The basic idea is to look at landslide losers in American Presidential elections, and see how much better they could have done under the circumstances.
For example: Al Smith in 1928.
It's pretty much an ASB for Smith to win that election. But he came relatively close to winning New York and Texas.
If we tip the balance in those states over to Smith, he wins 152 electoral votes-more than John W. Davis had four years earlier.
Any other thoughts?