Hmm.... this seems extremely naive on the part of the author. Let's take his approach for a moment. How do we look at the statistics he proposes? Do we look

Hmm.... this seems extremely naive on the part of the author. Let's take his approach for a moment. How do we look at the statistics he proposes? Do we look at the last election or the last decade or since the last redistricting? His analysis showed single sets of races, but US states have bicameral legislatures... do we somehow combine those results or pick and chose? Do we, as the author did, discount third party participation and if so isn't that also biasing redistricting efforts in favor of two parties that are not as different as most would like to admit? Do we choose our time frame ad hoc at the time of redistricting? How do we deal with this problem: we have a district that was itself gerrymandered at one point in time to ensure that a significant racial minority had sufficient electoral clout in at least some districts (including our example). Now that district votes for Party A consistently, but the mean vs. average approach shows that the redistricting unduly favors Party B... do we meet the aims of inclusion or do we try to ensure that political parties are not advantaged/disadvantaged at redistricting time? The real problem here is that democracy has become this important at all. Rather than worry about how to best preserve the rights of each individual from undue interference from anyone (including government officials) we are more worried about ensuring that we are not unduly disadvantaged in having access to the force that government represents: and in so doing ensuring that we can use it against those we don't favor for the benefit of those that we do.