Michigan, Colorado or Faithless Electors Could Scuttle Clinton Coronation

“Beds are damp. There is a crack in the blue wall, and it has to do with trade. This is the ghost of Bernie Sanders.”

Van Jones on CNN re: Michigan

Either former Methodist Sunday schooler Hillary Clinton or former New York Military Academy cadet Donald Tump will claim the mantle of the next U.S. Presidency within the next forty hours.

Or perhaps both will. We’ve had two Popes, even three. Why not two Presidents?

Clinton has a 45.4% to 42.4% average national poll lead by my calculations. Those calculations almost exclusively involve strictly averaging the latest results from all pollsters in the field within the last ten days. Furthermore, for months, the mathematical geek aggregators’ consensus has been that Trump’s chances of upsetting Clinton are slender even on his best days. If Clinton wins her “firewall” of states totaling 272 electoral votes, states where she’s led and usually led handily for most of the way, or if she loses one of them like New Hampshire, but ekes out a victory in hard fought Nevada, as now widely expected, or a win in Florida, North Carolina, or even Ohio, the candidate Democratic operatives, major media, and Barack Obama have fought so hard to help coronate will finally have won her imperial crown.

If, on the other hand, Trump can breach the firewall without leaving behind states he is expected to win like Ohio, Iowa, and Arizona, the Donald will assume the position of Groper-in-Chief, a position that has moldered vacant these sixteen years since Clinton’s consort William Jefferson and his merry band of playboys left the West Wing bereft of any keyboards with the letter W, with a “Jail to the Thief” sticker stuck to a presidential filing cabinet and with obscene anti-Bush slogans vandalizing the loos.

To breach the firewall, Trump must claim victory in Florida and North Carolina and win one of a handful of states like Michigan, Nevada, Colorado, Pennsylvania or, in a real stretch, Virginia, New Mexico, Wisconsin, or Minnesota. These last four states are ones where polling averages, mine and the average of averages for FiveThirtyEight, RealClearPolitics, Huffington Post and Drew Linzer’s Votamatic, see Clinton maintaining a lead of 5.0% to 6.5%.

By my count, just eight state contests over the last three presidential cycles have seen RealClearPolitics (RCP) average miss by 4.5% or greater. Bush outperformed in Florida and Hawai’i in 2004. Obama underperformed the RCP average in Iowa, still winning handily, and overperformed in New Mexico and Nevada in 2008. Finally, Obama handily beat polling averages in Oregon and Michigan in 2012. This last polling miss, by 5.5% in Michigan last time around, is particularly relevant for what lies immediately ahead.

Michigan was, of course, also the site of the “one of the greatest upsets in modern political history,” when Bernie Sanders out performed polling averages by more than 20% to win the Democratic primary in March of this year.

correctly predicted the Sanders’ upset and am predicting, once again, that Clinton will defy polling expectations, including my ten-day average modelling, and lose Michigan.  This does not necessarily mean I am predicting that Clinton will lose the election as a whole. I will not bore you here with a long explanation of my reasoning, but it is clear that the Clinton camp is also seeing the very real potential of a loss in Michigan. This tracks very well with what I am hearing from family and friends and extended networks in the state where I vote.

For virtually all other states, I am sticking with the data I and others have collected, though I should note that I have a strong suspicion that Trump may win in Colorado and that Clinton could pull off a big surprise in Arizona on the strength of Latina and Latino voters. This latter possibility could take weeks to settle given how badly Arizona runs its elections and election counting. Early voting in Colorado has not gone nearly the way Democrats would have liked it to go and the best pollster in the state projects a 39% to 39%tie with lots of third party and undecided voters. Of note, I am projecting that Donald Trump will split out Maine’s second congressional district, winning it by a few percentage points and picking up an extra electoral vote. I should also note that my polling averages, as of now, project a 0.4% Clinton win in North Carolina, but I am going with the average of other aggregators to pick a small Trump win in the Tarheel State.

Here is what that data, which I’ve been tweeting about (#10at10) for a few months, looks like as of 3:30 p.m. eastern. Other poll aggregators’ data in the chart was last updated at 10 p.m. last night. I intend to update the chart around 10 p.m. eastern tonight when I expect that all polls will have come in.

ten-at-ten-states-penultimate

As an additional wildcard, at least one, if not two Electoral College voters in Washington state are insisting they may well not vote for Clinton on December 19. These potential unfaithful electors mean Clinton needs at least 272 Electoral Votes, if not a few more to guard against the possibility of further defections.

Florida is too close to call. My prediction is a 0.5% gap in one direction or the other,triggering an automatic recount, a 2000 redux if you will. If absolutely forced to pick who will win Florida and thus the presidency, I would waver for a long while and finally suggest Clinton by a nose, sticking with my data that projects a 0.4% Clinton win.

Here’s what that electoral map looks like, Florida a toss-up, courtesy 270towin.com.

final-pick-2016

Doug Johnson Hatlem writes on polling, elections data, and politics. For questions, comments, or to inquire about syndicating this weekly column for the 2020 cycle in your outlet, he can be contacted on Twitter @djjohnso (DMs open) or at djjohnso@yahoo.com (subject line #10at10 Election Column).