Actually, the overall vote difference is supposedly 300,000. That number represents one quarter of one percent of the 120 million votes cast. I cannot believe that the way votes are counted produces accuracy to that level.

If we used the popular vote to decide presidential election we would have recount after recount in this case until it ended up in the courts who would probably decide it to be a tie. In that case, congress would elect the president.

If you think about it only one eighth of one percent of the votes plus one would have to be flipped in order to turn the results around.

Remember 2000 when we recounted the Florida votes. That was about six million votes. Can you imagine recounting 120 million votes?