The graphic above was sampled from David Brady and Douglas Rivers’s post at the Hoover Institute on why the various predictive models probably won’t be effective in predicting the outcome of this particular and peculiar presidential election. I include it because it highlights a number of important points.
First, if Donald Trump wins Ohio, Pennsylvania, North Carolina, and Florida, he wins the election. Can he do that? Beats me.
Second, will Hillary Clinton’s divide ut regnes strategy be effective in holding the states Barack Obama carried in 2012? The Clinton Campaign certainly seems to think so.
Third, the Clinton Campaign’s strategy requires it to spend enormous amounts of money in some expensive media markets. If she is to continue that strategy, she’ll need the prodigious amounts of money she’s been raising ($350 million) and then some.
Regardless, it will be an election like no other. A man running against a woman. The combined ages of the two candidates is the greatest in American history (particularly when neither is an incumbent). Two candidates with such low approval ratings and high disapproval ratings have never faced off. The Clinton Campaign, at least, is likely to spend more money than any campaign in history.
I think an initial indicator of the effectiveness of previous prediction models or mindsets will be in the convention bounce. There seems to be an assumption that the convention bounces will be relatively equivalent. What if Trump ends up with a 3 to 4 point bounce from his convention, but Clinton ends up with none, or maybe even a loss? I don’t know that it would necessarily mean the Republican convention was better (though this possibility cannot be discounted), but it would be a good indication that the usual patterns do not apply to this election.
Two months ago Trump’s negatives were in the 70% range, now appear to be around 6 in ten. Clinton, however, appears to keep getting worse with no bottom yet.
Seems like each party has nominated the only person the other candidate can beat. Clinton is the only person Trump could beat and Trump is the only candidate Clinton can beat.
Arnold Kling made an interesting point about voting third party this year:
“One way to think about this year is that Mr. Trump is the 3rd-party candidate, in the tradition of Ross Perot and George Wallace (with a core constituency inherited from those two), who happened to capture the Republican nomination. Johnson is a more credible major-party candidate who happens to be on the ballot as a Libertarian.”
Unless something drastic changes, I will likely vote for Johnson this year. I found out the hard way that this troubles many of my social media contacts especially since I live in Florida, and have been barraged by a litany of tired arguments about “throwing away” my vote. The lesson learned for me is keep politics completely off of my Facebook page.
George Washington for President
Third-party stuff wasn’t really discussed in the linked piece and could be a significant factor. Looking at Nate Silver’s projections for Johnson a few days ago, there were some interesting states where he was projected to get more than 10% of the vote:
Indiana
Maine
New Mexico
Vermont
There are some other states with double digit showings for Johnson, but where the Democratic Party is moribund and unlikely to be influenced by third/fourth party voting. Other than Vermont, the other three states project a race within the margin of error under minimal polling. There has been one state poll in New Mexico, one in Indiana within the last three months, and one in Maine within the last four months. I think the amount of statewide polling outside of the big three states is probably going to be deficient this cycle.
Also, Silver indicates that his model depresses third-party voting this far out, since traditionally their support erodes as the election nears. He also indicated that one of the reasons his projections are more favorable than competitors is his model prefers to use polls that give the option to vote for third/fourth parties. Clinton does relatively worse right now under these polls relative to Trump, which seems counter-intuitive.
I don’t believe that either Johnson or Stein is capable of carrying any state let alone enough states to throw the election into the House. The old saw about a mile wide and an inch deep is a pretty good description of third party support in the United States. It’s diffused throughout the states and local areas rather than being strong in some region, state, etc.
I don’t think either will carry a state, but they could serve as a Nader-factor and be seen as throwing the election in what might be an unexpected direction. But perhaps more importantly, if they get 10-15% of the votes, they make a lot of states more competitive than are traditionally focused on.