After lagging for months at historic lows, Mitt Romney's personal popularity has advanced in the final weekend to its highest of the 2012 campaign, rivaling Barack Obama's. But Obama pushes back with greater enthusiasm among his supporters - and the race itself remains a tie.
Fifty-four percent of likely voters in the latest ABC News/Washington Post tracking poll express a favorable opinion of Obama overall, the most basic measure of a public figure's popularity. Yet 53 percent now see Romney favorably - a majority, remarkably, for the first time.
It's a dramatic gain for Romney, who emerged from the Republican primaries as the least popular major party candidate in polling back to 1984 and remained there up to the debates. Just 40 percent saw him favorably as recently as late August, and it was essentially no better, 44 percent, after the party conventions.
The presidential debates clearly helped him: Sixty-two percent of likely voters describe Romney's performance in the debates as a factor in their vote, and those who do so are a broad 23 points more likely to see him favorably overall, 61 percent vs. 38 percent. Substantially fewer cite Obama's handling of the response to Hurricane Sandy as a factor in their vote, 49 percent.
While the gap in overall favorability has closed to naught, and the debates boosted Romney's personal popularity, Obama holds advantages in strength of support. Thirty-eight percent of likely voters see him "strongly" favorably, 8 percentage points more than say the same about Romney. "Strong" approval of the president's job performance, at 34 percent, ties its highest since 2009, and is nearly double its low in August 2011. And the share of Obama's backers who say they're "very" enthusiastic about him has reached its highest of the campaign, moving ahead of Romney's by a significant margin, 6 points, for the first time in two weeks.
Notably, with 69 percent of his supporters very enthusiastic, the president has regained the strength-of-support levels he achieved in the 2008 election, after running behind that pace earlier this fall. Romney's come even farther, more than doubling strong enthusiasm among his supporters since the spring, to 63 percent now, also numerically a new high.
THE RACE - These elements play out in a race that couldn't be closer: Obama and Romney receive 48 percent support apiece among likely voters in this survey, produced for ABC by Langer Research Associates. By several standards their long-running deadlock has constituted the closest race in pre-election polling in decades.
As if it's needed, independents - potentially swing voters in national elections - provide a further exclamation point. They now also divide precisely evenly, 46-46 percent. That matches Obama's best to date among independents, and it's a new low in this group for Romney, who'd reached 58 percent support among independents just a week and a half ago.
One reason the bottom line has not been affected is that, as he's slipped among independents, Romney's bulked up among Republicans, with a new high of 97 percent support in his own party. If that holds Tuesday it'd surpass the in-party record in exit polls dating to 1976 - 93 percent for George W. Bush in 2004.
Obama has 91 percent support from Democrats - not shabby, but the fact that he's losing 8 percent of Democrats to Romney, while winning only 3 percent of Republicans, is a challenge. John Kerry suffered a similar disproportion in 2004; the difference is that, in this poll, Democrats slightly outnumber Republicans among likely voters, by 4 points, while turnout between the parties was dead even in '04.
THE VOTE - While the focus is on Election Day, for many that's passed: Twenty-five percent of likely voters say they've already voted, and an additional 16 percent intend to do so in advance. Obama is +7 points among those who've voted or plan to vote early, trailing the wide 18-point advantage among early voters he held in ABC/Post's 2008 pre-election polling. Romney, for his part, is +4 points among those who plan to vote on Election Day. While this year's differences aren't significant given the sample sizes, these kinds of numbers, if they hold, could be a key element in the outcome of the race.
Then again, with the contest this close overall, the result could turn on truly any factor - for example, how many women vote (now +6 Obama) vs. men (+7 Romney); how many whites vote (+20 Romney) vs. nonwhites (+59 Obama); how many young adults vote (+25 Obama) vs. seniors (+12 Romney); how many voters with post-graduate degrees vote (+20 Obama) vs. the larger proportion with undergraduate degrees only (+5 Romney); and whether evangelical white Protestants (+70 Romney) participate in their customary numbers.
In a close contest, in fact every group matters, which explains the campaigns' hyperactive voter-contact efforts. Nationally 43 percent of likely voters say they've been contacted by either or both campaigns; in the eight states now classified as battlegrounds by the ABC News Political Unit that soars to 65 percent. Contact rates have been nearly equal by both camps, a measure on which Obama outpaced John McCain four years ago.
ECONOMY - The issue of the race is the economy; Obama's vulnerability on it has been Romney's key to the White House, yet Romney's been unable to grasp it firmly. Today likely voters divide by a close 49-46 percent, Romney-Obama, in trust to handle the economy. A significant lead on this issue is what Romney's needed to break away.
Obama, for his part, is back to a slight 6-point advantage on economic empathy - another element of his pushback against Romney, yet again, not nearly as wide as the president's advantage on empathy earlier in the campaign.
With no strong breakout by either candidate on the economy or on economic empathy, a range of other concerns, predispositions and candidate attributes come to the fore. Obama generally led on those before the debates; since then the divisions have tightened. Favorability is a prime example: a measure on which the candidates, as with the horse race itself, are tied.
METHODOLOGY - This ABC News/Washington Post poll was conducted by telephone Oct. 30-Nov. 2, 2012, among a random national sample of 1,809 likely voters, including landline and cell-phone-only respondents. The survey, conducted after Hurricane Sandy hit the East Coast, includes interviews with 283 likely voters in the Northeast. Results in that group are in accord with their pre-storm levels.
Results have a margin of sampling error of 3 points, including design effect. (Question 11 was asked Oct. 31-Nov. 2 among 1,485 likely voters; those results also have a 3-point error margin. Question 29 was asked Nov. 1-2 among 1,164 likely voters; those results have a 3.5-point error margin.) The survey was produced for ABC News by Langer Research Associates of New York, N.Y., with sampling, data collection and tabulation by Abt-SRBI of New York, N.Y.
Partisan divisions, Democrats-Republicans-independents, are 33-29-34 percent among likely voters; they were 39-32-29 percent in the 2008 exit poll. The ABC News Political Unit now designates "battleground" states as Colorado, Florida, Iowa, Nevada, New Hampshire, Ohio, Virginia and Wisconsin.