Trump v Romney

[Tetro Kornbluth]

This is a map I made quickly comparing Trump's performance to Romney's four years ago. The blue states are those where Trump overperformed Romney in terms of percentage points. The red states are those where Trump underperformed Romney in percentage points terms. The shading represents a 2 point difference - so the lightest blue shade is where Trump overperformed Romney by 2 points or less, the 40%> Blue shade is 2 to 4 points, 50%> blue is 4 to 6 shade, and so on and the reverse for the 'red states'. This is all using the current figures on this website. I was too lazy to figure out the differences for the Maine and Nebraska Congressional districts. Errors are possible.

A couple of comments:

1) Despite Trump winning and Romney losing, if this were an electoral map 'red' would win comfortably with over 300 EVs. This, more than anything else, really shows the impact third parties had. Don't forget that Stein outperformed Johnson 2012, 2016 was an unusual race.

2) Actually the Utah shading underestimates the massive drop in the vote for the Republican candidate from 2012, but unfortunately Dave's calculator does not have a D 140%> shade

3) Apart from Utah, the areas where Trump dropped the most were the Sun Belt and the Northwest.

4) The state which Trump gained the most from Romney wasn't West Virginia, it was North Dakota... and this despite a strong Johnson vote. What's going on here? EDIT: Actually, no, this is wrong. It was West Virginia, then Iowa, and then North Dakota. This is what I get from relying memory. Speaking of which, Iowa WTH?

5) Yes, despite Trump almost winning Minnesota he actually dropped 0.04 percentage points from Romney, it's just that Third Parties ate into the Clinton vote.

6) lol Chicago

7) Amazingly Minnesota wasn't even the closest 'gap', that would be Arkansas - whose change is less than 1/100th of a percentage point, and required me to put all the figures into excel to calculate the difference.

[Young Conservative]

Some of these definitely aren't right if you did it by margin (as is genuinely calculated and accepted).... how did you calculate this?

[vileplume]

4) The state which Trump gained the most from Romney wasn't West Virginia, it was North Dakota... and this despite a strong Johnson vote. What's going on here? EDIT: Actually, no, this is wrong. It was West Virginia, then Iowa, and then North Dakota. This is what I get from relying memory. Speaking of which, Iowa WTH?

Well the pattern nationally was the rural white Democratic vote collapsed to Trump while he did badly in most big metros. These three states, especially Iowa, had significant white rural democratic votes coming into 2016 and even in the Dakotas the white rural areas were significantly more Democratic than their counterparts in Nebraska and Kansas so Trump had more potential to gain there. None of the three you mentioned have big metro areas either to counter the swing to Trump in the rural areas.

[catographer]

Some of these definitely aren't right if you did it by margin (as is genuinely calculated and accepted).... how did you calculate this?

He did it by vote share. Trump got a smaller share of the vote nationally than Romney: 46.0% to 47.2%.

[Mike67]

What do y'all think the chances of Mitt Romney running as a Third Party Candidate in 2020?

[Tetro Kornbluth]

It is changes in vote share, not margin... you can find those maps elsewhere.

Anyway, I made a series of these (because bored at work), first of all:

Trump vs McCain

Blue wins this map, and the problems we had with Utah continue and now will start to apply to West Virginia. D/R 90%> is anything above 12 percentage points ftr. Of interest, here: Indiana and Alaska, which show the greatest GOP drop in this period outside of Utah mainly down to Palin no longer being on the ballot I would suppose.

Trump v Bush 2004 (i.e. The Last GOP victory)

One thing to note about these four elections: Oklahoma has been ridiculously stable in its GOP vote share. But not even in the Rust Belt did Trump really improve upon W, only in Appalachia did he make big gains. We must not forget that West Virginia is an anomaly.

Trump v Bush 2000

Now this is more interesting, as Bush lost the popular vote too we can get a better picture which isn't so distorted by the national result. A few things: lol Florida, note the DC Metro area, and Vermont really is unusual in New England.

Now, let's get strange...

Trump vs Ford 76

Here there are quite a few states - in the West, in the South and New England where the change is more than 12 points and many more where it is above 10. This looks much more like a modern election map (unsurprisingly) although there are some strange ones, like Nebraska. Also clear that at this point Florida, Texas, and North Carolina were not like the rest of the South. And Virginia too, obviously. Also is Pennsylvania really unambiguously part of the North East? Trump did worse in Wisconsin than Bush either time or Gerald Ford, but unlike them all, he won the state.

And Finally, Nixon 1960

Florida, North Carolina, and Virginia again. Rhode Island also stands out, and Wyoming. All three West Coast states are 12%+ changes against the GOP, as is Maine and Vermont (RIP Alf Landon). DC couldn't vote in 1960, but I'm pretty sure that Nixon would have done better than Trump's 4% - Nixon in 1968 got 18% after all.

[bandg]

Maybe you could use two-party vote share? The high 3rd party vote this year kind of messes with the comparisons. For example, even though Trump got a lower vote share in MN than Romney, it's hard to argue that he actually did worse, when you take into account the 3rd party vote.

[jaichind]

Seem to fit the narrative that Romney was a Western Republican and Trump is an Eastern Republican.