I read in two sources that the 2016 popular vote was Democratic by 2.09 percentage points (48.18% Clinton, 46.09% Trump). I also read it was by 2.22 percentage points (51.11% Clinton, 48.89% Trump). I assume the second one is the two-party vote. Is that correct?

  • 2
    As an aside, I can't help but read any such "won by X percentage points" in comparison to how one tends to measure "games behind" in sports. The latter being closer to "how much of the opponent's win/losses [votes] have to be converted into the other's in order to make them equal". And that's always struck me as the real margin of victory: how much you could have afforded to lose before victory is lost. Of course third parties make that a bit weird, but if we look at the 2-party votes only it just halves things, so a 1.11 percentage points margin for Clinton. Jun 19, 2020 at 6:28
  • 1
    @zibadawatimmy: Note that the "games behind" measurement is something I have never encountered outside of US sports, while percentage points are found in many other contexts.
    – DevSolar
    Jun 19, 2020 at 9:56
  • 1
    @zibadawatimmy If everyone who was eligible always voted then that usage would make more sense, but that's not the case. You could have gotten people who didn't vote to vote for you, not only people who voted against you. Jun 19, 2020 at 20:48
  • @CaptainMan That's equivalent to the third party problem I mentioned, yes. Jun 20, 2020 at 7:53

2 Answers 2


Yes - the official totals according to the FEC were:

Trump - 62,984,828 (46.09% of all 136,669,276 votes)

Clinton - 65,853,514 (48.18% of all 136,669,276 votes)

So in terms of difference in popular vote percentage, this gives us your first figure, 2.09 points. However, in terms of the percentage of the votes given to either of the two main parties, this gives us:

Trump - 62,984,828 out of 128,838,342 = 48.89%

Clinton -  65,853,514 out of 128,838,342 = 51.11%

Calculating the difference between these percentages gives us your second figure of 2.22 points.

  • 4
    So the difference is whether or not you count votes given to 3rd party candidates?
    – Philipp
    Jun 18, 2020 at 15:26
  • 14
    @Philipp not quite - Nevada also includes a 'None of These Candidates' option which is counted in the FEC's overall vote count.
    – CDJB
    Jun 18, 2020 at 15:28
  • 2
    Yes it is based on whether or not we count 3rd parties. With 3rd parties it's 2.1. With third parties factored out, it is 2.2. Jun 18, 2020 at 17:01
  • 2
    @CDJB I wasn't aware of that option in Nevada and I thought it would be an interesting bit of data to look at the idea that both candidates were really bad in 2016. Sure enough 'none' increased from 0.57% in 2012 to 2.56% in 2016. Gary Johnson also more than quadrupled his vote share from 1.08% to 4.61%.
    – Eric Nolan
    Jun 19, 2020 at 10:35
  • 2
    Depending on what one wants to spin, one can get even more different values by basing the percentages on a) votes for one of the two major parties (as above), b) valid votes for any party c) votes including invalid votes and/or "none of the above" votes d) number of registerd voters e) number of eligible voters. -- IMHO, d) and e) are irrelevant unless on e wants to point out specifically low turnout; b) and c) have their justtifications, wheras a) would be considered improper in any other country, but has its justification in the US where other parties are practically non-existent [cont] Jun 19, 2020 at 12:08

Vote totals for 2016 are

                Donald Trump   Hillary Clinton
Popular vote      62,984,828        65,853,514
Percentage             46.1%             48.2%

Which gives Clinton a 2.1% advantage.

  • 7
    This is imprecise. The difference between 46.1% and 48.2% is 2.1 percentage points, not percent.
    – GManNickG
    Jun 19, 2020 at 18:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .