In the USA, registered voters are Democrat, Republican, or Independent.

In the 2016 presidential election, Trump won the traditionally Democratic Rust Belt states of Michigan, Pennsylvania, and Wisconsin by narrow margins, which effectively decided the election by just 107,000 votes.

Did he win those states because of significant differences between the turnout of registered Democrats or Republicans, or because an unexpected number of Democrats voted for him, or because an unexpected number of Independents voted for him, or some combination of those?

In other words, for those three states, are there statistics showing, for each of the three party categories: the number of registered voters, the number of actual voters (turnout), and (most importantly) how many voted against their party affiliation? Was the biggest swing because of Democrats voting for Trump, or was that insignificant compared to the number of Independents voting for Trump?

  • 4
    In 2016, the Democratic vote stayed roughly flat nationally while the Republican votes rose by 2 million. It was less than 107,000 votes. And, WI and MI don't have party registration. Turnout is a key factor. Aug 31, 2020 at 15:03
  • I think mentioning national results distracts from the focus of the question. I think the 107k was true for the 3 states and those 3 alone could be considered a swing cause. For WI/MI, treat those voters as independent, assume their 2012 votes were for their preferred party, and compare with their 2016 results to get a sense of how many flipped.
    – joe snyder
    Aug 31, 2020 at 16:01
  • 1
    77.7k, not 107k. Trump got more republicans to vote for him than Mitt, when the nomination of a rich and proper man was felt to be out of touch.
    – dandavis
    Aug 31, 2020 at 19:14
  • Would voter disenfranchisement and/or suppression by Party B, aimed at demographics traditionally friendly to Party A, be considered Party A causing the result by not turning out, or Party B causing the result by suppressing turnout? Sep 8, 2020 at 14:02

2 Answers 2


Democrat voters switched to 3rd-party candidates

Note that the figures online are very erratic. For example, census.gov says 2016 had 4,713,000 Michigan voters (to nearest 1,000), but michigan.gov says 4,874,619, and wikipedia says 4,799,284. The counts of registered voters are similarly inconsistent, and michigan.gov even shows a 63% turnout while their actual numbers compute as 65%.

Nevertheless, the swing in votes for parties (as opposed to turnout) appear to be the biggest factor. I didn't have the patience to track down party affiliation figures, but for a quick and dirty comparison let's assume voters voted for their preferred party in 2012. Using wikipedia figures we have:

                  2012                           2016
        Votes   Obama  Romney Other     Votes   Clinton Trump Other
Mich  4,730,961 54.21% 44.71% 1.08%   4,799,284 47.27% 47.50% 5.23%
Penn  5,753,670 51.97% 46.59% 1.44%   6,165,478 47.46% 48.18% 4.36%
Wisc  3,068,434 52.83% 45.89% 1.28%   2,976,150 46.45% 47.22% 6.33%

In Michigan, the Democrat got about 296,000 fewer votes in 2016 than 2012, but the Republican only gained about 164,000. Pennsylvania and Wisconsin were similar. It appears a lot of former Democrat voters switched to Other, which hurt Clinton enough to give the win to Trump.

  • 1
    Polling shows third party voters did not vote for Clinton because they didn't like her personality. That swung the election enough for Trump to win electoral college. Sep 2, 2020 at 14:45
  • 1
    Is there any indication that these "other" voters were otherwise Democratic in affiliation? Seems like Trump drove a lot of GOP voters to vote Libertarian, but he also attracted a very different demographic (anti-establishment) to turn out, while the problem with Clinton was more getting the vote to turn out. 2012 --> 2016, Libertarian votes were up by over 3M, while Green Party was up by less than 1M. That seems to indicate the increase in "other" was coming more from the 2012 GOP side than the Dems. Sep 2, 2020 at 16:30
  • @PoloHoleSet I think it doesn't matter who contributed the most to Other, only that there were enough Dem defectors to drop Clinton's % by so much. Obama Dems either didn't vote, went to Trump, or went to Other. Although Trump outperformed Romney in the 3 states, Other growth was bigger. The biggest delta was the Clinton drop, and the next biggest was Other growth, so it would seem the simple explanation is one helped feed the other. The Wiki quote and links by Joe Boost address the affiliation %, though what happened nationally vs. the 3 states in questions might be 2 different things.
    – joe snyder
    Sep 3, 2020 at 1:42
  • @joesnyder - But if the Dem defectors subtracting from the Clinton vote are outnumbered 3 to 1 by GOP defectors subtracting from the Trump vote, then, clearly third-party defections are not the determining factor in the GOP winning. In your own comment, you are talking about other factors that played a much more prominent role, but your question does put the onus on that particular factor, which would be far, far down the list. You can't only look at third party defections from one candidate and claim it's a factor while ignoring them for the other. Sep 8, 2020 at 14:00
  • 1
    @WitnessProtectionID44583292 - Of course they do. Libertarian and Constitutional parties draw from traditionally aligned GOP voters just like Green Party draws from traditionally Democratic voters. If Trump turned out more voters who traditionally don't vote, or didn't for Romney in 2012, super duper, but THIS answer is looking at the uptick in third party numbers, and the third parties that draw from what is generally considered the GOP pool showed much bigger increase in votes than the Green Party. If the answer claimed a multi-variate cause, it wouldn't be an issue, but it doesn't. Sep 9, 2020 at 19:11

Descriptions of party hopping tabulated in joe snyder's answer can be found in places such as these:

Wikipedia's page on Obama–Trump voters states:

Obama-Trump voters [...] are people who voted for Democratic Party nominee Barack Obama in the 2008 or 2012 presidential elections (or both), but later voted for Republican Party nominee Donald Trump in 2016: these voters comprise 13% of Trump voters and 9% of Obama voters. In contrast, 7% of Obama voters did not vote and 3% voted for a third party candidate. While some analysts consider them to have been decisive in Trump's victory, others have disputed this conclusion.

You must log in to answer this question.

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