Certainly the notion of eliminating the Electoral College and electing the President via a national popular vote is an interesting approach. It would eliminate some portion of state-by-state campaigning, which has both pros and cons. But I wanted to take a different look at the issue, and asked myself the questions: What if we kept the Electoral College, and how could we make it better?
In order to find a middle ground of sorts between the current Electoral College and a national popular vote, I decided to create a new Electoral College formula. This formula has three guiding principles:
- Maintain the basic state-by-state structure of the current Electoral College and the 538 electoral votes.
- Establish a more proportional structure for the electoral votes so that they more closely resemble each state's popular vote.
- Maintain a small incentive for a state "winner".
(It's worth noting that I came up with this idea and the resulting formulas before looking at any data.)
In the end my formula was fairly simple:
- First, allocate one electoral vote in each state to the overall winner (measured by popular vote) of the state.
- Second, allocate the remaining electoral votes in that state to each candidate based on their percentage of the popular vote.
For example, the state of Wisconsin has 10 electoral votes. As such one vote would be allocated for the overall winner and nine votes would be allocated proportionally. If the popular vote in Wisconsin for the 2012 Presidential election was 55.5% for Obama and 44.4% for Romney, then Obama would receive 6 electoral votes (5 votes of the 9 based on his popular vote percentage + 1 vote for being the state's winning candidate) and Romney would receive 4 electoral votes (4 votes of the 9 based on his popular vote percentage). Obviously this outcome is significantly different than the winner-take-all approach in 2012 election where Obama won the popular vote in Wisconsin and collected all 10 electoral votes.
- Using the new Electoral College formula results in a 2012 Presidential election where Obama's margin of victory is significantly more representative of the national popular vote. In fact, they would be almost exactly the same except for the popular votes that went to 3rd party candidates.
- Large states look very different using the new Electoral College formula. For example, California isn't quite as "blue" and Texas isn't quite as "red".
You can view and interact with all the results here. This area also includes more detail around my assumptions and formulas. Time permitting (which unfortunately is not likely) I would like to add the time dimension to this analysis by including data for the previous 3-4 elections. I think it would be interesting to see how the states change from one election to the next using the same formula for calculating electoral votes. If you have any suggestions/questions/comments please include them in the Comments section below.