An interesting look at not just the 2016 election, but also at roughly 60 years of U.S. presidential elections.
We know that in 2016, Trump won 46% of the popular vote and Hillary won 48%.

But this graph looks not just at the ratio of those who voted, but their smaller ratio as a subset of the total population, in elections back to 1960.




That graph could include the total population (including ineligible children and non-citizen legal immigrants, of which my stepmother was, from Germany, she lived here 60 years but never felt a need to become a U.S. citizen and vote).
Or it more likely is the total adult and U.S. citizen population eligible to vote. Which I assume it to be.

But either way, that the election victory in any election could be vastly changed if everyone eligible voted. Our presidents over 60 years represent, at most, the choice of 38% of all citizens. And generally the choice of 26% to 31% of all citizens.

40 to 48% of the public in election after election don't participate. That amount participating could elect a third alternative candidate by a landslide. In 1992, 19% voted for Ross Perot, which inflated participation that year to 64% of the public.

Perot died a few days ago. The deficit and free trade issues he pressed in 1992 and 1996, along with the issues of sovereignty, free trade, offshoring of jobs and border security raised by Pat Buchanan in 1992, 1996 and 2000, are the issues that Trump championed and won on in 2016. Trump's presidency is the enduring legacy of both Perot and Buchanan.