As I write this, there are 183 days until Election Day. There are 15.8 million more seconds of tweets, fights, ads, calls, debates, conventions and other nonsense.
A lot of time, but time flies.
It's also 183 days during which the race can change. In recent contests, the day-to-day fluctuations of the contest haven't really made much difference toward the outcome, but there's one way in which a significant change over the next 16 million-odd seconds could be important: If Republicans finally embrace Donald Trump and, as a result, his favorability numbers increase.
There has been a lot of discussion over the two likely nominees' favorability ratings recently, for the simple reason that the numbers are terrible. Both Trump and Hillary Clinton are viewed negatively by more people than they are positively — in other words, the net favorability for each is negative. And since 1992, the candidate who has the strongest net favorability in the last Gallup poll before Election Day has won the popular vote five out of six times. (The sole exception was George W. Bush in 2004.)
Clinton is doing better on net favorability than is Trump right now, thanks to Trump's terrible numbers with women and non-white voters. (His numbers among Republicans are also lower than Clinton's are among Democrats, in part thanks to women viewing him negatively.) This suggests, based solely on this metric, that Clinton is better positioned than Trump.
But those numbers are not as static as they may seem. Plotting net favorability for the winning (green) and losing (yellow) candidates over time during those six elections using Gallup's data, it's clear that there's a lot of variability.
More than half of the time in 2000 and 2004, the candidate who lost the popular vote had better net favorability ratings than the winner. In 1992, Bill Clinton had better net favorability ratings than George H.W. Bush only about two-thirds of the time. Even in 2008, Barack Obama's net favorability was often lower than John McCain's — until the end.
Those dots can be confusing. So here are the average net favorabilities for each winner and loser in 50-day chunks before the election. Here, you can see that George W. Bush's net favorability ratings in the last 50 days were, on average, higher than Al Gore's — and Bush, of course, went on to win the tiebreak in the election.
Trump's favorability has been compared to W.'s in 2004. Bush's unfavorability numbers that year were much worse than John Kerry's at this point. That contest was very different than this one for a lot of reasons, but Bush's unpopularity didn't cost him reelection.
Trump has about 16 million seconds to engineer a similar miracle.