Making a Bad System Even Worse

The Electoral College is stupid. I think we can all agree on that. It’s an antiquated system for picking our president that hasn’t made any sense since about 1805. Because of our insistence on following what our founding fathers said to the letter or because we’re too lazy to do anything about it, a president can be elected while getting less overall votes than his opponent. Sounds like democracy to me. When the Electoral College was created, the only people who could vote were white men. That changed, thank god, but we still use a system that’s based on a fundamental mistrust of voters.

Railing on the Electoral College is fun, but it also presents some interesting possibilities to do some goofy and nerdy things with election results. This month, the primary focus of the nation is March Madness. At least it’s my main focus. I don’t know about you. I did some research last week on which states were the best at the tournament, both with respect to how many schools they’ve sent and how many wins those schools have accumulated.



In my mind, the next step was to use that information to make my very own Electoral College. So that’s what I did. I used the school and wins data to run the last four presidential elections.

As you can see, the graphic above shows how the 2012 Presidential Election between President Obama and Mitt Romney would have turned out if, instead of Electoral College votes being based on population (number of senators plus representatives), they were based on how many teams each state has sent to the tournament in the history of the contest. The President still carries this election rather easily. Maybe if Mitt had a binder full of basketball players, he’d have done a little better.

Number of schools might not be the best measurement, though. What really matters in sports is winning. There could be a state like Louisiana that’s sent 11 schools to the tournament, but those 11 teams only have 36 wins. Here’s the 2000 Bush vs Gore election with electoral votes based on wins. The numbers are a lot higher, but it’s basically the same idea.

In this case, President Bush wins by a much, much larger margin than he did with the actual Electoral College. In fact, over the course of simulating five elections, including the 200o election twice, I noticed some patterns develop.

Apparently, electing a president based on college basketball has a pretty strong Republican bias. Well, basically, a North Carolina bias, and North Carolina has gone Republican in three out of the last four elections. The only exception is President Obama’s victory over John McCain in 2008. North Carolina has always been a basketball powerhouse. Duke and UNC are both perennial national title contenders, and NC State and Wake Forest have each had periods in their history where they’ve been extremely competitive. NC State even won a couple of national championships. There are also about 50 little schools in the Tar Heel state that win the Big South or whatever and end up in the tournament. As a result, they’ve sent 14 teams who’ve won 290 games, a much larger proportion of total electoral votes than North Carolina accounts for in the actual election.

The fact is, most of the states that are overrepresented in basketball generally vote Republican. Winning Kentucky when we're basing the vote on wins is worth 4.8 times the votes of the population based electoral college. Eight out of the top ten states overrepresented by the schools metric voted republican in at least three out of the last four elections. Same with six out of the top seven states overrepresented by wins. This helps explain why Obama beat Romney rather handily in the actual election, but the wins vote was extremely close. It also explains why the Bush vs Gore election was so close, but Bush dominates the basketball vote, even when I pretended the Supreme Court hadn’t elected him in 2000 and gave Florida to Gore. It’s worth mentioning that Gore won the popular vote but lost the election as well.

The bottom line is that this was a fun exercise that I spent entirely too much time on. It’s also clear that whoever gets the Republican nomination should wish that some constitutional scholar sees this blog post and decides that basing our elections on basketball is a good idea.