That's an excellent observation; I never thought about that. I keep meaning to sit down with some election datasets and experiment with different reforms to see if anything might have made a difference in certain close elections.

For example, as a *very* rough (but easily calculated) approximation of what you're talking about, we can imagine what would happen if we just removed the +2 spotting of electors each state gets because of the Senate (which would require a Constitutional amendment). While this would not have changed the outcome of the 2016 election, it would have altered the 2000 election, giving Al Gore a 225-211 Electoral College victory (and maybe then, who knows, we never get Trump 2016).