Burnt Orange Report

Microtargetting Voters -- Good idea, fad, or both?

By Jim Dallas

Newsweek last week ran a story about the increased use of market research to target individual voters, with the prerequisite "Karl Rove is god" ("Wizard of Oz" might be more appropriate appelation for Mr. Rove).

True, the Republicans still have a huge lead on us when it comes to micro-targetting voters, but we're catching up with Demzilla (we hope).

Additionally, I tend to suspect that the law of diminishing returns applies here; just because the Republicans know which color underwear you're wearing right now doesn't mean that information is going to win them your vote; and if they're going to spend money on that, then let them (because at some point, a marginal increase in information on a voter is just going to be a wasted investment).

All this talk all boils down to making an educated guess about your neighbors values so that you can most effectively persuade them. For the El Cheapo candidate (e.g. a justice of the peace candidate in Podunk), a totally free service like Claritas's ZIP Code search -- or census records, tax assessments, and voting histories -- ought to get them half way there. While computers and market segmentation databases are going to be very helpful, they're simply an extension of what campaigns ought to be doing -- effectively -- anyway.

(I suppose it's worth noting that the free stuff I suggested only allows an assessment of your audience in the aggregate; it can tell you about a neighborhood, but not about one particular neighbor. That's true, of course, and that's why all this new stuff can't be ignored. But nor will it improve a campaign's effectiveness by orders of magnitude, in my humble opinion.)

On the other hand, the road to victory is paved with the skulls of backwards-thinking pols who failed to get with the times. See for example the folks who pooh-poohed scientific polling or television or helicopters in the 40s and 50s -- and then got beat (sure, Lyndon Johnson may or may not have cheated in 1948, but by embracing "high-tech", Johnson was able to essentially tie Coke Stevenson, who was at the time probably the most popular politician in Texas).

Thanks to Kevin Drum for noting Claritas's site on Washington Monthly.

TOTALLY OFF TOPIC: The "you are where you live" meme came up in Drum's defense of David Brooks, who was attacked by Philadelphia Magazine's Sasha Issenberg for "not checking his facts." Umm, and I could say the same thing about Issenberg, who claims that "one of Goodwin's strongest markets has been deep-Red McAllen, Texas."

HAHAHAHAHAHAHA. McAllen might be geographically located in Texas, but every Texan knows that McAllen isn't really part of the sociopolitical construct of Texas, by which I mean George W. Bush's cowboy fan club (nor is Austin, nor San Antonio, nor Galveston, really, by that standard).

More to the point -- shouldn't Lyndon Johnson's biographer (that's who Doris Kearns Goodwin is) get lots of readers in South Texas, where some of LBJ's more infamous exploits occurred (see earlier reference to 1948 Senate Race)? I mean, duh.

OH. According to Issenberg, Galveston is part of "Red America," despite having a political culture somewhere to the left of Philadelphia's. ("Blue Americans have heard so much about Red America, and they've always wanted to see it. But Blue Americans don't take vacations to places like Galveston and Dubuque," Issenberg claims.)

Just look at the numbers -- McAllen's Hidalgo County went for Gore 61-38, which is about the same result tallied in "uber-blue" Montgomery County, Maryland, which went for Gore by a margin of 63-34. Galveston County, as a whole, is deeply "purple"; but the city of Galveston itself went for Gore.

The whole Philly Mag piece is a laugh riot. Obviously, Sasha Issenberg doesn't know anything about Texas, or Texans, and I'm starting to think that I don't want people like that in my state, anyway.

Once you scale up the "red/blue" hypothesis to geopolitical units the size of states, it starts to lose all meaning and sociological accuracy (as the post I made a while back about Starbucks in Vermont ought to indicate).