Once upon a time, there was a man named Gary Johnson who ran for president. Gary Johnson was never really expected to win the presidential election, but he was doing pretty decently. He had the good fortune to be running against two of the least popular people who'd ever helmed major-party presidential tickets, providing an opportunity, it seemed, for a less-extreme former Republican to make a run at it from the outside. We knew that third-party candidates often faltered as the election approached, but this was 2016! Gary Johnson was a former governor running against Hillary Clinton and Donald Trump. Nowhere to go but up.
But one day Gary Johnson was invited to appear on television to talk about his campaign and someone asked him about Aleppo and he said “What is Aleppo?” and that was that.
From the first debate on Sept. 26 until a few weeks ago, there were two trends: Johnson and Trump were slipping downward in the polls, and support for Clinton was rising alongside the percentage of people who weren't picking any of the four main candidates. (The other main candidate is a woman named Jill Stein; she is an extra in this little drama.) Then Trump stopped falling and then Clinton stopped rising. Then, suddenly and recently, Johnson started falling faster as the number of people who weren't picking any candidate started to drop.
Since the first debate, Clinton has gained 2.2 points and Trump has gained 1.6 — including a big increase over the past 10 days. The number of people who aren't settled on any of the four main candidates is now lower than it was then; support for Gary Johnson (Remember him? From the story?) sank more than anything else.
One side effect of the unpopularity of the two main candidates was that a lot of people were, at one time, willing to consider a third-party candidate or were unwilling to commit to Trump or Clinton. For most of the past few months, the percentage of people picking one of the two major-party candidates in head-to-head polling has been lower than in 2004, 2008 or 2012. That percentage dropped quickly, as noted, and it's now about in line with this point in 2008. In the last week of the election, we can expect it to drop further.
(The percentage not picking any of the four candidates has been more in line with past elections — a function of giving people non-Trump and non-Clinton options.)
What we've seen in our polling is that Trump is benefiting from wavering Republicans settling on their party's nominee for the presidency. The polling average (which is a pretty crude tool for gauging such things) would indicate that those wavering Republicans moved from unsure or backing Johnson back to Trump. If that interpretation is accurate and the trend of those who haven't made up their minds deciding on Trump continues, it's not good news for Clinton.
Which, for her supporters, is a much scarier story.