizes what a tough hill it is.”

Some Trump supporters have questioned the accuracy of the polls that serve as the underpinning for this analysis. Polls of all registered voters, they claim, consistently overrepresent Democratic performance compared to polls of likely voters. Another popular assertion is that Trump backers, fearful to express their support for Trump in a telephone poll, either say they will vote for another candidate or don’t answer the phone at all. And then there’s the blanket declaration that the polls are deliberately biased against the Republican.
There’s little evidence for any of these explanations, however. The data suggest moving to likely voters, which some pollsters are already beginning to do, won’t help Trump like it did for Mitt Romney four years ago. There’s scant reason to believe that “shy Trump voters” exist: Trump didn’t out-perform his polls during the primaries, and he isn’t running stronger in polls that offer an anonymous way to respond, like Internet polls, than in polls with a live phone interviewer. And the idea that a diverse set of pollsters and news outlets — from NBC News to the Wall Street Journal to Fox News — would produce polls that deliberately interviewed more Democrats similarly strains credulity.
For his part, Trump isn’t questioning the poll results as he has in the past. Trump, in a series of tweets on Sunday decrying media bias, tacitly conceded that he is trailing in the race.
“If the disgusting and corrupt media covered me honestly and didn’t put false meaning into the words I say, I would be beating Hillary by 20%,” Trump tweeted.
The three scheduled debates, with the first set for September 26 on Long Island, loom as the next pre-determined inflection points on the calendar. But Wlezien, the University of Texas professor, said even a dynamic performance in the first debate might be too little, too late for Trump if he can’t make inroads soon.
“As things get more and more baked, there’s less room for change,” said Wlezien, who added that the debates are “the second-most important set of events, but they’re far less important than conventions.”