The NY Times is wrong and here is why: elections don't happen in a vacuum. [View all]

[Drunken Irishman]

It's easy to suggest that, because of the Electoral College, Trump could lose by an even wider-margin in the national popular vote than he did in 2016 and still exceed the total Electoral Votes he won in 2016 - but this is unlikely.

Tthe general election is not, no matter what anyone tells you, 50 miniature campaigns. These localized electoral elections do not happen in a vacuum. The fact is, yes, Hillary won the popular vote by about two-points nationally - however, had she equaled Obama's popular vote margin over Romney in 2012, she is likely the president right now.

Here is why:

Again, I point to the fact that the election doesn't happen in a vacuum. Nearly almost every state will follow the national trend. There might be exceptions (more due to a candidate being from there that may drive up a larger than expected vote) but on the whole, the trend swings in conservative states and liberal states and there's no further evidence of this than margins.

Let's start by looking at the most Democratic states from 2012. Those states, where Obama won the highest amount of support, were (along with his percentage of support):

Hawaii (70.55%)
Vermont (66.57%)
New York (63.35%)
Rhode Island (62.70)
Maryland (61.97%)
Massachusetts (60.65%)
California (60.24%)

Those seven states (and you can throw in DC, which voted Obama with 90% of the vote) gave Obama 60% or greater in the 2012 election. Now compare that to 2016's results for Hillary in these states:

Hawaii (62.22%)
California (61.73%)
Maryland (60.33%)
Massachusetts (60.01%)

There were only four states Hillary won in 2016 with 60% or greater of the vote (compared to seven for Obama). Of those four states, Hillary only did better than Obama in one: California.

So, what about the three other states Obama carried with 60% or greater of the popular vote - how'd Hillary do?

New York (59.01%)
Vermont (56.68%)
Rhode Island (54.41%)

Despite being from New York, Hillary actually did worse all around than Obama there - by roughly four-points. That's a lot. She did nine-points worse than Obama in Rhode Island and roughly six-points worse than Obama in Vermont.

I hope it's becoming clear what my point is here: None of these states acted independently. Only California of those listed saw an increase - every other state saw a decline in support from 2012 to 2016 for the Democratic nominee. Even heavily Democratic states.

Why? Because, again, these elections aren't independent of each other. They're all still connected. So, if Hillary sees a four-point loss of support in New York, it almost reasons other states would see a similar shift. These weren't minor shifts, either. Hillary lost nearly two-points off Obama's total in Maryland. In New Jersey, which Obama won by with 58.38% of the vote in 2012, Hillary won with 54.99% - basically, rounded, a three-point decrease. It's not significant on its own, because these are blue states. She could absorb those margins and still win their electoral votes.

But what about closer states?

Obama won Pennsylvania with 51.97% in 2012. Hillary only managed 47.46%. That's a decrease of 4.5 points from Obama's total. That decrease seems fairly understandable now, right, in the context of some of these blue states where Hillary performed, at times, four-plus points worse than Obama.

Maybe there's no better example of this than Maine, where Obama did nearly ten-points better than Hillary. Maine has been a reliable Democratic state and likely will remain one but it followed the national mood and the tight, tight popular vote margins.

Of course, the margins shift depending on the state. Some saw larger shifts than others but the constant is that the shift happened across the board. It wasn't just on the coasts, as we've shown Hillary actually did worse than Obama on the East Coast, and while Hillary did marginally better in California than Obama, in 2012, Obama won Washington with 56.16% of the vote, compared to 52.54% for Hillary. In Oregon, Obama won it with 54.24% and Hillary carried it with 50.07%.

So, knowing this, knowing that the shift was pretty much across most every state from 2012 to 2016, it reasons that if there's a shift back to the Democrats that it will take place across the country - and won't be isolated just to the coastal states or the Democratic strongholds. That's the fallacy of the argument. It assumes that what makes Trump unpopular in Oregon isn't there for people in Pennsylvania when it is. Sure, Trump is far less popular in California than he is on Pennsylvania but he's still fairly unpopular in Pennsylvania, too.

To be sure, this doesn't guarantee anything. 2016 was so narrow that it literally came down to just three states. If the national polls indicate a very tight margin again, then absolutely Trump could manage to expand his grip on the map (most likely by winning New Hampshire, which was very narrowly won by Hillary in 2016). But that's about it. Of course, it doesn't matter if it's by 308 or 304, a win is a win. But if his national support, which is well underwater now, shifts even marginally worse for him, or even remains as it is right now, he's likely to lose Pennsylvania, Michigan and Wisconsin and therefore the presidency.

Because that's how it works. Wisconsin might be more conservative than Vermont but the same rules apply here. If Vermont shifts even marginally back toward the Democrats in 2020, it reasons so will Wisconsin because that's typically the nature of the game. And with the margins being so close in Wisconsin, Michigan and Pennsylvania in 2016, even a very slight shift should be enough.