I’ve been reading a lot of political coverage this primary season—too much for my or anyone’s health and sanity. And in the course of that reading I’ve developed a number of pet peeves about political reporting. Many of these, I realize, are common and long-standing complaints: from the echo-chamber aspect of it all to the too-predictable cycle that carries “news” from a campaign conference call on Day 0 to sites like TPM and Politico on Day 1 to articles in the newspapers and in Slate on Day 2.

But one major complaint that I haven’t seen aired before is that both the campaigns and the news media appear to share an assumption that seems to me mostly unwarranted. The assumption is that the decision to vote for a candidate is the sum total of lots of smaller decisions: what that voter thinks about issues a, b, and c; how that voter feels about the race, religion, and gender of candidates m, n, and p; and which candidate the voter thinks will be better suited to handle situations x, y, and z. The idea is that if you can find the right mathematical weight to assign to each of these micro-decisions, you will be able to predict to a high degree of accuracy the macro-decision of who a voter will chose in the polling booth. Contrariwise, if you want to explain why a voter chose the candidate she did, you should be able to break it back down into lots of little micro-decisions.

My problem with this assumption is that it doesn’t at all take into account the fact that at a certain point most voters make a decision to support their candidate, and from that point forward that macro-decision will determine in large part how they make their micro-decisions. In other words, once you pick your guy or gal, you start to adjust your thinking along the lines that are most friendly to that guy or gal. (You want them to win, after all.)

The most vivid example of this is the exit poll that showed that Hillary Clinton won 95% of the voters in Wisconsin who said that experience was the most important quality in a Democratic candidate. But the way this fact is always reported makes it sound like people were looking for someone with experience first, and only then chose Clinton. But the arrow of causation here seems confused at best. There’s no good reason not to think that the reverse was true: that people picked Clinton and then, when asked to justify that choice, said that “experience” was the most important quality in a candidate. And no one has to guess why they would choose that quality: the “experience” line is, after all, the one that the Clinton campaign has been feeding its supporters since the beginning.

I noticed this assumption at work again today in a post by Greg Sargent at TPM’s Horse’s Mouth. Sargent quotes a new poll by the New York Times that says 48% of Democrats think that the press goes harder on Clinton than it does on Obama. Even after acknowledging that 51% of Democrats think that the press is either neutral or biased for Hillary, Sargent calls it “eye-opening that nearly half of Dems say the media’s been harder on Hillary—it’s a far larger proportion than among the pundits, who rarely if ever acknowledge this to be the case.”

Here again, though, it seems to me that Sargent is putting the cart before the horse. He’s assuming that people are making an independent judgment about media bias that has nothing to do with their choice of candidate. But isn’t it at least a little suspicious that that 48% number is so close to the level of support for Clinton in national polls? If I were a Clinton supporter right now and someone asked me if she was being fairly treated by the press, I would say “no.” But I don’t trust that my reasons for that judgment would be independent of my support for her. A much more plausible scenario is that I would have decided at some point to support Clinton and would now be faced with the fact that Clinton is not doing well in the primaries or in the polls. Given that I would want a way to explain that, and that I would want some way that doesn’t involve accepting the messianic transfiguration of Barack Obama, why wouldn’t I accept the Clinton campaign’s media-bias storyline, especially when it’s fed to me from a NYT-hired pollster?

The point I’m trying to make here—somewhat convolutedly, I acknowledge—is that this assumption I’m attacking involves unwarranted ideas about the causes of our decisions. Political reporters almost always talk about a voter’s choice to support a candidate as if it were the final effect of a long and complex series of motivations and judgments. This is surely true for as long as a person remains undecided. But once a person makes a decision to support a candidate (even if that comes long before they visit the polling booth) that effect then itself becomes a cause of our motivations and judgments. I might vote for Clinton because she has the most experience, or I might think experience is the most important quality in a candidate because I’ve decided to vote for Clinton. But political reporters almost never acknowledge this back-and-forth. I wish they would.