One of the strangest spectacles in American politics is the press coverage of a major speech or debate. You get a bunch of media personalities who, by definition, follow politics far more closely than the average American speculating about how they think the speech would play with a typical undecided voter. Compounding the irony, there’s often a description of a nominal swing demographic — the white working class, the heartland, whatever — that couldn’t be further from the experience of a wealthy television celebrity. And even crazier, many of these events aren’t actually watched by that many people; their most important consequence is as a driver of media coverage. Thus, undecided voters are affected by the media’s coverage of what they think undecided voters would have thought had they watched the thing. Ryan Avent proposes a better way:
I don’t want to diminish the work of my journalist buddies, many of whom have written some really good, really insightful stuff on the speech. But I think that we’d get much more informative coverage by randomly selecting ten viewers and asking them to write what they thought of it.
This makes a ton of sense. So much actual convention coverage is, by contrast, baffling. CNN appears to have spent vast sums of money flying their “talent” out to Denver, putting them up in hotels, and constructing an entire new set in order to provide a venue for talking heads to “cover” the convention by talking over the speeches. And then next week they’re going to do it all over again! Suppose they’d just sent one reporter, not built a new set, showed more speeches, and let their talking heads discuss the speeches from CNN’s usual studios in DC and Atlanta. They could have saved a ton of money. And would substantially fewer people have watched? MSNBC, too. Why did Keith Olbermann and Chris Matthews need to be in Denver? What value did that add?