How to read these charts:
|(like Jezebel...||...and CNN)|
-After Donald Trump won the US presidential election last week, a lot of people (see above) have been blaming third-party candidates for tipping the election in his favor. I set out to determine if that was true.
|-There were 10 states in which Trump or Clinton's margin of victory was lower than the total number of votes cast for third-party candidates. 4 went to Trump, 6 to Clinton. Mathematically, these are the only states whose Electoral outcome might have been changed by the votes siphoned off by third-party candidates. They are listed in the left-most column of each tab on this sheet.|
-The two relevant third-party candidates in 2016 were Libertarian Gary Johnson and Green Party's Jill Stein. On each tab, their vote totals by state are color-coded in gold and green, respectively.
-Had there been no third-party candidates to vote for, everyone who voted for Johnson or Stein in real-life would have faced three alternative options: vote for Clinton, vote for Trump, or don't vote at all.
Everything in purple is an assumption about what portion of each group's voters would have taken which option in an alternate universe without Johnson or Stein on their ballot.
-There's no way to know for sure what these third-party voters would have done in this alternate universe, and just like any counterfacutal there's room for disagreement.
|-Nevertheless, we can make reasonable estimates based on what we know about the ideological tendencies of Johnson and Stein's followers. Each tab of this spreadsheet simulates how the election would have turned out under a different set of assumptions regarding Johnson and Stein voters' behavior.|
-Beneath the chart, each tab lists the assumptions on which that simulation is based, as well as the conclusion of that simulation.
|-The bottom line of my findings is that while third-party voters may have influenced the outcome in Michigan and New Hampshire, it is extremely unlikely that they changed the winner of the national election overall, and Johnson's campaign in particualy may actually have diverted enough votes from Trump to enable Clinton to win the popular vote (by plurality).|
|-If you would like to run your own simulations based on your own assumptions, without having to recreate the spreadsheet from scratch, shoot me an email at firstname.lastname@example.org and I'd be happy to send you the original Excel file!|
-All vote totals are as reported on CNN.com as of November 13th, 2016, and all simulations assume that whatever additional votes are tallied after that date would not be significant enough to change anything.