Taking a look at the rationales for the electoral college, it seems pretty messy on the surface, and the intended benefits are not all that clear or relevant today, in my opinion.

So I decided to take an empirical approach to see what the functional results were in the five cases where the electoral vote prevailed over the popular vote. One would assume that if there is a positive side to the electoral vote, then it should make an appearance in the quality of the presidents that were not winners of the popular vote.

Historical rankings of U.S. presidents

1824: John Quincy Adams (D/R)- Adams is an outlier due to the system being different at that time; he was actually chosen by Congress out of four candidates, none of whom won a majority of the popular vote or the required number of electoral votes.

1876: Rutherford B. Hayes (R) - ranks solidly within the 3rd quartile

1888: Benjamin Harrison (R) - ranks solidly within the 3rd quartile

2000: George W. Bush (R) - ranks 3rd-4th quartile

2016: Donald Trump (R) - ranks solidly within the 4th quartile

According to this information, the evidence indicates that the electoral college has never elected an average, or above average president. There also appears to be a downward trend in the quality of "popular vote loser" presidents.

You never change things by fighting the existing reality.
To change something, build a new model that makes the old model obsolete.
R. Buckminster Fuller