2 ms·

What you are saying is true. However, it is probably hard to find a real example of being to divide a state (e.g.) into X groups that all have that 75-25 mix. N

What you are saying is true. However, it is probably hard to find a real example of being to divide a state (e.g.) into X groups that all have that 75-25 mix. Normally, the city vs country or ethnic groupings will leave certain areas in opposite mixes. The gerrymandering that usually happens is that the group in charge wants to win 3/4 of the X seats at a small percentage like 60-40 and lose 1/4 of the X seats by a huge margin like 90-10. This would give them a huge seat advantage since they own most of the seats. In this case, the gap would be huge proving deliberate gerrymandering.

phailhaus 9y ago
It doesn't matter if it's unlikely, the point is that the efficiency gap is being proposed as a measure of partisan gerrymandering, but it clearly does not do what it claims because it thinks a state with all districts split 75-25 is fair. But it's not: one party is getting 25% of the votes but 0% of the seats.
- snarf21 9y ago
  I'm agreeing with you that a case like this isn't being fairly measured by this gap. I'm not trolling but the likeliness does play into it. This case is not much of a concern if it has never happened in any state in US history, e.g. I think there are other things that can be done to make it even harder to create all 75-25 like saying all districts must have a size or density or .... that is within a standard deviation (or two?) of the mean. I don't know what the right math is offhand but something like this adds another constraint.
  - phailhaus 9y ago
    My point is that there's clearly something missing if a 75-25 split is measured as 0 net wasted votes. That means that the efficiency gap treats 75-25 as the "perfect election", and is actually measuring how far any given district election is from that "ideal". Even worse, that means that if we actually use the efficiency gap to measure gerrymandering, then there is a strong incentive for parties to draw district lines to be as close to 75-25 as possible!

phailhaus 9y ago

It doesn't matter if it's unlikely, the point is that the efficiency gap is being proposed as a measure of partisan gerrymandering, but it clearly does not do what it claims because it thinks a state with all districts split 75-25 is fair. But it's not: one party is getting 25% of the votes but 0% of the seats.

snarf21 9y ago
I'm agreeing with you that a case like this isn't being fairly measured by this gap. I'm not trolling but the likeliness does play into it. This case is not much of a concern if it has never happened in any state in US history, e.g. I think there are other things that can be done to make it even harder to create all 75-25 like saying all districts must have a size or density or .... that is within a standard deviation (or two?) of the mean. I don't know what the right math is offhand but something like this adds another constraint.
- phailhaus 9y ago
  My point is that there's clearly something missing if a 75-25 split is measured as 0 net wasted votes. That means that the efficiency gap treats 75-25 as the "perfect election", and is actually measuring how far any given district election is from that "ideal". Even worse, that means that if we actually use the efficiency gap to measure gerrymandering, then there is a strong incentive for parties to draw district lines to be as close to 75-25 as possible!