4 ms·

The two party system is an inevitable outcome of the election system we have. The game theory dictates it. Switch to approval voting(http://www.electology.org/a

The two party system is an inevitable outcome of the election system we have. The game theory dictates it. Switch to approval voting(http://www.electology.org/approval-voting http://www.electology.org/approval-voting) and we'll have a much more responsive and ethical government. In a winner take all system, if there are more than two candidates, the dominant effect is the spoiler effect, where having more similar candidates makes it less likely that any will win. This means that anything more than two parties is degenerate, which implies that only the dominant eigenvector of political opinion will ever be reflected in an election. (Which in the US is typically the culture wars/identity politics, unless the economy is falling apart.)

farnsworth 15y ago
I've thought a system like this would be great, but it seems like we would end up with so many candidates that it would be difficult to make informed decisions for every one. Then congress would become fragmented with dozens of factions that don't work together. It seems like a system of a small number of large parties (but more than 2, perhaps) would be most efficient. I suspect that there's a solution but couldn't find one in the link.
- JoshTriplett 15y ago
  > Then congress would become fragmented with dozens of factions that don't work together. That sounds like an excellent outcome. They might actually have to talk to each other, work out solutions that work for everyone, and refrain from doing things they can't all agree on.
wvoq 15y ago
Could you say a little bit to justify your use of the term eigenvector in that sentence?
nandemo 15y ago
If I'm guessing correctly what you mean by spoiler effect, I guess a 2-stage election does minimize its effects. For instance, in Brazil, a presidential (or gubernatorial, etc) candidate must have a majority (that is, more than 50% of the votes) to win. Suppose in the first-stage we have candidate A with 30%, B with 20%, C with 15%, etc. Then a couple of months later we have a second-stage with candidates A and B. So it's possible that B wins. I believe in the approval voting that you mention, there's a situation where candidate C would be able to win, which is also interesting.

farnsworth 15y ago

I've thought a system like this would be great, but it seems like we would end up with so many candidates that it would be difficult to make informed decisions for every one. Then congress would become fragmented with dozens of factions that don't work together. It seems like a system of a small number of large parties (but more than 2, perhaps) would be most efficient. I suspect that there's a solution but couldn't find one in the link.

JoshTriplett 15y ago
> Then congress would become fragmented with dozens of factions that don't work together. That sounds like an excellent outcome. They might actually have to talk to each other, work out solutions that work for everyone, and refrain from doing things they can't all agree on.

wvoq 15y ago

Could you say a little bit to justify your use of the term eigenvector in that sentence?

nandemo 15y ago

If I'm guessing correctly what you mean by spoiler effect, I guess a 2-stage election does minimize its effects. For instance, in Brazil, a presidential (or gubernatorial, etc) candidate must have a majority (that is, more than 50% of the votes) to win. Suppose in the first-stage we have candidate A with 30%, B with 20%, C with 15%, etc. Then a couple of months later we have a second-stage with candidates A and B. So it's possible that B wins. I believe in the approval voting that you mention, there's a situation where candidate C would be able to win, which is also interesting.