Some interesting PV/EV differences (not counting Jackson/Adams).

[Mr. Smith]

Firstly. All the true winners were Democrats, while the system-gamers were Republicans

1876:

1. Tilden loses by less than 10 EV
2. Tilden tops the PV by more than 1%

1888

1. Cleveland loses by over 50 EV
2. Cleveland tops the PV by less than 1%

2000

1. Gore loses by less than 10 EV
2. Gore tops the PV by less than 1%

2016

1. Clinton loses by over 50 EV
2. Clinton tops the PV by more than 1%


...WTF...

[Bismarck]

Firstly. All the true winners were Democrats, while the system-gamers were Republicans

1876:

1. Tilden loses by less than 10 EV
2. Tilden tops the PV by more than 1%

1888

1. Cleveland loses by over 50 EV
2. Cleveland tops the PV by less than 1%

2000

1. Gore loses by less than 10 EV
2. Gore tops the PV by less than 1%

2016

1. Clinton loses by over 50 EV
2. Clinton tops the PV by more than 1%


...WTF...

Would you have preferred in the 1880s to let the south decide presidential elections while the rest of the country votes the other way?

[jaichind]

What is interesting is that 1876 1888 and 2000 a flip of one state would have shifted the election winner.  In 2016 no shift of any state would have given the election to Clinton (assuming MI goes to Trump.)

[Xing]

And that's why we're going to have a hard time getting rid of the electoral college. The party that has only benefited from this system is not likely to want to do away with it. I think we'll need a Democrat to win the EC while losing the PV before more Republicans start considering getting rid of it.

[jaichind]

And that's why we're going to have a hard time getting rid of the electoral college. The party that has only benefited from this system is not likely to want to do away with it. I think we'll need a Democrat to win the EC while losing the PV before more Republicans start considering getting rid of it.

There is an argument that JFK lost the popular vote in 1960 so in theory the GOP has in theory been a victim of the PV/EC split going against them.

http://www.realclearpolitics.com/articles/2012/10/19/did_jfk_lose_the_popular_vote_115833.html

[Mr. Smith]

The more concerning thing is that there isn't a consistent pattern. If all of them had turned out like 2000, this wouldn't be so bad.

If all of them had turned out like 2016, this'd be even more egregious.

But no, all colors of the rainbow have happened.

[Mr.Phips]

The reason for the split in 1876 and 1888 is likely that black voters were prevented from voting in most of the South (other than those 3 states in 1876). The percentage differences in the national popular vote were small enough that this could have made the difference, as the Republican Party was the choice for pretty much all black voters. Plus, Democrats barely lost important Northern states.

It's also worth noting that Lincoln would have been toast in a national PV runoff with Douglas, so the EC at least has that going for it.  On the other hand, Wilson (yuge net negative for the country IMO) would have been toast in a national PV runoff with Teddy Roosevelt.

And the EV/PV split in Kennedy's favor is a real thing using the most mathematically correct way of allocating votes in Alabama.  I don't know why historians have allowed that one to get memory holed.

?  Kennedy won Alabama by nearly 100K votes.  

[/]

What is interesting is that 1876 1888 and 2000 a flip of one state would have shifted the election winner.  In 2016 no shift of any state would have given the election to Clinton (assuming MI goes to Trump.)

Texas would have. And it was decided by less than 10% with a massive swing towards Clinton.

[catographer]

2016 is definitely the largest discrepancy in national popular vote and electoral college winner margins. Trump got more EVs than any Republican since 1988, yet he performed worse than the last two times the GOP won the election (2000 and 2004). Trump got a smaller percentage of the vote than Romney, and apparently only <1% more than McCain.

It appears as though the GOP has a different range of typical support than the Democrats.
In elections since 2000, they've received 47.9%, 50.7%, 45.7%, 47.2%, 46.4%
Whereas for Democrats, they've received 48.4%, 48.3%, 52.9%, 51.1%, 47.9%

[jaichind]

It seems that Clinton is on track to win by 1.7%.  Since Trump won PA by at bit less than 1.2% then if we assume a national uniform swing Clinton would have to win by 2.9% in the PV to win the EC overall.  If this that means achieved Trump achieved the largest EC advantage relative to PV since 1948 where in theory Dewey could have won the EC if he kept Truman's PV margin to 3.6% assuming national uniform swing.  

This thread which I bookmarked a while ago contains the data

https://uselectionatlas.org/FORUM/index.php?topic=167153.0

[○∙◄☻¥tπ[╪AV┼cVê└]

You know which one stands out? Gore, he wasn't a New York Democrat.

[SingingAnalyst]

Firstly. All the true winners were Democrats, while the system-gamers were Republicans

1876:

1. Tilden loses by less than 10 EV
2. Tilden tops the PV by more than 1%

1888

1. Cleveland loses by over 50 EV
2. Cleveland tops the PV by less than 1%

2000

1. Gore loses by less than 10 EV
2. Gore tops the PV by less than 1%

2016

1. Clinton loses by over 50 EV
2. Clinton tops the PV by more than 1%


...WTF...

Of course, as we now know, Clinton won the PV by 2.09% and still lost.

Not only has the GOP benefited all four times (beginning in 1876) from the PV/EC discrepancy, there were three other elections in which it "almost" happened (that is, just a few thousand well-placed vote flips would have given the election to the PV loser): 1916, 1948, and 1976-- and in all three of those cases, the GOP would have benefited. One has to go all the way back to 1884 to find an election in which a similarly small number of vote flips would have given the Dems the election despite their losing the PV.

This is all the more remarkable considering how the relative strengths of the parties, by region and urbanization, have changed so much over time.