How far, under ideal circumstances, can the popular vote diverge from the electoral college victor? Or loser?
Under actual circumstances:
Gore won by 500,000 votes nationwide. What is the largest popular vote margin that one can win by and still lose? Under ideal circumstances and real ones. Is it possible to win the country by five million votes and still lose the Electoral College?
Kerry lost the national vote by three million but lost Ohio by only 120,000 votes. If Kerry had won Ohio in '04 (possible, without GOP control of the voting apparatus) while losing the popular vote, would his presidency be legitimate? At what degree of closeness is this decided?
What are the actual mathematical limits? And then, what are the plausibility limits, beyond which predictions fail.
Cheers for a box to think outside of and for an Obama victory so far outside the margin of error they wont even try.