Quote:Almost always???
Again, I'd like to see two electoral votes subtracted from each state and the District of Columbia so that the college better reflects the population. This would make it less likely that a candidate could win without a plurality of the popular vote -- but it wouldn't make it impossible.
A winning candidate could eek out slim pluralities in a bunch of states while the losing candidate rolls up huge majorities in a few big states.
On Nate's question: This has happened at least three times before:
-- 1888, when Benjamin Harrison (5,439,853) beat Grover Cleveland (5,540,309).
-- 1876, when Rutherford B. Hayes (4,036,298) beat Samuel J. Tilden (4,300,590). This result was widely seen as crooked. Four states' electoral votes were in dispute. Boiled down, Congress gave them all to Hayes, the Republican, in exchange for an agreement to end Reconstruction. As a result, Harrison won by one electoral vote.
-- 1824, when John Quincy Adams (108,740) beat Andrew Jackson (153,544), his main opponent. (This election was decided by the House of Representatives because no candidate could get a majority of the electoral vote. Jackson actually received more electoral votes).
No records of the popular vote were kept before 1824, according to the National Archives.