Monday, November 28, 2016

Election: I was wrong. . . and right.

“my prediction that Clinton was a 100% favorite was a glaring miss. My confidence was not partisan spin. It was based on public data, voting history and some sense of the Clinton campaign’s own models. I played with various state scenarios, and even in the most generous outcomes, could not get Donald J. Trump to 270 electoral votes.  [emphasis added]

David Plouffe, manager of Barack Obama’s winning 2008 campaign who oversaw Obama’s winning 2012 campaign from the White House.

It’s not surprising someone would emerge to take on Obama’s elite rule, with all its faults.  I’m surprised Trump was that person.

I expected the polls were right — Clinton up by 3.3% on election day meant Hillary would be president.  If the polls had instead shown her ahead by 1.7%, her current popular vote victory margin, I would have been less sure.

On June 9, 2016 — six months ago — I demonstrated how Trump could beat Hillary in the Electoral College, while still losing the popular vote.  Here’s that prophetic blog post in full:

Conservative John Podhoretz makes the case that Clinton’s path to victory is much smoother than Trump’s. He’s right. We showed how Trump can win, but added that it importantly depends upon Trump holding every state Romney won, plus Florida, Ohio, and Pennsylvania.

It’s important to understand that, contrary to normal thinking, Trump has a better chance of winning the electoral vote than the popular vote.  [emphasis added] That’s counterintuitive because Democrats have more big electoral vote states they will likely carry than does Trump.

The size of the unmarried women + minorities + youth coalition, which we first wrote about in 2010 and which keeps expanding with our minorities' population, shows up best in national polls. But the election will be won within individual states, not in a nationwide referendum. 2016 could see Democrats beating Trump by millions, but still losing in the electoral college. [emphasis added]

Here’s our map showing in brown the eleven states that will decide America’s next leader (click map to enlarge):
If you live in the other 40 (including the District of Columbia) states, the ones marked blue or red, your vote won’t really matter this November.

To win, Trump must hold the Romney states of Arizona (57% non-Hispanic white, 26% Mexican-American), Georgia and North Carolina (two “New South” states with transplanted liberals and large African-American populations). North Carolina, which voted for Obama in 2008 and barely for Romney last time, is particularly at risk for Trump.

Trump then must add Florida, Ohio, and Pennsylvania to win. Pennsylvania, the Keystone State, is THE key state, as was Florida to Bush’s win in 2000 and Ohio to Bush’s re-election in 2004. Obama carried all three twice, and Pennsylvania, 79% white and with its western half the anti-Obama Appalachia coal country ripe to turn to Trump, hasn’t voted Republican since 1988.

Right now, Clinton leads Trump in the poll average for all three, though not by large margins. Trump’s anti-Mexican rhetoric is expected to complicate his winning Florida, which is only 57% non-Hispanic white (however, only 3% Mexican-American).

But if Trump loses Florida’s 29 electoral votes, there is one other narrow path to victory. That path accounts for the remaining five battleground states on the map above.

Race will be very important in this election. Forget talk of non-border states Colorado or Nevada being in play. [emphasis added] Their relatively heavy Mexican-American populations will help send them to Clinton. (Arizona is different because like Texas, Arizona's long border with Mexico means whites there hold together in the face of the rising Hispanic immigrant tide.)

The race issue cuts two ways. If race is big and noisy, heavily white states may vote more Republican, more anti-Obama/Clinton than they have in the past. [emphasis added] New Hampshire (92% white) voted for Bush in 2000, and Iowa (88% white) voted Bush in 2004. The two states are targets for Trump pick-ups.

If Trump loses Florida and wins both Iowa and New Hampshire, victory will depend upon picking up two of three upper Midwest states: Michigan (16 electoral votes), Minnesota (10), and Wisconsin (10). Wisconsin is 83% white, Minnesota 82%, Michigan 76%. All are mining or Rust Belt and variously suffering in the Obama economy.

The go-white election strategy won’t win over white but progressive Maine, Vermont, and Oregon, but every other state more white than Michigan is either in Trump’s column or on our battleground list. [emphais added] A smart Trump would focus his campaign on the 11 states our map colors brown.

Evidently, Trump was smarter than we realized.  He won Arizona, Georgia, North Carolina, Florida, Ohio, Pennsylvania, Iowa, Michigan, and Wisconsin, barely lost New Hampshire, lost Minnesota by just 1.5%, and won the election.

No comments: