Week 2: Plurality and "Blue Moon Journal" Article

I dedicated the beginning of this week to doing more of an overview of the various voting systems using "The Mathematics of Elections and Voting." I studied plurality, instant runoff (also known as the Hare method), the Coombs rule, Borda count and other pointscore methods, and the Condorcet method. I will give a short summary of each of these systems and share a few questions that came up as I was reading.

Plurality is fairly straightforward: the candidate with the most votes compared to the others wins the election. I will go into more detail about plurality below, as this was what I have been focusing my research on for the past few days.

Instant runoff utilizes the preferential voting system, meaning that in this method, voters are required to provide a preference list of the candidates in order from their favorite to their least favorite. Once the totals are tallied for each candidate using the preference lists, the candidate with the fewest first place votes is eliminated. Then, the votes are tabulated again as if the eliminated candidate had not existed, and once again the candidate with the fewest first place votes is eliminated from the competition. This continues until two candidates remain, and the overall winner is decided by a majority vote. The Coombs rule is an alternative to the Hare method; essentially, the candidates with the greatest number of last place votes are eliminated one at a time.

As referenced by their name, pointscore methods set a fixed number of points for each candidate (example: if there are 3 competitors, a voter would give their favorite candidate 3 points, their second favorite candidate 2 points, and their least favorite 1 point. All the points given to each candidate would then be added and the person with the most number of points wins.) When the points go in uniform steps, the method is called a Borda count. There are instances, though, when the scales aren't uniform, such as the one used in the Indy Racing League.

The last method I studied this past week was the Condorcet method and Condorcet's extended method. For the Condorcet method, you conduct "runoff" elections between pairs of candidates. For example, if there are three candidates (A, B, C), there would be three competitions: A vs. B, A vs. C, and B vs. C. The candidate who wins is the one who beats the others at the runoffs. Say A beats B 12-8, B beats C 7-5, A beats C 16-10. A would then be the Condorcet winner. There are situations where the Condorcet method does not yield results (example from"The Mathematics of Elections and Voting": A beats B 8-4. B beats C 9-3, C beats A 7-5. This is a cyclical scenario and there is no Condorcet winner.) In these cases, Condorcet's extended method must be used. In order to save the in-depth description for the post in which I focus on the Condorcet method, for now I will just say that you rank the paired runoffs based on the gap of votes won. So, the winner who beat another candidate by the biggest majority is ranked first, and so on. Apologies if this doesn't make sense! Look out for the post specifically on the Condorcet method in the upcoming weeks to learn more!

Some questions I thought of as I was doing my research (which I hope to answer by the end of this project) are:

1. Is it fair to change voting methods at different stages of an election?

2. What is the rationale behind non-uniform point methods?

3. How can you prove that the same winner will come about in both the Condorcet method and Condorcet's extended method?

4. In what ways are the Hare method and the Coombs rule different? Is one more fair than the other?

This week, I have been focusing on plurality research. The method is pretty simple, so I didn't find a lot of novel information. It is the most widely-used voting system in the world--it's efficient, convenient, and not as costly--but has many flaws. For example, it fails to acknowledge the true popularity of the candidates because voters do not provide a preference list. Say there are three candidates (A, B, C) and the following rankings are true:

30 people: A, B, C

20 people: B, C, A

18 people: C, B, A

Candidate A would win by the plurality method, but in reality Candidate B is the most popular candidate.

Another drawback of the plurality method is that some voters may feel pressured to vote for more popular candidates or not vote at all if they prefer less popular candidates because they might feel as if their vote doesn’t matter. The more candidates involved, the more confusing things can get. The spoiler effect (https://en.wikipedia.org/wiki/Spoiler_effect) can really mess things up for candidates of the same party under the plurality method. If there are 2 candidates, one Republican and one Democrat, who currently are splitting the vote 40% and 60%, respectively, but another Democrat candidate joins the race and splits the 60%, the majority vote could now go to the Republican. This situation is similar to the one that occurred in 2000, where Green Party candidate Ralph Nader "took" votes away from Al Gore, and George W. Bush in turn won the presidency.

I am currently working on an article on plurality for "The Blue Moon Journal," a student-run STEM journal. I plan on explaining what plurality is, where it is used, its drawbacks, and alternative methods.

My mentor also recently sent me this link about the recent Third Congressional District's Democratic primary election in Massachusetts, in which Dan Koh lost to Lori Trahan by a mere 145 votes after a recount. Advocates of ranked choice voting (providing preference lists) are now pushing for a change in the way Massachusetts conducts its elections. https://www.nytimes.com/2018/09/17/us/politics/massachusetts-voting-lori-trahan.html

After finishing my research on plurality and the Blue Moon Journal article, I will be looking at the instant runoff method and will start my second paper!