“The margin of victory is razor thin but a win is a win and Mitt Romney will take it.” -Nora Raum, NPR News.

The absurdly close results from Iowa provide a worthwhile moment to take stock of what actually gets decided on voting nights in the presidential nominating season.

Think of a normal election night: Eight votes more means that candidate is elected. A win is a win is a win, indeed.

But in presidential nominating contests, the voters in one state after another merely apportion their state’s delegates to the national convention. And technically, that’s the point where someone actually wins.

Normal election nights give us the clear information of a victor, and it makes for clear next day stories. Not so with presidential nominating season contests.

But while they rarely decide much individually, they do impart information.

And so, in wake of the caucuses (and it will happen again after the New Hampshire primary, and depending on how things shake out, for weeks or months after that) journalists sift through the numbers looking for other meanings, stories and narratives.

That’s appropriate and unavoidable, because it is clear that there is information to be found: Michele Bachman got one kind of message, and withdrew from the race after her weak finish. Rick Santorum and Ron Paul showed that there is a wide swath of Republican voters who prefer them to Romney.

But the maddening thing is that this search for meaning can verge into head-twisting metrics, assertions, and debates that call to mind Calvinball, the comic strip game whose rules are endlessly revised so that only the rulemaker can win. So: Did Romney win Iowa? Or did he emerge battered when his essential draw is weighed against some set of expectations about how he really should have done? And, relatedly, did he make a strategic error in deciding to whole-heartedly (or perhaps just more-heartedly) contest the state late last year? And did Paul’s very close third place represent a win or a vast disappointment? And perhaps most absurdly, does the lack of nine votes from Santorum supporters belie the strength of the threat he now presents to Romney?

All those questions add up one big one: Who now has momentum?

This question and the others may or may not be interesting to consider, but there’s no doubt each one is measured against an enormously subjective yardstick.

Of course there is something concrete that is decided on caucus night: the delegates. The parties deploy a kaleidoscope of methods—caucuses, primaries, winner take all, proportional allocation, etc.—to divvy up these representatives. Here’s how Iowa does it: the precinct caucuses meet to elect delegates to county conventions (slated for March 10), who will in turn select delegates to congressional district conventions (April 21) and the state convention (June 10).

That later date is when the national delegates, the ones who will travel to Tampa in late August for the Republican convention that will formally choose a nominee, are actually selected. At no stage are each of these caucuses or conventions bound to distribute their delegates by any set procedure, nor are they obligated to select those delegates based on each body’s presidential preference.

That’s complicated. The Associated Press, and its delegate guru Stephen Ohlemacher, admirably explained it in a stand-alone article, which translated last night’s results into an eventual Iowa crew of 13 Romney delegates and 12 Santorum delegates, who will join 2,286 other delegates in Tampa.

But you’d be hard pressed to find those numbers relayed elsewhere. The absence is illustrative, because it shows that we long ago decided that the impact of the Iowa caucus is far more important than its mathematical results. Of course that’s the case: since its nearly whole cloth creation in 1976, Iowa has held the position of the most meaningful but yet meaningless event in presidential politics. (The Ames straw poll is perhaps the platonic reflection of the real event, certainly more meaningless, yet still scandalously meaningful.)

Reporters are used to covering campaign events that do not literally decide the nomination: each new poll, new attack ad, or new fundraising number brings reams of copy and minutes of chatter about how it might change the race, and often, more self-referentially, how it might change the narrative of the race.

This strand of speculative campaign coverage, with its endless dissection of strategy, draws some readers and repels others. But one of the hoary old conceptions of much day-to-day journalism is that it is meant to be a mirror of events, not a shaper of them.

That is clearly not the case when candidacies rise and fall after the tiniest of results are amplified through the campaign press’s churn. (How tiny? Iowa’s delegates represent just 1% of the Tampa vote; the Republican caucus-goers could have fit into a football stadium; even coming in last among the major contesting candidates, Bachmann was no more than 13 delegates out of the lead.)

But voters do look to earlier results to coordinate their choices. The present campaign offers a perfect example, as Republican voters who prefer a candidate besides Romney may now turn to Santorum as the candidate most likely to edge him.

There’s nothing wrong with a certain class of journalists trying to anticipate these shifts. And by doing so, they will inevitably amplify the very phenomenon they seek to describe, as there is no way to report on momentum without also contributing to it.

As last night’s votes trickled in, cable viewers likely heard analyst panels asked variations on this question: “What’s tomorrow’s headline?” Unpack that (somewhat quaint) query’s reflexivity, and it shows that journalists sometimes try to anticipate and cover the changes that their and their colleagues’ coverage will create.

Such discussions are a mixed blessing; they reflect a consciousness of the dynamic, while inevitably playing into it. As Brendan Nyhan argued for CJR just before the caucuses, reporters need to be conscious of the way that they create as much as report on momentum, and straightforward with their audiences that that’s what’s going on. Talking about changes to the narrative, even if it is part of a feedback loop and couched as an exercise in predictive expertise, may be as full a reckoning of journalists role in creating momentum as can be hoped for.

Politics, you may have heard, is a messy business.

If you'd like to help CJR and win a chance at one of 10 free print subscriptions, take a brief survey for us here.

Clint Hendler is the managing editor of Mother Jones, and a former deputy editor of CJR.