Do you have a dumb question that you're kind of embarrassed to ask in the main thread? Is there something you're just not sure about?
This is your opportunity to ask questions. No question too simple or too silly.
Culture war topics are accepted, and proposals for a better intro post are appreciated.
Jump in the discussion.
No email address required.
Notes -
@marinuso provides the classic election theory answer, which is supported by both game theory and observational evidence (most FPTP systems become two-party systems in real world).
Why Pareto principle does not apply? I can't explain it properly, but here is something interesting. Pareto distribution over things is not a physical or social "law", Pareto distribution is observed when frequency of things is produced by a phenomenon which is described by a power law. Many mathematical processes can be constructed that generate power laws, and lots of ink spilled to argue some of those processes describe one social behavior or other. See Newman 2006 Section 4 for a review. It's a long list, and the length of the list is the reason I think it's tricky to explain. When things are exponentially, normally, or log-normally distributed, it is easy to say why it is so (things occur at constant rate; things occur as sum of random variates; or things occur as multiplication of random variates, respectively.) Which is why I am a fan of the explanation given on page 23: most things that appear Pareto are not really Pareto, just log-normal distributed in a way that looks like a Pareto distribution.
It's not that difficult to think a hypothesis why popularity of consumer products in a coffeeshop would be log-normal. Suppose potential customer-events for each product depends on word of mouth. Word of mouth is a multiplicative process. First guy to introduce coffee to European peoples introduced it to U(0, Dunbar Number) of people. Each of whom can introduce it to U(0, Dunbar Number) of people. Continue this for generations. Due to some reasons that are essentially random, it turned out to be possible to cultivate Arabica in colonial plantations --> random multiplicative high rolls. Due to reasons that can be thought as random for the purpose of this hypothesis, Japan was closed off until commander Perry arrived (another random event), each of which had multiplicative effect close to 0, hindering the global popularity of matcha. Naturally this is just a just-so story, but it is reasonable to assume lots of social phenomena is multiplicative. And multiplication of random variates yields log-normal distributions.
In elections, the election system induces a need for strategic behavior which is more important than random individual voter preferences. However, in case it helps, I think it likely that popularity of individual politicians is roughly Pareto-like. Competition between the parties encourages them to put forward politicians so that they stay competitive in elections.
More options
Context Copy link