Charles Franklin is hardly a household name, but for a few minutes Wednesday, the full spotlight of the 2016 presidential campaign focused on the veteran pollster and political scientist as he unveiled the latest results from his closely followed poll of Wisconsin voters.
When he announced the verdict: Hillary Clinton leading Donald Trump 46%-40% — almost identical to the margin the poll had found in early October — sighs of relief from Democratic activists could be heard from Brooklyn to Santa Monica.
For Democrats, who, as a tribe, seem more inclined to election-season night terrors than their Republican foes, these are anxious days.
After the final presidential debate last month, they had begun to pinch themselves and talk of landslides. No more. In recent days, once high-flying polls for Clinton have come back to earth.
Indeed, in an ironic — yet, in hindsight, predictable — fashion, the 2016 campaign, after all the controversy and twists, is ending in a familiar way, almost identical to the last days of the race four years ago. The Democratic nominee is moving down the final stretch with a small lead while the Republican scrambles to find a way to flip blue states.
That scenario is testimony to the tremendous pull that partisanship exerts over presidential elections, almost regardless of the specifics of the candidates.
In a closely divided country, it means yet another contest that will leave many voters feeling whiplash.
Even though far fewer polls have been conducted this year than in the last election, social media has made them more available and ubiquitous than ever. News headlines highlight the variations from one survey to the next — the inevitable product of the laws of probability — driving many voters to distraction as they check Twitter for the latest bulletin.
But interviews with leading pollsters in both parties Wednesday found widespread agreement on several points:
Although Trump seems all but certain to perform better with blue-collar white voters than Mitt Romney did four years ago, those gains are being offset by losses among suburban, college-educated voters, especially women.
The lineup of states may shift slightly from 2012 — Clinton may win North Carolina, which President Obama lost, but lose Ohio, which Obama carried — but Trump, so far, has not been able to break the Democrats’ hold on the states they need for an electoral majority.
Last Friday’s news from FBI Director James B. Comey that agents would be examining additional emails that may be related to Clinton has had a small, but mostly transient, influence on the race.
And Clinton holds a lead roughly similar to Obama’s 4-point margin of victory from four years ago.
Some polls, of course, continue to show a path to victory for Trump. A tracking poll by ABC and the Washington Post, which last week showed Clinton holding a double-digit lead, now finds her and Trump tied at 46%.
And the USC/Los Angeles Times "Daybreak" tracking poll, which has consistently shown a better result for Trump than most other surveys, gave him a 5-point lead, 47.8% to 42.4%.
In both parties, however, experienced pollsters don't see the race that way.
Neil Newhouse, who served as Romney's chief pollster four years ago, said Comey's news "put the brakes" on what was starting to look like a runaway victory for Clinton, but that she was still ahead by 2 to 4 percentage points.
"Hillary is still likely to win," Newhouse said, even if that means "limping across the finish line."
On the Democratic side, Anna Greenberg, whose firm polls for many of the party's Senate and congressional candidates, called the race "pretty stable."
"There has been narrowing, which was to be expected. Trump was always going to consolidate more Republicans, and I don't think Clinton was ever going to win by 10 to 12 points" — as some surveys had shown in the days after the final presidential debate Oct. 19 — she said.
The big impact of a tighter race, she and Republican pollsters said, could be to improve the chances of Republicans in down-ballot races by encouraging GOP turnout.
Indeed, the Wisconsin poll, released by Marquette University Law School, found that the Senate race between Republican incumbent Ron Johnson and former Sen. Russ Feingold had narrowed to a near tie. Other Senate races, such as those in North Carolina, Indiana and Missouri, also appear headed to extremely close finishes, with control of the chamber up for grabs.
But, as is the case in Wisconsin, Trump appears to be running several points behind the Republican Senate candidates in most key states.
In Florida, for example, most polls show Sen. Marco Rubio headed for reelection even as the presidential race in the state remains very close.
Data on early voting in Florida shows a strong upsurge among Latinos. That almost certainly benefits Rubio, whose political base is in Miami's Cuban American community.
But a high turnout for Rubio among Latinos also seems likely to help Clinton. Strategists in both parties cite polls showing a significant number of Cuban American voters who are Rubio-Clinton crossovers, driven from Trump by what they perceive as animus toward Latinos.
The underlying stability of the race, which even big headlines have trouble shifting, stems from the reality that both Clinton and Trump are extremely well-known to voters, who have deeply ingrained opinions about both of them.
Both candidates have gained a tiny bit in voters' esteem since the start of October, according to surveys by Gallup, which has tracked their standing all year. But public opinion about Trump remains far less favorable than opinions of Clinton.
By just short of 2 to 1, 63% to 34%, Americans have an unfavorable view of Trump, Gallup found. For Clinton, the net view is still negative, but closer, 54% to 43%.
Just over half of Americans, 51%, say Clinton has the "personality and leadership qualities a president should have," Gallup found, while only 32% said so about Trump — far lower than the other major-party candidates going back to 2000 about whom Gallup has asked that question.
At the same time, approval of Obama's job performance has continued to rise, a fact that boosts his party.
At 54% approval, Obama's standing is now almost identical to that of Ronald Reagan just before his vice president, George H.W. Bush, won the 1988 election.
One shift in the race clearly has happened: Support for Libertarian candidate Gary Johnson has dropped. Johnson has lost about half his backing since his high point in late August, when some polls had him getting just over 10% of the vote.
Whether that drop for Johnson has provided a net benefit for either candidate remains uncertain. Polls indicate that he was drawing roughly equal amounts of support from both sides.
Beyond that, however, the big movement shown in some polls in the last week may be nothing more than a mirage, some leading experts on public opinion say.
All that's really happened in the past several weeks are "phantom swings" caused by shifts in which side is more eager to talk to pollsters, Stanford professor Doug Rivers and Benjamin Lauderdale of the London School of Economics wrote in a blog post Tuesday.
Using data from the YouGov surveys, which Rivers helps direct, the two found that after the "Access Hollywood" video surfaced, on which Trump could be heard boasting that he could get away with assaulting women, people who had previously identified themselves as Trump supporters were 4% less likely than Clinton supporters to respond to a poll.
The reverse happened after Friday's news about Comey's letter to Congress. Suddenly, it was Clinton supporters who were less likely to respond to a survey, they found.
"When things are going badly for a candidate, their supporters tend to stop participating in polls," Rivers and Lauderdale wrote. But that doesn't mean they've decided not to vote.
"Clinton was never as far ahead as many published polls suggested," they concluded. "Equally … she has not lost as much by recent events as some published polls suggest."
"The truth is more boring: Real change mostly happens slowly, and the impact of campaign events is much less than the media makes out."
For more on politics and policy, follow me on Twitter @DavidLauter.