A good way to judge how fair a state's congressional map is, is to look at the popular congressional statewide vote totals per party, e.g., 52% GOP, 48% Dem, and compare that ratio to the party composition of the state's congresiional delegation, e.g., 6 GOP/5 Dem. The closer the two ratios are to each other, the fairer, and less partisan the map, and converserly, the greater the disparity between the two ratios, the more partisan the map. The Texas map of the last ten years is an example of the latter -- the GOP gets more congressional votes statewide, but the Texas congressional delegation is currently split 17/13 in favor of the Dems. This was accomplished by packing Republican voters into a limited number of GOP supermajority districts, while spreading Dem voters across a greater number of modestly Dem districts. Arizona is another example of a partisanly gerrymandered state, only in the GOP's favor, with a 5/1 delegation split in favor of the GOP, in contrast to a popular statewide 60%/40% vote split in the GOP's favor. Texas was controlled by the Dems 10 years ago, and Arizona was controlled by the GOP, hence the partisan maps.
I wonder if it would be possible to do some variation of the 'one cuts; the other chooses' strategy here? Obviously that strategy won't work in its normal form (since once the districts are cut out, it's the distribution of voters which 'chooses' who gets each district) but I wonder if something like it might. I'll have to think about that.
A good way to judge how fair a state's congressional map is, is to look at the popular congressional statewide vote totals per party, e.g., 52% GOP, 48% Dem, and compare that ratio to the party composition of the state's congresiional delegation, e.g., 6 GOP/5 Dem. The closer the two ratios are to each other, the fairer, and less partisan the map, and converserly, the greater the disparity between the two ratios, the more partisan the map.
While it might be 'fairest' for the ratio of party representation to be proportional to the ratio of party votes, a disparity between the two hardly indicates a 'partisan' map. Indeed, imagine that in a state with ten districts, voters were placed into the districts on the basis of the last digit of their SSN. Such a 'mapping' could hardly be regarded as 'partisan', but if either party holds more than 51-52% of the vote it's likely that such a 'mapping' would give that party 100% of the representation.