Instead of a poll, let’s start today’s Election Update with some actual votes. According to the estimable Nevada journalist Jon Ralston, Democrats have a 20-percentage-point turnout edge so far based on early and absentee voting in Clark County (home to Las Vegas), Nevada. And they have a 10-point edge in Washoe County (home to Reno).

Nevada is one of a number of states where Democrats usually do better in early voting than in the vote overall, so this shouldn’t be taken to mean that Hillary Clinton and the Democratic U.S. Senate candidate in Nevada, Catherine Cortez Masto, are going to win their races by double digits. But Nevada is an interesting state, insofar as both Clinton and Donald Trump can find things to like about its demographic makeup: In Clinton’s case, the growing number of Hispanic and Asian-American voters bodes well for her; in Trump’s case, there’s the fact that only about one-third of Nevada’s white voters have college degrees, according to FiveThirtyEight’s estimates. Furthermore, Nevada has shown tight polling all year, with Clinton having only pulled ahead since the debates — surprising given that President Obama won Nevada by 7 percentage points in 2012 and that Clinton is beating Obama’s numbers in other Western states.

Those early-voting numbers, though, don’t look good for Trump. Democrats are matching their 2012 pace in Clark County, according to Ralston. And they’re beating it in Washoe County, a place where the demographics ought to be relatively Trump-friendly. If Clinton is pulling in her marginal voters and Trump isn’t getting his, things could go from bad to worse for the GOP.

One needs to be careful about drawing too many inferences from early-voting data. There are a lot of states to look at and a lot of ways to run the numbers, and we’ve seen smart analysts trick themselves into drawing conclusions that didn’t necessarily hold up well by Election Day. But it seems fair to say the data is mostly in line with the polls. Democrats are seeing very strong early-voting numbers in Virginia and reasonably encouraging ones in North Carolina, two states where Clinton has consistently outperformed Obama in polls. They also seem set to make gains in Arizona and Colorado, where the same is true. But Democratic numbers aren’t all that good in Iowa or Ohio, where Clinton has underperformed Obama in polls.

The problem for Trump is that taken as a whole, his polls aren’t very good — and, in fact, they may still be getting worse. An ABC News national poll released on Sunday morning — the first live-caller poll conducted fully after the final presidential debate — showed Clinton leading Trump 50 percent to 38 percent. Clinton’s 12-point lead in that poll is toward the high end of a broad range of results from recent national polls, with surveys showing everything from a 15-point Clinton lead to a 2-point Trump edge. But the ABC News poll is interesting given its recency and given why Clinton has pulled so far ahead in it — Republicans aren’t very happy with their candidate and may not turn out to vote:

The previous ABC/Post poll found a sharp 12-point decline in enthusiasm for Trump among his supporters, almost exclusively among those who’d preferred a different GOP nominee. Intended participation now has followed: The share of registered Republicans who are likely to vote is down 7 points since mid-October.

I’d urge a little bit of caution here, given that swings in enthusiasm can be transient and can sometimes exaggerate the underlying change in voter sentiment. Our polls-only forecast has Clinton up by about 7 percentage points instead of by double digits — and our polls-plus forecast would still bet on the race tightening slightly.

But you can easily see how the worst-case scenario is firmly on the table for Trump and Republican down-ballot candidates, where the bottom falls out from GOP turnout. Consider:

  • Trump is getting only about 80 percent of the Republican vote, whereas candidates typically finish at about 90 percent of their party’s vote or above.
  • Furthermore, the Republicans missing from Trump’s column tend to be high-education, high-income voters, who typically also have a high propensity to vote.
  • Voters are increasingly convinced that Clinton will win the election, and turnout can be lower in lopsided elections. (Although, this presents risks to both candidates: complacency on the part of Democrats, despondency on the part of Republicans.)
  • Republicans and Trump have a substantial ground game deficit, with Clinton and Democrats holding a nearly 4-1 advantage in paid staffers.
  • Trump’s rhetoric that the election is rigged could discourage turnout among his own voters.
  • Trump’s base is relatively small, especially if he underperforms among college-educated Republicans.

The nightmare scenario for the GOP is that high-information Republican voters, seeing Trump imploding and not necessarily having been happy with him as their nominee in the first place, feel free to cast a protest vote at the top of the ticket. Meanwhile, lower-information Republican voters don’t turn out at all, given that Trump’s rigging rhetoric could suppress their vote and that Republicans don’t have the field operation to pull them back in. That’s how you could get a Clinton landslide like the one the ABC News poll describes, along with a Democratic Senate and possibly even — although it’s a reach — a Democratic House.

That isn’t the only scenario in play, but it’s an increasing possibility. Overall, Clinton’s chances of winning the presidency are 87 percent according to our polls-only model and 85 percent according to polls-plus.


Source: FiveThirtyEight

Leave a Reply