This is my second post on this matter. The first post is here.
In 2014, the Washington post extrapolated state Presidential voting trends from 2000 to 2012 (I regularly consult this link for reference).
It showed all states as either solidly Red or Blue, except for the following swing states (the combination of all three, plus the states Romney won, is sufficient for a Republican to win the general election):
North Carolina: .165 percentage points more Republican than the projected 2020 popular vote.
Pennsylvania: .387 percentage points more Democratic than the projected 2020 popular vote.
Ohio: .427 percentage points more Democratic than the projected 2020 popular vote.
Yes, the extrapolations show Ohio as more Democratic than Pennsylvania, due to PA being filled with more coal mines.
The closest Slippery Small Swing States that were previously necessary for a Republican to win the Presidency are
Iowa, which is 3.875 points more Democratic than the projected 2020 popular vote.
New Hampshire (due to its anti-Bush swing in 2004) is 4.686 points more Democratic than the projected 2020 popular vote.
Virginia will be squarely out of reach for any Republican in the future except in case of severe recession or unpopular war under a Democratic president. This will be due to urbanization around D.C., as well as the expansion of the franchise for felons.
Bush used New Hampshire (plus a bunch of other states that will never vote GOP again, such as Nevada and Colorado) to win the election of 2000. He used Iowa (plus New Mexico, which will definitely never vote GOP again) to win re-election in 2004.
So these are the possible Trump victory maps for 2016:
Trump could either win Virginia plus New Hampshire or Pennsylvania. He must win Ohio and Florida, in any case.
None of these are implausible. Trump won more votes than Clinton in the Ohio primary, where Republicans won more primary votes than Democrats. He also won more votes than Clinton in the New Hampshire primary, where Republicans won more primary votes than Democrats. Trump and Republicans narrowly won fewer votes than Clinton and the Democrats in Pennsylvania, but, perhaps, as in Louisiana, that was due to the closed primary there, and the Democrat party registration advantage.
I predict Trump will win New Hampshire and Pennsylvania, but lose Virginia. Trump is the Light of the East. His best primary performance so far to date was in Rhode Island, a state he’ll certainly lose in the general.
Trump said he’d win Pennsylvania (reasonable) and put Michigan and New York into play (impossible). Trump likes to exaggerate expectations in hopes that they will become reality.
In any case, the vast majority of Trump’s advertising and campaigning must be concentrated in Ohio and Pennsylvania. Maybe leave some for North Carolina, Indiana, and Florida. Virginia is a sixth priority.
I call the strategy focused on winning Florida, Pennsylvania, and Ohio the three-state strategy. As you can see from the Washington Post article, on its face, it’s clearly the best one.