Skew the polls

I see Lynn Sweet, the woman who single-handed derailed an important national conversation about health care reform and generated a weeks long media circus with her ill-timed, off the wall, out of context question about the Henry Gates arrest at a major presidential news conference, is flogging yesterday's buzz on Romney's alleged surge in the polls. Today's polling has already made those polls somewhat irrelevant.

Of course, regular readers know I put little stock in the polls at any time. They're useless as predictive tools. They only encourage horserace coverage over policy analysis and I'm coming to believe they really are skewed. Not for partisan purposes. Strictly for business reasons.

Polling used to be rare and done by only a few trusted organizations. Now they're a big business and lots of people making serious money on them. If they all asked the same questions, phrased neutrally, they would all get the same results and eventually one or two big firms would eliminate the smaller competitors. Delivering wildly conflicting results is one way to ensure that the horserace obsessed will slavishly follow every damn one of them. And a lot of websites making money on apps that aggregate the results.

I'm not suggesting there's any collusion, but in the brave new world of digital media, traffic stats equal money. I see no incentive to challenge the whole premise of daily polling and plenty of incentive to encourage it. Which only serves to engender hysteria among nervous partisan observers. Ultimately, it's the electorate, and indeed our system of government, that loses when the focus is on who's winning which demo because of the way they said it, instead of reporting on the real world ramifications of the policy proposals being put forward. Not I expect it to change in the foreseeable future.