[TheLoneAdmin]

I agree we should switch to Proportional Representation. Unfortunately this would weak our duopoly parties, allowing third parties to win seats. So they (D&R) will never change the laws.

In a FPTP election, 10-20% of the votes are wasted on a 3rd party. In a PR election, 10-20% of the votes get 10-20% of the seats.

[dragonwriter]

> Unfortunately this would weak our duopoly parties, allowing third parties to win seats. So they (D&R) will never change the laws.

(1) They will change the law if the choice is being displaced as major parties (i.e., if people care enough), and

(2) for state government, in many cases the elected members of political parties would not need to act, the citizenry could change the laws directly,

> In a FPTP election, 10-20% of the votes are wasted on a 3rd party. In a PR election, 10-20% of the votes get 10-20% of the seats.

Assuming you mean party list PR, that's true, but then also individual candidates aren't accountable to the general electorate and no one has anyone they can point to as “their” representative. That's a significant trade-off; there are systems which achieve rough proportionality (far better than FPTP) that don't have these problems (i.e., STV with small, e.g. 5-member, districts), but if you look simply at a naive “wastee vote” metric as you've constructed here they may not be better than FPTP.

> In a FPTP election, 10-20% of the votes are wasted on a 3rd party

In the US, the number is much lower; of course, the problem is as big or bigger if you add in abstentions and tactical votes.

[bobwaycott]

> ... individual candidates aren't accountable to the general electorate and no one has anyone they can point to as “their” representative.

Alternatively, individual candidates would be accountable to the voters who placed them there. Similarly, the voters who placed them there would think of them primarily as their representative. Those voters who are represented by more than one person in their ideological camp would find they have more than one representative they can reach out to on issues. Those who are more independent but find themselves more closely aligned to a set of reps in one of the camps will also find they have multiple representatives to reach out to, ask to cooperate on something, etc. I live in a state that has GOP dominance at all levels. I’d love to feel like there was anyone representing me the last 20 years at the state and federal levels.

EDIT: I’m not advocating for such a system per se, I’m just suggesting its efficacy is somewhat a matter of perception/perspective. Both voters and candidates can find accountability and notions/feelings of representation in a district-free proportional system. It is surprising to me there’s no way for voters to challenge gerrymandering directly when, in many states, voters aren’t represented by someone who shares their views for decades (on either side of the ideological aisle).

[brianolson]

I advocate STV with larger sets, 8 or more. The wasted vote is smaller, any constituency that can show up and get out the vote for 1/N of the total gets one of the N seats. 5 seats or fewer biases towards the major parties. Lots of 5 seat sets will go 3-to-2 or 2-to-3. I'd like to get in other voices and split it 5/3/1 or 4/4/1. And I bet over time that 1 would grow to 2, but we have a bootstrapping problem in the current system that always throws away that smaller voice.

[dragonwriter]

> I advocate STV with larger sets, 8 or more.

The trade-off with more seats with STV is better theoretical proportionality vs. larger candidate pool that voters must evaluate; because the number of viable parties as well as the number of candidates each can sensibly run scales roughly with the number of seats per district, the attention burden varies roughky with the square of the number of seats.

> The wasted vote is smaller, any constituency that can show up and get out the vote for 1/N of the total gets one of the N seats.

With the usual formula, pedantically, the smallest integer number of votes greater than a 1/(1+N) share of the electorate, rather than a 1/N share.

> 5 seats or fewer biases towards the major parties.

I think you overestimate the degree to which support for the major parties in the US is fundamental rather than a artifact of tactical voting based on the realities of FPTP; the US is, by affiliation rather than typical vote, 39% independent, 31% Democrat, 28% Republican by the most recent Gallup poll numbers, and at least the Democrats consist of two apply conflicting, roughly balanced factions in loose tactical alliance. (The Republicans have factional divisions, too.)