Bishop's "The Big Sort": Two Snapshots of Harris County
Pretty pictures to add some color to the discussion of Bill Bishop's "The Big Sort" ...
The color-coding is as follows:
Dark Red: 0-25% Dem
Red: 25-45% Dem
Purple: 45-55% Dem
Blue: 55-75% Dem
Dark Blue: 75-100% Dem
The 2006 map is just a straightforward calculation of the Moody-Willett results for Texas Supreme Court. The 1994 results are slightly more complex - I just averaged out the results of the Senate race (KBH v Richard Fisher), the Governor's race (Bush v Richards), and the State Treasurer's race (irrelevant). Basically, a representative statewide election result of the year, a close loss and a close win ... all from statewide races.
Taken together, it shows the lowest point for Harris County Dems and the latest measurable state that we're in. There are elements in there that, to me at least, both emphasize the power of place that Bill Bishop describes, but also elements that call it into question.
Precinct 149 (furthest west purple blotch on the current map) is shown as going from deep red to purple in these two snapshots. The bulk of this area in 1994 was precinct 1393 and it went 22.3% Democratic. Yet, in 2006, precinct 149 went 46.3% for Bill Moody. Yet, at the same time, there are many consistent themes in where one vote categorization moved the direction it did. Namely - proximity matters. You can see areas where the shading shows a growing region of Democratic vote (Alief in the southwest, for example). Alief might be viewed as more likely to trend Democratic since it is proximate to Democraticly-inclined precincts around it under that logic.
Obviously, Harris County had even more Dem-friendly areas during much of the 1970s and 1980s, when countywide elected officials were predominantly Democratic. So over the course of Bishop's timeline, we've seen several swings and fluctuations all within this one single county.
What to make of it? I don't pretend to have gone into the depth of research that Bishop has, but this is pretty much where we'll go with the posts planned for the next day or two. For now, enjoy the pretty pictures.

If you use the Lampson-Gibbs and/or Matula-Davis results for South East Harris County for 2006, you will see much more purple than with the Moody race.
I'm not likely to use the Lampson results here just yet. I've done a district map on that race and it's heavily skewed. But you're right about Matula in SE. Add to that, Ellen's (134), Scott's (137), and Trautman's (127), there is an argument for including them within a county-level view and it adds to the reasons we're optimistic about the county this cycle.