The Pessimistic Outlook for Trump

October 28, 2016

Happy Friday. So far, more than 16 million votes have been cast in this year's election.

Yesterday you got the optimistic outlook for Trump. Now for the pessimistic outlook . . .

The early-voting numbers look ominous for Republicans in a lot of states. Let's make all appropriate caveats — we don't know for certain that a registered Republican will vote for the Republican, and that a registered Democrat will vote for the Democrat. And yes, there's always the question of whether a high early-vote total is a genuine surge or just an early-vote operation cannibalizing party regulars who would otherwise vote on Election Day. But every voter who votes early is one less for the campaign to worry about on that Tuesday. Generally speaking, a Republican candidate would want to see his party having a nice, solid lead in early voting.

Let's start in Arizona, where Hillary Clinton led two of the last three polls.

Four years ago yesterday, registered Republicans had a 21,179 vote advantage over registered Democrats in early voting. This year, registered Democrats are ahead by 4,116 votes. This doesn't mean the state is lost; the early vote is roughly 440,000 so far, and last time Romney won the state by about 200,000 votes, out of 2.3 million cast. But clearly the Clinton campaign thinks the state is in play and is willing to spend money there.

Quinnipiac puts Trump ahead in Georgia by one point. That's a shocker, but the previous three polls had Trump only ahead by 2 or 4 points. Early voting is up 40 percent from 2012. The state doesn't register voters by party but keeps demographic information about them; 28.3 percent of the early voters are African American and 56 percent are women. In 2008, the last cycle we have exit-poll election results for, African Americans were 30 percent of the electorate in the Peach State and women were 54 percent. McCain won the state by 5 points that year. Right now you would expect a narrow Trump win, but when a state is this close, a loss is not out of the question.

In Iowa, registered Democrats continue to outpace registered Republicans in returned absentee ballots, but it's a smaller margin than last time. Quinnipiac and Loras have shown the state tied, although the most recent Des Moines Register poll puts Trump up by four points. Based upon how Trump has led here for most of the year, you would expect a win . . . but if the Democrats get-out-the-vote effort works well and the GOP one stumbles . . .

It's not certain that Trump will win one of Maine's two congressional districts. Registered Democrats continue to outpace registered Republicans in returned absentee ballots in both districts. Trump plans to campaign there again, so obviously the campaign feels it is winnable but not yet locked down.

Mike Pence was in Omaha, Nebraska yesterday. You don't send the vice-presidential candidate to Nebraska less than two weeks before Election Day if you're not worried about the Second Congressional District, which Barack Obama won in 2008.

For the rest of the map, just give Hillary Clinton the states where she leads in the polls. As noted yesterday, Trump has never led any independent general-election poll in Michigan, Minnesota, New Hampshire, New Mexico, or Wisconsin. She's led most of the recent polls in Florida, Nevada, and North Carolina.

Right now, you would expect Trump to narrowly win Ohio, but there have been much more surprising outcomes than Hillary Clinton winning in a state where she has 36 offices and Trump has 16.

Assume Evan McMullin wins Utah. Again, the campaign schedule reveals the perceptions within the campaign; the Trump camp sent Pence there recently.

Assume Hillary Clinton wins everywhere on this list and you have the true worst-case scenario: 375 electoral votes for Clinton, 157 for Trump, and 6 for Evan McMullin. (Under that outcome, Trump would have the fewest electoral votes by a major party candidate since Michael Dukakis.)

The best-case scenario and the worst-case scenario are both unlikely; the most likely outcome is somewhere in between.

It's a Cliché, but an Accurate One: Which Side Can Bring Out More of Their Voters?

At the heart of this question is whether Trump needed a major investment in a data-driven get-out-the-vote system with more professionals than volunteers, lots of field offices across swing states, and whether he could afford to concede television advertising for long stretches during the year.

Yes, the Romney campaign's Project ORCA had major glitches and management problems on the morning of Election Day 2012. But by the end of the day, the campaign had reports from about 90 percent of the counties they were targeting. Ultimately, it wasn't a crashing web site that cost Romney the 5 million vote margin.

In 2011, the Obama campaign saw that they could very well lose, and went out and targeted unregistered voters who were likely to support their candidate and got them registered. The Trump campaign thinks they're going to bring out new voters, but the data — at least in Ohio — suggest that it isn't happening, or at least not in the areas of the state you would expect to be most Trump-friendly. The theory of finding disaffected or unregistered blue-collar whites and adding them to the voter pool might actually work. But you need to do the work, not follow the Field of Dreams strategy, of building it (or putting forth a candidate who appeals to them) and waiting for them to come.

Most Voting Machines Will Work Fine . . . But Check Anyway.

For those who suspect the machines will flip their votes from their preferred candidate to the other candidate . . .

A voter who experienced a problem while early voting contacted the AJC, saying it took three tries Tuesday on a machine at the county's administration complex in Richmond Hill before it correctly recorded his choice of Democratic presidential nominee Hillary Clinton.

He said on the first two tries, he initially selected Clinton but the touchscreen on the machine then changed to show his selection as Republican presidential nominee Donald Trump — which he canceled before trying again. He said his wife had a similar experience on the same machine.

The voter requested that the AJC not publish his name.

The Georgia Secretary of State's Office has opened an investigation into what state officials said was isolated incident.

"We are confident that machines are not 'flipping' voters," said David B. Dove, the office's chief of staff and legal counsel. "It appears with this particular machine that the county did not properly conduct logic and accuracy testing on this unit. That test ensures the geographic areas on the unit's screen corresponds to the underlying ballot format. This testing is required by state law."

Bryan County Election Supervisor Cindy Reynolds told the AJC that the machine was one of eight in operation for early voting. At least 20 people had previously used the machine that day; no one reported any problems to poll workers but, she said, "I went ahead and took it down just to be sure."

ADDENDA: In lieu of the pop culture podcast this week, a slice of suburban dad life. In Authenticity Woods, Fairfax County, we don't have a Halloween parade at the highly ranked public elementary school, because some students don't celebrate Halloween. No, we have a "Fall Spirit" parade. Fine.

No toy guns, knives or other weapons in costumes? Fine. We live in tense times, and if you give a child nunchucks, they will inevitably start whacking each other with them.

But now the theme of this year's parade for my older son's class is Virginia industries, and he wanted to be Newport News Shipbuilding. I think the kid down the street has to be a bale of tobacco or something. "Hey, Daddy thinks you should be a giant bottle of whiskey, that's a Virginia product!"

My younger son had to be a storybook character, and he chose the Duckling from The Duckling Gets a Cookie!?, part of Mo Willems' painfully insightful series about a misbehaving pigeon who has no impulse control and emulates every toddler trick for emotional manipulation in the book. The Duckling is actually a next-level manipulator, teaching children the valuable lesson that you need to at least appear to be behaving well in order to get what you really want.