If Clinton wins the presidency by the current margin the polls are giving her right now, yes the House falls.
Not necessarily. See: 1972. Even 1964 and 1984 yielded unimpressive congressional gains for the winning party relative to the size of their margins in the national popular vote for president. I doubt that even with a popular vote margin of 10% or more for Clinton would be enough to overcome the structural Republican advantage in the House.
That was back when split ticket voting was far more common then it is today though.
Could those conditions be coming back?
1964 was a year where a significant number of GOP officeholders openly declined to endorse Goldwater (although few endorsed LBJ). It was also a year where a number of elected Southern Democratic officeholders endorsed Goldwater, including Rep. John Bell Williams (D-MS), Rep. William Colmer (D-MS), and Rep. Albert Watson (D-SC). Watson switched to the GOP in 1965 after the Democratic caucus stripped him of his seniority.
There is a correlation between ticket-splitting by voters and party-bolting by elected officials in that party.
