A bad argument against the Electoral College is that because it is sometimes out of step with the “popular vote,” the “real winner” sometimes ends up “losing.” This is a bad argument for many reasons, but especially because presidential elections are not fought for the “popular vote” total, so it is extremely silly to draw any meaningful conclusions from it.
The largest ten states in America are California, Texas, Florida, New York, Pennsylvania, Illinois, Ohio, Georgia, North Carolina, and Michigan. Of these, Florida, Pennsylvania, Ohio, Georgia, North Carolina, and Michigan are competitive in presidential elections; Texas is sort of competitive (increasingly so, especially this year); and California, New York, and Illinois are not only forgone conclusions, but effectively one-party states. In practice, this means that Republicans voters in three of the six largest states have little incentive to vote, while Democrats in seven of the ten have a real incentive to vote. The “popular vote” reflects this.
A good argument against the Electoral College is that this varied incentive structure is itself a problem. I don’t agree with this — or, rather, I do, but I don’t find it persuasive enough to turn me against the Electoral College on balance — but I think it’s an argument worth reckoning with. “If you change the rules and assume a game that wasn’t actually played, my candidate won” is not.